PEAP Projects

From OMAPpedia

Jump to: navigation, search

All project posted here are visible to public users. Please refer to PEAP guidelines for more information.


Contents

Project Headline Template

Title:

Description:

Give a long description of what the project involves.
Provide details on the gap/void in the community the project is addressing

Project/URL (if exists):

Expected results:

Give a details of final out of the project & what will be implemented

Background & work by projects submitter/s:

Give details of your background and all relevant information that will help the judges determine you have the necessary skills, interest & background needed to execute your proposal

Contact email:

SPC - Super Portable PC (A PC in a Pocket)

Title: SPC or Super Portable PC - A PC in Every Pocket!

Description:

There are of course multiple computers in the Market in various sizes and shapes. Owing to its incredibly small size, the objective of this project will be to develop a PC so small, it could literally be indistinguishable size-wise kept beside a 3.5 inch portable disk. The 'SPC' can run Android 3.0 possibly higher versions in the future, making room for cost effective computing, which not only saves space, but power and energy resources.

The HDMI output would make it easy to connect with TVs and Monitors, including the Monitors/LCD screens in Automobiles. Ability to use this PC on the go using Portable Pico Projectors and Eye wear with Displays With its superior OMAP architecture and Dual Processors, streaming HD and playing HD clips should be a breeze. Streaming HD content via Wireless would make it an ideal entertainment center. The SPC will also have backward capable compatibilities to help developing countries to use older TVs as monitors. The 'SPC' Would have android plugins which could help any user key in data using his/her Cell phone via Blue-tooth or a Blue-tooth keyboard. Owing to its expansion ports and connectivity options, It will include optional sensors like smoke detectors and noise level detectors, constantly keeping the user informed about his/her surroundings and warn the user appropriately. It not a new Idea, though inclusion of a camera to help developing countries send photos of medical conditions using this device can go a long way.


Android Plugins would help this SPC to automatically sum up a users daily routine, Keep health records, Call/phone records, Back up data from other sources regularly via bluetooth or wireless,

To sum it up, We have for you an Ultra Portable, high performance, low cost, multipurpose PC.

Project/URL (if exists):

Expected results:

A multirole, multipurpose, portable android PC that can be run via various output devices and input devices

Background & work by projects submitter/s:

We are a team of software professionals with 10 years of experience with embedded software design & development (designed & developed a RS232 controlled PCB), Software Security, Web development & Internet marketing.

Contact email: ,


Porting Android 3.0 and integration of Scan Engine in to OMAP4

Title:Porting Android 3.0 and integration of Scan Engine in to OMAP4

Description:

The objective of this project is to port the Android 3.0 and integrate the scan engine as well the third party software scanner aka Zebra Xing in to OMAP4. As Android 3.0 supposed to be Enterprise enabled this project show cases the enterprise segment features in to OMAP4.

Project/URL (if exists): NA

Expected results:

* Android on OMAP4 is Enterprise ready
* By using the Mutli Cores can show case the performance improvements

Background & work by projects submitter/s:

Had over 12 Years Experience in Embedded field.worked closely with all TI chip sets and done many base ports on all the chipsets.partcularly Symbian as well the Android.Worked on DSP bridge.

Contact email:

Versatile enclosure case for the PandaBoard

Title:

Versatile enclosure case for the PandaBoard

Description:

The goal of the project is to design an enclosure case that will help to integrate the Pandaboard in other projects, and also allow it to be used as a one of the smallest open nettop computer.
Milestones:
- Market research: the first stage is to evaluate existing enclosures and plataform similar to the one proposed (i.e.: Fit-PC2, eBOX530-820, etc.).
- Project forum: an URL will be set up where people interested in the project can join and help us with their suggestions.
- Study of requirements: we will encourage the community to send their ideas for the design and make a list of requirements in order to evaluate the design alternatives.
- Generation of alternatives: taking into account the requirements a series of alternatives will be generated and evaluated by us and the community involved.
- Development of the selected alternative: we will get in contact with providers to analyze the manufacturing issues and refine the design.

Project/URL (if exists):

Expected results:

The result will be a versatile and very cool enclosure for the PandaBoard suitable to be manufactured and deliverable to the community at reasonable prices.
At the same time the enclosure should be a good option to deploy a PandaBoard based nettop for anyone.

Background & work by projects submitter/s:

We are a team of creative designers providing consultancy and development of interactive installations, visual arts and multimedia.
Some of our areas of interest are: computer vision, robotics, augmented reality, mixed reality, real 3D sensing, etc.
Most of the tools and platforms we like to use are open: gnu/linux, processing, openframeworks, puredata, etc.
We are also open to collaboration in interesting projects with other creative people with common interests.
David S.K.:
I’m Industrial Design Engineer and Master in Visual Arts and Multimedia, PhD candidate at the Polytechnic University of Valencia, Spain. I’ve been working in interfaces design and programming for ten : years and at the same time doing product design and prototyping. Portfolio: www.dasaki.com [1]
Francisco S.P.:
I’m PhD Professor at the the Polytechnic University of Valencia, Spain. I teach real time video applications and hardware oriented to interactive artworks at the Fine Arts College.
Some of the experiments we’re cooking can be visited at: www.therandomlab.com [2]

Contact email: [.]es

OMAP4 Integrated media and communication system for automobiles

Title:

OMAP4 Integrated media and communication system for automobiles

Description:

Project aims at building an integrated system of video, GPS, music and browsing facility in automobiles with touch screen. In order to give user the flexibility to use his own application and personalize, Android OS is used.
This system would make the best use of the benefits OMAP4 has to offer. High Definition video will provide the better picture quality. Low power audio with a better audio output and ::power back-up. Saving the power during traveling. Browsing facility to reduce the need to carry ::laptop. Android to explore the world of innovative applications.

Project/URL (if exists):NA

Expected Result:

* Easy navigation
* Better video quality
* A compact system at low cost

Background:

Engineering graduate with three years of experience in embedded systems. Currently working on OMAP3.

Contact Email:

Android Based Automobile Navigation and Entertainment System

Title: Android Based Automobile Navigation and Entertainment System

Description:

ABANES seeks to provide the car PC community the ability to utilize the Android OS and its built in features to fabricate an automobile based entertainment and navigation system. The project aims to develop a platform that enables: GPS turn-by-turn navigation, music playback, 1080p video playback, rearview (backup) camera display, internet connectivity (through WiFi, tethered cellular phone or dedicated WAN hardware), BlueTooth paring/hands free operation of compatible cellular phones, FM radio playback, satellite radio playback, OBD2 diagnostics capability and community application support. Many of these technologies or applications exist on other platforms (cellular phones, full x86 based car pc) but are unwieldy, impractical or, in some jurisdictions, illegal to use in a moving vehicle.
ABANES will leverage the current car pc community (sites such as mp3car.com) and their resources to unify these technologies with an open platform. A full hardware prototype will be developed and installed inside an automobile. Where possible the project will use off the shelf components and drivers to minimize redevelopment (eg. Leopardboard camera module). The focus will be on incorporating novel elements (FM radio, Satellite radio, etc.) and leveraging already completed Android work for the PandaBoard. The PandaBoard is the first affordable non-x86 development platform to offer the requisite processing power and communication accessories.

Project/URL (if exists):

Expected results:

ABANES intends to incorporate the onboard WiFi, BlueTooth, FM, audio and video processing modules. Rearview camera functionality is to be accomplished via the Leopardboard camera module. GPS, Satellite radio, WAN and OBD2 modules will be sourced and drivers will be written where necessary. Existing power supply options for x86 based car entertainment systems will be incorporated and supported at the driver level. The system will initially use a double DIN mountable touch screen that is commercially available, although, it is hoped that a higher resolution screen can be sourced (through the developer’s industry connections) and supported as a secondary objective. The end result will be an open source car entertainment system more powerful, user friendly and customizable than anything currently offered commercially.

Background & work by projects submitter/s:

3 x Computer Engineers:
-6 years experience as software developers for North America’s largest smartphone (and soon to be dual core XXXXX based tablet) manufacturer.
-5 years experience developing applications in the mobile space.
-6 years experience in electronics fabrication and debugging.
-Specialization in cellular radio calibration/testing, GPS fabrication and test, Personal Area Networks, microprocessor interfacing, embedded systems and computer vision.

Contact email: Yoda1[.]

Embedded Master on PandaBoard

Title:

Embedded Master on PandaBoard

Description:

Embedded Master (EM) is a project of Open Embedded Software Foundation (OESF) which aims to make Android becomes a common platform for various embedded systems. EM complies with the Android framework, retains compatibility with API of Android, and extends various functions commonly necessary for embedded system of home information appliances.

Project/URL (if exists): http://developer.oesf.biz/em/developer/

Expected results:

A working EM3 version on PandaBoard, with some extensions such as:
- OESF boot screen
- Usb wifi (WPA supplicant)
- USB Keyboard
- User Interface extension
- ALSA Sound
- DLNA Extension
- Home Media Service extension
- OESF MarketPlace extension
- OESF Platform Builder support

Background & work by projects submitter/s:

Open Embedded Software Foundation (OESF) is an organization focused on standardization, co-development and expanding market reach of Android platform for embedded devices. We are focusing on proliferation of this powerful open source software into embedded products beyond cellular phones, and it is driving for reclamation of a new market and activation of embedded market as a leader of embedded technologies.
Released version of Embedded Master (EM):
- EM1 (Codename: blueberry)
- EM2 (Codename: cinnamon)

Contact email:

DLNA server

Title: DLNA server

Description:

The Digital Living Network Alliance (DLNA) is a non-profit collaborative trade organization. Alliance members have stated the common goal of using standards-based technology to make it easier for consumers to use, share and enjoy their digital photos, music and videos.
This project is meant to create a fully DLNA compliant Home Network Device.
The system could use HDMI as video output in order to visualize high resolution multimedia content (videos, music and photos).

Project/URL (if exists):

TBD

Expected results:

PandaBoard will run a DLNA media server on the top of a linux based OS.

Background & work by projects submitter/s:

Team of 6 engineers with in-depth experience in a broad array of embedded software development technologies working for leader TLC companies since 2000.
Deep knowledge on TI C6x DSP’s and ARM architecture.

Contact email:

Implementing control of LCD alphanumeric displays

Title:

Implementing control of LCD alphanumeric displays

Description:

PandaBoard has not a display embedded in it. You must have some external (maybe expensive) hardware connected, to control it. At least you must have one of this:
- a DVI monitor connected to DVI port on PandaBoard
- a PC (or something else) connected to serial port
The goal of this project is to have a cheap alphanumeric display working on PandaBoard so it can be controlled in a "standalone" fashion. You will have a (minimal) Linux shell on the LCD where you can digit with an USB keyboard. The driver will work with different geometries of display (2x16,4x14,1x40,...).
At first I will implement Hitachi HD44780 protocol. In a second phase I would like to manage other protocols as well.
This project will be heavily based on the lcd-linux project (http://lcd-linux.sourceforge.net/). The difference is that PandaBoard has not a parallel port so I will try to connect the LCD to some unused GPIO of the PandaBoard (or even to serial port through a microcontroller). Some soldering has to be done here, but nothing very difficult.

Project/URL (if exists):

not yet but I'm planning to create a new gitorious project or collaborate with existing project.

Expected results:

working LCD alphanumeric displays on PandaBoard.

Background & work by projects submitter/s:

I'm an Electronic Engineer and I've been working on Linux and open-source for 10 years, developing device drivers and making modifications to the Linux kernel and various bootloaders (Qi, XLDR, U-boot). I've also good skills on C and x86/ARM assembly. I've been working on x86 and ARM architectures (Intel pxa270, Samsung s3c2442/Freerunner GTA02, Freescale iMX51, TI OMAP3530/Mistral EVM board). I've good skills on microcontrollers (PIC, H8,...) as well.

Contact email:

Porting Qi bootlader on PandaBoard

Title:

Porting Qi bootlader on PandaBoard

Description:

The goal of this project is to port Qi bootloader to the PandaBoard. Qi is a tiny, fast, "no frills" bootloader for ARM architectures. It permits direct loading of Linux kernel without intervention of u-boot. Its aim is to stay at the minimum needed to load and boot Linux kernel (without boot-menus or additional peripheral init).
At the moment there are Qi's versions for Samsung s3c* and omap3 (in a different branch) but there is no omap4 version. This project will try to develop and maintain a working version for omap4.
This bootlader will give feedback to the user by turning on/off LEDs on the PandaBoard in a way similar to FreeRunner GTA02 (for example turn on LED on successful kernel pull). It will also accept input from the user by pressing some buttons (for example adding debugging parameters to the kernel command line). To this end maybe I will need to add some push-buttons or LEDs connected to GPIOs
In a second phase I'm planning to add JTAG support to PandaBoard to debug early stages of boot. In this phase I will implement, configure and document a complete open source based JTAG debugging development chain for ARM Cortex A9 in OMAP4 used on PandaBoard. This includes final port of open source JTAG software OpenOCD for OMAP4 on Panda, and then configure and document all software (and hardware) components involved.

Project/URL (if exists):

not yet but I'm planning to create a new branch in the existing Qi project in gitorious (http://gitorious.org/0xlab-bootloader/qi-bootloader/)

Expected results:

- working (and fast) version of Qi on PandaBoard
- feedback and input to/from the user through buttons/LEDs

Background & work by projects submitter/s:

I'm an Electronic Engineer and I've been working on Linux and open-source for 10 years, developing device drivers and making modifications to the Linux kernel and various bootloaders (Qi, XLDR, U-boot). I've also good skills on C and x86/ARM assembly. I've been working on x86 and ARM architectures (Intel pxa270, Samsung s3c2442, s5pc100, Freescale iMX51, TI OMAP3530/Mistral EVM board). I also own a Freerunner GTA02 Phone on which I've already installed and modified my own version of Qi and debugged it with JTAG and OpenOCD.

Contact email:

Title: Maemo 5 on PANDA Board

Description: Maemo 5 BSP for Zoom2 will be ported to PANDA board. The LCD will be integrated with the board and the full set-up will be brought-up on Maemo 5.

Project/URL (if exists): NA

Expected results: The project involves porting the Maemo 5 BSP from OMAP 3 to OMAP4 and integration of OMAP4430 PANDA Board with LCD. This will be a clone out of OMAP4430 with Maemo 5. The output will be have development kit and Maemo BSP

Background & work by projects submitter/s: I am designated as Software Architect having 18 years of experience in embedded software development. I was involved in Platform Enabler Software (BSP) development for multiple ARM, ARM Cortex, Microchip, NXP Microcontrollers. I have handled multiple System Integration activities on OMAP and Imx21 basically to integrate the multimedia and Telephony components to the Linux and Symbian. I am also a hobbyist and worked on bringing-up uC/OS on PIC.

Contact email:


IP Surveillance Camera

Description:

OMAP4 could be used as IP Surveillance Camera reference design ideal for applications ranging from surveillance to remote monitoring like shopping malls, warehouses, airports, homeland and many more This project basically demonstrates the capabilities of OMAP4 in High Definition Video processing, to provide OEMs with the ability to deploy IP Surveillance Camera reference Design. It uses advanced video processing technologies supporting H.264/MPEG4 video compression and intelligent video content analytics simultaneously

Project/URL (if exists):NA

Expected results:

The final project will be demonstrated for IP Surveillance Camera as below. The project will involbe bring-in up Android (Eclair or Froyo) on this PANDABOARD. The Android OMX video stack will be ported to this PANDA and the web application on Host will be written to demonstrate the IP camera capability

Background & work by projects submitter/s:

I am qualified Electronics Engineering professional with 3 years’ qualitative experience in embedded Software/ Hardware Product Development and Device Drivers). I have rich skills in the development and testing across OS like Embedded Linux (programming of Device Drivers, Middleware and Applications). I am designated as Senior Software Engineer involved in Platform Enabler Software (BSP) development for OMAP series. I have recently worked on Beagle I have handled multiple System Integration activities on OMAP .

Contact email:

Title: QNX Neutrino on PANDA Board

Description:

• In the first phase - QNX Neutrino 6.4.1 BSP for OMAP3530 will be ported to PANDA board. The electronics will be made to integrate the LCD with the board and the full set-up will be brought-up on QNX Neutrino RTOS. • In the second phase - QNX CAR will be ported onto this board


Project/URL (if exists): NA


Expected results: The work product of this project will be OMAP4430 PANDA Board with LCD. This will be a clone out of OMAP4430 with QNX car running. The output will be have development kit and QnX BSP


Background & work by projects submitter/s:

I am qualified Electronics Engineering professional with 18 years experience in embedded Software/ Hardware Product Development. I have rich skills in the development across OS like QnX, Nucleus and Embedded Linux (programming of Device Drivers, Middleware and Applications). I am designated as Software Architect involved in Platform Enabler Software (BSP) development for OMAP series, Imx21, MXC300-30 mobile platform, AT91SAM, Luminory Micro, LPC1100, Pic32. I have handled multiple System Integration activities on OMAP and Imx21 basically to integrate the multimedia and Telephony components to the Linux and QnX CAR platforms.


Contact email:

Compact ground control station from Ardupilot

Description:

Ardupilot is open source Arduino based autopilot radio controlled airplanes. Many of these planes are equipped with video cameras for FPV (first person view)-flying. Necessary OSD-instruments are usually drawn onboard to video feed using some proprietary system like ezOSD. Ardupilot includes support for transmitting telemetry data live to ground station. While there currently exists many projects for ground stations, to my knowledge none of them handle recording of video.
PandaBoard's strength would be possibility to construct small and rugged system for handling simultaneously video recording and OSD flight instrument rendering to live video. Rendering OSD on the ground station would save weight on the airplane by eliminating proprietary system and would allow saving unmodified video separately with telemetry data. I've previously tried this with BeagleBoard, but it turns out it's not quite fast enough for the live feed.

Expected results:

Ground station software for receiving telemetry data, saving video and rendering flight instruments to live video for FPV flying.
Enclosure for ground station (PandaBoard, video receiver and battery)

Background & work by projects submitter/s:

I already have preliminary code ready which currently runs on the laptop. I'm interested to port it to PandaBoard since the laptop is not rugged enought to be used outdoors. I'm familiar with BeagleBoard and I'm currenly working for the MeeGo project.

Contact email:

NANNY

NANNY (Home Automation and Robotics Management System)

Description:

This project is meant to explore the creation of an open platform to bring personal robotics and their automation beyond just that of a simple vacuum cleaner. NANNY will focus on the interaction of multiple robotic appliances and humans within the house through an interactive interface accessible through a wireless network.

The idea of having one personal robot try to do a multitude of things is slowing the proliferation of robotics into the home. Nanny will support the idea of a home that is connected together by a multitude of devices that specialize in specific tasks. You don't cut your food with a spoon, nor drink with a fork. Why should we expect a robotic vacuum cleaner to be able to make your dinner or keep your house secure.

I believe that the panda board is an ideal platform for the creation of a networked system that is powerful enough to provide a usuable interface for humans both locally as well as remotely, yet small enough to be situated throughout the home much like that of a thermostat or security monitoring device.

Using an open software platform, creators of other robotic devices incorporating the NANNY platform would be able to tap into NANNY's mind, and they would be able to share each others resources.

Project/URL (if exists):

http://nanny.medelman.net (future page on personal website if chosen)

Expected results:

Ideally a platform for you and your family that will automate and ease the workload in your life.

While the goal may seem lofty, with some of you having visions of HAL9000 running their home, I feel that in order for the majority of humans to accept robots into their lives, they need to feel like they serve a greator purpose other then just being an amusement.

There is still much progress to be made in bringing more automation into the home, it is inevitable and standards will need to be created in order for devices to communicate with each other and their owners.

With that notion, I aim to bridge the gaps and overcome the barriers of bringing robotics into the home. Someday everyone will have a "Rosie the Robot", but we have to start somewhere.


Background & work by projects submitter/s:

I recently graduated with my Bachelors in Electrical Engineering, but I have 8 years of experience in software development including work that involved a WinCE based kiosk device that users interacted with. Since graduating with my E.E. degree, I work in the medical device industry creating solutions for devices that ultimately save people's lives. I have experience in embedded systems that require human interaction and must meet strict safety guidelines set by the FDA. My personal interests lie in robotics and automation and I am always looking for ways to improve my skill set now that I formally earned the title of electrical engineer.


Contact email:

Community PandaBoard Documentation

Title: PandaBoard Documentation

Description:

This project aims to develop extensive documentation regarding Linux support for the PandaBoard at elinux.org wiki.
The documentation will be similar in content and style to the BeagleBoard documentation.
This documentation will complement any future documentation posted at pandaboard.org.

Project/URL: http://www.elinux.org/PandaBoard

Expected Results:

Completion of a comprehensive PandaBoard Beginners Guid.
Multiple contributors working on wiki pages.

Contact Email:

Home Multimedia Manager

Home Multimedia Manager:

Description:

This project will support a multimedia manager based on Android OS in order to play HD video and audio for Home and office environments.
The system will be able to play music, video and photos to HDMI/DVI display devices such as flat panel LCD computer displays and digital
projectors. The system will support USB, SD,MMC or Bluetooth connection to external devices in order to play different multimedia files.
Furthermore the system count with an internal mass storage device in order to give the user the possibility to create and manage
its own multimedia library. Due to fact that the system is based on Android OS and panda board has wire and wireless network connection
give the user the opportunity to connect to the Internet and watch HD multimedia from youtube, Google TV and Pandora among many others.
Android will give the users the freedom of download new applications on this system and a new branch of innovation is open to the developers.
All these give to Home Multimedia Manager a wide versatility to develop new multimedia technologies.

Project/URL (if exists):

http://homemultimediamanager.webs.com/

Expected results:

To develop an android application with the capacity to play, store and manage pictures, audio and video in hight definition.
This application would be able to play the multimedia files from external USB, SD or MMC in the beginning and would be able to connect to any HDMI/DVI device.

Background & work by projects submitter/s:

Developing audio drivers (ASoC) for OMAP3 and OMAP4 chips
Develop of initial support for Hawkboard / Omapl 138
Development of Android applications
Testing at kernel level for core drivers on Android and Ubuntu for OMAP3 and OMAP 4
Development and testing at kernel level for core drivers (Keypad and touchscreen) on Ubuntu for OMAP3 and OMAP 4
Specially eager to develop new embedded systems

Contact email:

Port of Android Froyo 2.2 to PandaBoard

Title: android-pandaboard

Description:

This project will aim at working on adding support for running Android 2.2 on the PandaBoard.

Project/URL (if exists):

Not yet, but I probably will use a TI supplied repository or a gitorious hosted project repository.

Expected results:

Initial baseline port of Android 2.2, with core features enabled, that will allow others to work on PandaBoard derived custom boards running android for custom requirements.

Background & work by projects submitter/s:

I started work on porting Android in Nov 2008. I ported Android 1.5 to the Gumstix Overo Earth platform sometime in March 2009 and published the sources for download in June 2009. Recently I have also ported Android Froyo 2.2 for the same platform.
A link to the old port can be found here: Initial port of Android 1.5r3 for the Gumstix Overo platform.

Contact email:


Multimedia system for cars

Title: BeaCar (not the final name)

Description:

The goal is to design a multimedia system that fits in the place where you have your radio/cassette player in the car.
It will have support for Bluetooth, Wireless LAN, FM radio and will play all kinds of video and audio. Based on an embedded debian GNU/Linux it consists of very lightweight components and a GUI optimized for small touchscreens. Possible and mostly planned features are audio/video playback, surfing the web, connect various USB gadgets, navigation or other stuff that would be useful in a car like a connected rear camera.
Because the PandaBoard has most hardware features the project needs espacially the FM radio, it would perfectly fit as the hardware plattform.

Project/URL (if exists): http://xengi.ath.cx/beacar (still under contruction)

Expected results:

While in developement phase I expect many Wiki pages describing how to get debian stuff working on the PandaBoard.
As the final system I expect a Linux multimedia system for cars that you can't live without.

Background & work by projects submitter/s:

I'm an apprentice for a IT specialist in a company in Potsdam, Germany. I work with different Linux systems for 5 years and I'm interessted in ARM devices for more than a year. My skills for embedded ARM devices are not expert like but I'm very interessted and have fun working with it.
I want to create a small community around the device so everybody can help to improve things and add ideas. I believe there are many experienced people out there who want such a thing in there car.

Contact email:

Wine on ARM

Title: Wine on ARM

Description:

Windows apps are mostly compiled for x86 and they won't run on ARM with bare Wine, also there are WinCE apps which should run with some modifications to Wine and maybe the Kernel.
This results in two ways to go:
  • use the finished port as a base for something like WineMobile, which might be able to run PocketPC/WinCE/WinMob apps. Actual Problem is that all CE apps come along without relocation table, so they should be loaded at their baseaddress or we have to do some relocation magic. (i already had some success running wince apps)
  • run the assembler parts of x86 in qemu and call function in Wine compiled for ARM, which might speed up some things (darwine tried that before, but they stopped because the byteswapping (big endian <-> little endian) was too much work, but ARM is mostly little endian, so we have a chance if everything works fine with packed structures. I already gathered some information how darwine tried to do it)

Project/URL: http://wiki.winehq.org/ARM

Expected results:

Try hard to push both descriped things forward.

Background & work by projects submitter/s:

I'm coding since 9 years, the used language and skills of course moved through this time. At some point i started coding for Wine to enhance my skills. This month i sent my patches for ARM support and they got accepted, that was great as i spend a whole year(not fulltime) on that little piece of code. So now it's time to do something more usefull with the ARM port of Wine.

Contact email:


Gentoo Linux support

Title: Gentoo Linux support

Description:

Gentoo Linux is one of the Linux distributions that currently support ARM. It was first supported in 2003 and development hasn't stopped since. With so many ARM devices being released to the market recently, we like to support it in our distribution. Therefore i'll provide documentation for installing Gentoo Linux and prepare everything needed for getting it working. I'd also like to help OMAP developers testing new fixes. The board will also be used to test packages and build stages(root filesystems).

Project/URL: http://www.gentoo.org

Expected results:

Full support for the pandaboard on the distribution, with install guides and everything needed to get it working.

Background & work by projects submitter/s:

I've been a Gentoo developer since 2007, i'm the lead of the ia64/sparc teams and a senior member of the alpha/arm/sh/s390/x86 teams. My work on the distribution is making sure the packages work fine and provide support for whatever i can help with. I've written install guides for all the embedded hardware i've had in my hands.

Contact email:

Port Amahi to PandaBoard

Title: Port Amahi to PandaBoard

Description:

Port the Amahi Home Server to PandaBoard using the T2SDE distrbution.

Project/URL (if exists):

N/A (yet!)

Expected results:

Add to T2SDE the PandaBoard with kernel and uboot patches.
Add Amahi into T2SDE.
Complete integration PandaBoard and Amahi into T2SDE.

Background & work by projects submitter/s:

I'll preface this by saying that I am a software engineer, with 10 years experience in Linux .

Contact email:

WL1271 support on mainline Linux kernel

Title: Linux mainline support for the wl1271 WLAN chip

Description:

The idea is to make sure the wl1271 chip works out-of-the-box with the mainline Linux kernel. There is always some work to be done in order to get the chip to work correctly, such as setting the MUXes correctly in the board files etc. Also, minor differences always exist between different platforms, and these details can cause problems, so someone needs to test and make sure all the usual testcases pass also on the Pandaboard.

Project/URL (if exists):

No URL for this specific project. But the "official" (and unfortunately very outdated) official page for the wl1271 Linux kernel driver is: http://wireless.kernel.org/en/users/Drivers/wl12xx

Expected results:

Full support for the WLAN chip on the mainline Linux kernel, thus automatically adding support for it in all distributions that use recent Linux kernels.

Background & work by projects submitter/s:

I'm the original creator and maintainer of the wl1271 driver in the mainline Linux kernel. Lately I've been working closely with TI on this project. At the moment I'm working with Beagle board support for the wl1271 chip in the mainline kernel. Panda board also needs this kind of support.

Contact email:


Panda Game Console

Title:Panda Game Console

Description:

Project involves running games on a existing/ported Ubuntu distribution which can be controlled via Bluetooth HID joysticks, connected to a TV via HDMI.
The idea is to pick up the existing open-source stack BlueZ to enable the Bluetooth chip WL1271 on panda, make use of the Ubuntu ports or port Ubuntu for panda, thereby providing an out-of-the-box product for which panda can be used.

Project/URL (if exists):

Expected results:

Project would involve integrating various open-source components like the latest BlueZ stack support, HDMI display and Ubuntu. An HID solution would be implemented for the joystick on top of the existing HID support on BlueZ.

Background & work by projects submitter/s:

Have been working towards upstreaming Wi-Link Combo devices driver to mainline Linux, and have been working with a leading Indian IT service provider as a contractor to a leading semi-conductor company. Have been involved with Wi-Link devices and OMAP platforms since 2008 as part of my day job.

Contact email:


Symbian for PandaBoard

Title:Symbian for PandaBoard

Description:

Symbian is the largest SmartPhone device operating system on the planet, and available on the most handsets. It is now fully open source.
Now that we have a fully open source PDK, we want to use an open source device to enable open source handset and other project development on the platform. This has already been done with BeagleBoard, as part of our Wild Ducks project, and we have the PDK running on other reference platforms and simulators so far. It would be fantastic to harness the power of OMAP 4, in addition to taking advantage of much higher degree of component integration on PandaBoard.

Project/URL: http://developer.symbian.org/wiki/index.php/Wild_ducks_project/About http://developer.symbian.org http://www.symbian.org

Expected results:

Porting of Symbian to PandaBoard. Availability of all new Product Development Kits for PandaBoard. The design of reference hardware designs to augment PandaBoard functionality.

Background & work by projects submitter/s:

See the Wild Ducks page for progress with BeagleBoard so far.

Contact email: (and others)

Panda 3D Video Recorder

Title: A 3D-Video Recorder

Description:

A 3D video recorder using panda board and two cameras interfaced with pandaboard.
3D video recording is probably in its nascent stages and a open platform on this front would open up amazing possibilities, like say, 3D chat?

Project/URL (if exists): None yet. But will put up one soon.

Expected results:

The idea is to develop a 3D Video recorder with two identical cameras interfaces to panda board and record videos simultaneously from both the cameras. Software on pandaboard will then 'stitch' both the videos to produce a 3D video output. The video can then be recorded onto media or even played out on HDMI.

Background & work by projects submitter/s:

Would like to call myself a ARM/ Linux hobbyist. Have been playing with Beagleboard and Hawkboard; built myself a media server on Beagle. Would like to explore the possibilities of more such interesting stuff using open source hardware/ software.

Contact email:

Network DVR

Title: Network DVR for IP Cameras

Description:

Integrate the existing open-source projects (GStreamer, ffmpeg, Live555 ...) to create a reference software platform to record audio/video from IP cameras, convert it using omap hardware acceleration, perform video analytics (motion detection, face recognition...), remote control with a web interface, view it on screen. Core functionality is already implemented in the mentioned open source projects, "glue code" is what is missing.

Project/URL (if exists): None

Expected results:

Create a reference software platform (openembedded based) and develop one or more "Controllers" to integrate all the specified functionalities.

Background & work by projects submitter/s:

I have worked on system integration for video surveillance for three years on x86 and arm achitectures. I have already developed an opensource project, MoBlock.

Contact email: [.]de

Headless Home Server

Title: Home Server accessible though RDP

Description: The goal of the project is two fold:

The server, running on the PandaBoard, provides the RDP service. RDP (Remote Desktop Protocol) is now an open protocol. Devices that connect to the server have their own X11 instance and the UI (optionally with sound) is forwarded to them through RDP. Low compression / high bandwith video travels fine on RDP. Uses cases are:

Project/URL (if exists): None yet.

Expected results: xrdp, a Linux RDP server already runs on ARM. First:

Then:

RDP clients exist for almost any platform, including Android & iPhone. The proposed setup greatly simplifies access to home resources since administration only takes place on the Pandaboard. The multi-core feature of the chip is also essential.

Background & work by projects submitter/s: 10 years experience SW engineer, currently in the digital video field. Minor contributions to a few projects. Proficient with (embedded) Linux.

Contact email:

Clang/LLVM Support for OMAP4

Title:

Clang/LLVM Support for OMAP4

Description:

This project involves adding C-language support for OMAP4 processors to the Clang compiler, with initial focus on supporting 100% correct generation of ANSI C code. Eventual goals include support for C99 and C++.
It is crucial for the community to have powerful development tools at its disposable that are not encumbered by v3 of the GPL. This is especially important if the PandaBoard hopes to see adoption in commercial products, where companies need the ability to ship software while not being encumbered by GPLv3's patent licensing requirements.
A high quality, modern C compiler is the first step towards laying a powerful and free foundation for development on the PandaBoard platform. By targeting the BSD licensed Clang compiler, we're able to do this without alienating commercial interests and software projects that actively avoid including GPL'd code, such as FreeBSD.

Project/URL:

Development will be hosted on GitHub.

Expected results:

A working implementation of Clang that can compile executable ANSI C code for the PandaBoard.

Background:

Computer science student with several years of low-level Linux/BSD experience. Some experience with implementing a compiler on top of LLVM. Numerous contributions to FLOSS projects, mostly consisting of small bug fixes.

Contact Email:


Veggie growing web-controller

Title: Senior Pomidor (Senior Tomato)

Description:

The idea is to create an application for standalone self-controlling system that grows plants.
It is going to have a few sensors and meters (ph, temperature, humidity, etc), full control to power system (turn on and off different devices like lamps), water and even special chemicals.
Ultimately, the system should guide, teach and alert novice grower. The user interface is going to be helpful and friendly to quickly show status of the plants, schedule and adjust lightning scheme, control watering, etc. Web camera is going to track all things happening in the garden to have a way to see growing evolution and have appropriate photographs to show with alerts back in time.

Expected results:

Fully automated full-circle system that helps and does growing of various plants, vegetables or flowers. It will teach, journal and decide things for you if you're away or let you control the process. Gathering data every day will help to not make same mistakes and keep your veggie pets healthy and happy.
Technically, it's going to be a linux web server with ruby 1.9 on board running lightweight framework (like this) to handle different requests. Beautiful web application with vector graphics user interface and charts (computed on clientside) aimed on modern browsers will deliver amazing and pleasant experience easily compared to well thought out Mac application.

Background & work by projects submitter/s:

I'm an expert clientside web developer originating from Russia, now living in Bali, Indonesia. Doing scientific web applications for my day job and a javascript user interface framework as hobby . A few years ago I started building growboxes for different plants (tomatoes, lemons, ethnobotany) in a free time and got quite hooked on the thing. But after a few frustrations and various failures I decided that things can go better with automated system that knows-it-all and can actually record your decisions on the go so you don't screw up again like that. I have a friend that thinks playing classic music for plants would help a lot. Also worth a try. The thing suits me ideally because it's embeddable even in small room and can use quite a few devices with a possibility to scale out (real monitor, touch screen).
People in medicine have their ipads, we, growers, can have nice things too! We don't need portability that much, so no touchscreens. But being able to reach control from far away is very important (yay for web apps!).
Personally, this project is very important to me. I always wanted to tinker things more closer to real world, but so far I only did things for internet. This is ideal fit for me, because it's powerful enough to run my language of choice (ruby, where we need a lot of memory, but 1gb is godsent) and act like a real linux server. Awesome stuff!
All things will be open source from day 1 (I have quite a lotof that stuff), I'm pretty sure that I can build a platform for people to use to build similar applications.

Contact email:

Hexakopter

Title: Hexacopter project

Description:

Our aim is to create a hexacopter that is both open source and hardware. Our goal for the hardware is that we would be able to lift about 1 kg and have about 30min of flight time. We would need a powerful embedded system to stream video down and to do all the necessary calculations to keep it stable. The whole project is built as a platform for learning about the different aspects of technology needed for the project.
There is alot of hexakopters and quadkopters out there but none that was both open source and cheap to build. Although hexacopters are mostly used as toys this one could be used to film professional material in HD.

Project homepage:

hexakopter.fik1.net

Expected results:

A working hexakopter with a camerasystem and streaming capabilities.

Background & work by projects submitter/s:

We are two IT stundents from the Finnish School Arcada. Our project started after we saw a video of a working hexakopter and we thought we could build something bigger and with more functionality. After our school granted us some funds for the project we got even more determined to finish the project. We have been playing around with a Chumby but we soon found out that its not fast enough for video streaming. Our experience lies in electronics and media.

Contact email:

Feasible Mobile Augmented Reality Toolbox

Title: Feasible Mobile Augmented Reality Toolbox

Description:

Couple of high-end RISC cores and real time DSP on-board make implementation of precise 3D reconstruction from video sequence on PandaBoard plausible. Having such an algorithm will open perspective on intersting applications like augmented video calls etc.
We are going to use following ideas to make it happen. Computer vision algorithms are difficult to write, debug and maintain. We propose to develop special-purpose extension to the Commmon Lisp language in spirit of "sheet" language from the paper "Programming language support for digitized images or, the monsters in the closet". What we are going to make different is to target the DSP core not the general purpose one. This will allow us to off-load computing of image processing kernels from the ARM to DSP (and graphics accelerator in future).
We are going to use recently announces ARM port of Clozure Common Lisp as our vehicle.
Coding complex vision algorithms in Lisp have well-known advantages over other "conventional" languages e.g. dramatically shortened the edit-compile-test cycle which is invaluable if one targets best speed to market of their applications. None the less we do not tie users to one language. Clozure Lisp provides excellent foreign functions interface.
Current mobile augmented reality libraries like ARToolKit lacks precise positioning from the video sequence (with or without not-so-accurate GPS). Markers are used at the present. These are showstoppers on the way to augmented reality "in the wild". We want to fill the gap in order to enable markerless applications

Project/URL: []

Expected Results:

Minimal compiler and run-time for our "sheet"-like language enough to program modern feature detection algorithm like SURF: Speeded Up Robust Features.
Implementation of such an algorithm. SURF or SIFT
On top of these it will be possible to write 3D reconstruction algorithm. SMP opens new perspectives with respect to optimization here.
These will be enough to bootstrap.
This project is quite ambitious, we hope to make minimum valuable implementation ASAP and grow it later. ;-) It will not be possible to use all the features of the versatile CPUs right from the start. We want to show soundness of the approach in fact.

Background & work by projects submitter/s:

I, Yakov Zaytsev, do system and DSP programming in C and Erlang for living in a small company. I study automatic control in the Baltic State university. I started this project in stealth mode on github few month ago. ;-) I learned augmented reality field reading awesome Zisserman's Multiple View Geometry in Computer Vision a year ago. Programming tms320c64x+ (and other floating point special purpose processor) is my current paid activity. I'm excited about PEAP and I have decided to make my project public therefore.

Contact Email:

Lego Robotics Club Interface

Title: Lego Robotics Club Interface

Description: I run a Jr FLL Lego Robotics Team at River Place Elementary in Austin, TX. The board would be used as a teaching device to show students highly integrated, powerful while power-mindful performance computing. The initial goal would be to use the device as a platform to program the Lego Mindstorms NXT controllers. More advanced goals would be to use the board in conjunction with an lcd display to provide an interface with the robot, possibly data logging equipement, etc., ... really whatever the students can dream up and we can help implement.

Project/URL: http://firstlegoleague.org/

Expected results: Initial results would be to get the board functional with a linux-based OS then use available software to interface with the NXT controllers. This would allow students to easily program their robotic actions. It would be useful to demonstrate an inexpensive platform for other NXT robotics clubs/enthusiasts. Longer term results would be to build an "app" interface for robotics projects. This board is ideal for Lego Robitics given its size and performance [imagine a Lego robot trying to carry around a laptop :-) ]

Background & work by projects submitter/s: I have an MS in EE and Computer Engineering and 14+ years professional experiance in high tech industry. This is my first year as a FLL team coach and have formed a team from the neighborhood's local elementary school children and parents. My personal goal is to grow with this team over the years for full robotic team competitions in High School. I have a co-coach with 20+ years programming experience including OS's, applications, and web-based apps.

Contact email:


Port of OpenCV ComputerVision Library

Title:Port of OpenCV ComputerVision Library for PandaBoard

Description:

OpenCV is a popular computer vision library which is very useful for image processing and recognition.
Applications include face detection, Optical character recognition, object movement tracking etc..
This library is currently optimized for x86 architecture. The proposed project intends to port OpenCV to PandaBoard thereby allowing community to build OCR,Face Recognition and object tracing applications. With the availability of ARM SMP processor and camera interface in PandaBoard, the applications are endless for with OpenCV.
This project intends to fill the gap of Free Open Source optimized image processing library for ARM based processors.

Project/URL :[]

Expected results:

As a first step OpenCV would be ported to PandaBoard without any optimization and would be applied to simple image recognition.
The next step would be to optimize this library to fully utilize the ARM SIMD instructions and DSP capability present in PandaBoard. This shall improve the performance of OpenCV in an ARM based target.

Background & work by projects submitter/s:

I had created a Free Open Source OCR gTamilOCR. Refer Website[[3]]. This was created with a different set of image processing library. I am currently porting this application to use OpenCV to use SIMD optimized library with multiprocessing capability.

Contact email:

Port XBMC for Omap4

Title: XBMC

Description:

XBMC is a popular media center which provides the users the power to stream nearly any content to the client over many protocols. The graphical user interface is extremely versatile and allows a skinner to change the look and feel of the program completely. XBMC is designed to be powerful and give the user the ability to manage and watch the media while still being very user friendly. XBMC's media players are known for being advanced and uses advanced algorithms for scaling and syncing that rivals the upper segment of consumer electronics.
XBMC has been ported to ARM and have been shown to work on the Omap3 platform (beagleboard) and on Tegra2 platform. The problem XBMC has faced on both of these platforms is not being fully able to do 1080p. This is why a project to port to Omap4 would be interesting. XBMC was a GSoC for the BeagleBoard community this year and proved that while its viable to run XBMC on it, I was even able to get it to run the GUI in 720p fluently with a specialized skin.
The Omap4 (pandaboard) would be a great stepping stone for XBMC as it would be the first hardware which is ARM based that could actually be used to power fully HD capable set-top-boxes (or TV's with Omap4 in it) with XBMC as the frontend.

Project/URL (if exists):

http://xbmc.org/
http://elinux.org/BeagleBoard/GSoC/2010_Projects/XBMC
http://xbmc.org/topfs2/
http://forum.xbmc.org/showthread.php?t=35139
http://forum.xbmc.org/showthread.php?t=71297

Expected results:

Tap into the 1080p powers. Depending on the GPU power create a skin which is able to make XBMC on Omap4 fully 1080p able.
What PandaBoard would gain from XBMC is the ability to base a set-top-box on omap4 and allow XBMC to be the presenter and manager for said system.

Background & work by projects submitter/s:

I was a part of the GSoC for BeagleBoard and I was able to make XBMC dirty region render based, tap into the yuv to rgb hardware overlays (which made 480p viable). Have been a project member in xbmc for over 3 years. Forth year student in computer science.

Contact email:

Multimedia Streaming with Context Switch

Title: Multimedia Streaming with Context Switch

Description:

This project based on the Mobile Phone multimedia streaming connected through different pairing devices like wifi, bluetooth and wired devices with context switching.
The main feature of this project is when user wants to play the audio it saves the context while switching between display systems as well as music files.
If you enters into home with your Mobile Phone it automatically connects to home wifi and play the audio/video on your TV/home theatre system/PC.
When you entered to car it uses bluetooth and by using streaming it plays the audio/video on car multimedia system.
By using Image scaling, color conversion, dynamic audio post processing we are converting the video to support any existing display system.
Dynamic audio post processing depends on the connected audio system. Should be able to choose preset equalizer.
If we have FM transmitter we can use it for playing music in car FM radio.

Project/URL : none

Expected results:

Capacity to play all video formats upto full HD resolution over any display system based on the display resolution and color format.

Background & work by projects submitter/s:

Initial support for Omap4(blaze), Samsung s3c6410.
Camera interface in middleware and hardware level.
Integration of multimedia products on android and symbian.
Worked on multimedia codecs.
Audio routing and audio post processing.

Contact email:

Car Multimedia Center

Title:Car Multimedia Center

Description:

I haven't found any Double-Din-Radio with some kind of a mobile OS. So i thought, you could attach a board like this to the rear-camera-input of a radio with touchscreen.
Because no radio i have found has a touchscreen-"output", i had to mod an existing radio - maybe you can attach a touchscreen-controller with a loadfree amplifiercircuit,
so that this one touchscreen delivers the signal to a standard-ts-controller without disturbing the existing controller. If anyone would give me the money, for a RNS-510 i would have
an almost existing solution due my work.

Project/URL (if exists):

Expected results:

So in the end, i would like to have a free usable pc in my car. You could install free software and/or with a can-bus-interface, you could use all the OBD-messages for a graphical
"trip-computer" with some extra features and diagnosys.

Background & work by projects submitter/s:

Thats maybe some kind of problem. Until now, i have just worked with windows embedded on X86-systems and low-level-coding on some microcontrollers. With arm-hardware i would need support or :documentation for this. In the next time, i will beginn a job as a system integrator as my college-time is nearly over. So then, there will be some more knowledge in this scene.
So, yes, I am a beginner, but that may change.

Contact email:


Android School Tablet System

Title: Android School Tablet System


Description:

This project try to avoid the current repetitive educational system, a big fail. The success is knowledge sharing.
This project has a UI interface and services to share and expose problem solutions inside class (or a limited network, you can choose your classmates) exposing data streaming as educational platform.
The Android School Tablet system will be able to share documents. multimedia sound/video and data in different pairing devices like wifi and bluetooth, meanwhile the system is able to synchronize the data with all your classmates.
So, Android School Tablet system need applications to manage the camera, audio in/out system, a service/client to expose/read a streaming hereby an application to simulate a blackboard in real time. Also It needs applications like progress trackers, to do tasks, email readers, wikipedia queries & reader, multitouch apps, notebooks, calendars, etc, etc.
Other feature of this system will be the ability to install modules for that educational platform thanks to an application server, from module to listen and share music, till modules to develop code together, to make quizzes, to print reports and much more.
For this project is necessary (I think so):
Porting Android Froyo to Panda Board
Gstreamer in Android
Camera interface middleware/hardware.
Multimedia codecs.
Android applications.
Server/Client streaming app.
Application main server & web services.


Project/URL : none


Expected results:

The main feature of this project is all the interaction with Professor and student or everybody to desire share data with different people on an educational platform.


Background & work by projects submitter/s:

Android applications.
Multimedia Applications Android & Gstreamer.
Client streaming apps.
Automation and Industrial real time systems on QNX.
Application server & web services.
GNU/Linux desktop applications.


Contact email:

Android Security Car System

Title: Android Security Car System


Description:

This project is based on the current crash prevention automotive market.
The Android Security Car System will be able to recognize objects around the road/highway and take decisions to prevent crash accidents.
The system has many camera extensions and is able to recognize what object is near/far, it can calculate the speed who the object has and if the object is dangerous or has a imminent crashing state the car can take decisions like stop, activate air bags security system, etc, before the impact. For that the system will be able to recognize images trough the cameras and to know if the object is a human, a car or other stuff, all of this to know if It can active or not the security system.
The system has applications to know what is the state of the road to take decisions to keep the stability of the car, thanks to the sensibility of the accelerometer. In case of crash accident the system should take decisions and keep the stability of the car.
The system will keep in memory (black box, hard drive, etc) all the information log after a crash accident to help to understand what was the cause.
For this project is necessary (I think so):
Porting Android Froyo to Panda Board
Accelerometer interface.
Camera interface.
Multimedia codecs.
Develop or port an Image recognition system.
Android applications.


Project/URL : none


Expected results:

The main feature of this project is the image recognition and stability control for the car in case of a crash accident.


Background & work by projects submitter/s:

Android applications.
Multimedia Applications Android & Gstreamer.
Client streaming apps.
Automation and Industrial real time systems on QNX.
Application server & web services.
GNU/Linux desktop applications.
MS Windows drivers.


Contact email: ,


Port of Robot Operating System (ROS)

Title: ROS

Description:

ROS is an open-source, meta-operating system for your robot.
It provides the services you would expect from an operating system,
including hardware abstraction, low-level device control, implementation
of commonly-used functionality, message-passing between processes, and package management.
Port base system and relevant stacks from http://www.ros.org to OMAP4 architecture. ROS is structured in so called nodes that communicate via sockets. Each node can perform a complex task (mainly data filtering) therefore ROS can already leverage the power of a dual core processor. Most nodes heavily deal with geometric transforms and streamed data processing, so some core routines should be extended to use the OMAP4 DSP. OpenCV is also integrated into ROS but handled in a separate proposal on this page.

Project/URL (if exists): http://www.ros.org/wiki/pandaros

Expected results:

Background & work by projects submitter/s:

Contact email:

Audio Player with Digital Room Correction

Title: Audio Player with Digital Room Correction

Description:

This project aims to detail the procedure to create a high quality audio player for the living room.
In a first phase, the objective will be to put together all the pieces of software required (remote player daemon, high quality resampling, convolver, ...).
In a second phase, it will be tested if the system can keep up with multiamplification.
In a third phase, it will be studied if it is possible to get the convolver to use the DSP.

Project/URL: TBD

Expected Results:

To create the documentation describing the procedure.
To create an integrated distribution.
To enable other people being able to enjoy a high quality audio player.

Background & work by projects submitter/s: Aerospace engineer. Using linux for more than ten years. First experience with Arm platform. Not a software engineer, but I have some experience as a developer (python). Several years involved in the the world of DIY audio.

Contact Email: josemaria[.]

Creating an iSCSI-bootable distribution

Title: Creating an iSCSI-bootable distribution for OMAP4-based boards

Motivation

ARM-based boards are known as low power consumption hardware
SATA / IDE is missing on PandaBoard

Description:

Like the LTSP (Linex Terminal Server Projekt) it should be possible to use ARM-based machines to boot from network an ARM-based distribution
Unless every Linux Distribution with ARM-based images (Ubuntu, Gentoo etc.) should be bootable, the first distro to be adapted adapted for iSCSI-boot will be Gentoo for best test-abilites. After proof of concept all ARM-based distributions should be easily applicable for netboot on PandaBoard-based hardware.

Project/URL: will come if board arrives ;-)

Expected results:

First of all we will proof the network-bootability of Pandaboard
Adopting u-boot arm-bootloader / ethernet-based boot for ARM-based board (WiP)
Adopting boot process of Gentoo OpenRC
Transfer knowledge for Booting ARM-Ubuntu
Maybe it could be interesting using Wine on ARM for starting WinCE-applications via network ;-)

Background & work by projects submitter/s:

Using Linux/FreeBSD since 1997
Working for Gentoo-Community since 2003 (Founder of Munich Gentoo Linux User Group)
Long experience in netbooting
Long experience in virtualization (e.g. kvm, RadioTux interview)
Experience with Beagleboard-clone IGEPv2)

Contact email:

RObot ModULes on USb

Title: ROMULUS

Description:

Most of the micro controllers nowadays supports USB OTG connectivity. With USB HUBs
and cables we can simply deliver power and communications with the main "brain" (PandaBoard)
to the remote modules (sensor readers, motor drivers or other different kinds). A set of
Linux drivers and board blueprints would facilitate robot assembly. Using “modulespace”
libraries a module can be simply connected and recognized by the OS and then used trough a
userspace library that automatically detects the connected modules and their capabilities,
which facilitates main logic implementation and connectivity with the world outside, which
is enabled by the Wifi and Bluetooth chips on PandaBoard.

Project/URL (if exists): none

Expected results:

The result would be a set of basic Linux drivers, communication protocols, userspace
libraries, Arduino/PIC frameworks for ease of use and a proof of concept implementation of a
robot, that would provide the blueprints and examples necessary for the newcomers.

Background & work by projects submitter/s:

Open source enthusiast
Gstreamer contributor - http://gstreamer.freedesktop.org/
Former member of Cyberpipe team (a slovenian hacklab and much more) - http://www.kiberpipa.org
Video filter development for Gstreamer
Codec and multimedia protocol implementations for Gsteramer
Cassio:pia art project (software and hardware developer) - http://www.kiberpipa.org/index.php?name=News&file=article ,
Robotics hobbyist

Contact email:


MiniMyth

Title: MiniMyth


Description: MiniMyth is a small Linux distribution that turns a diskless computer into a MythTV frontend and slave backend.

A MythTV system is composed of a central backend that retrieves TV program listings and coordinates recordings amongst itself and various slave backends. A frontend system provides a user interface and retrieves and displays recordings, photos, CDs/DVDs/BluRay. MiniMyth however integrated various other components, like Internet video (YouTube, etc.), Console and Arcade Emulators, a web browser, UPnP player, alphanumeric LCD interface and others.

The MiniMyth project compiles a specific Linux distribution and all software components from source. It is designed to run on a diskless client, booting from the central server via PXE.

The latest version of MiniMyth contains:

   * MythTV 0.23.1 with fixes through SVN 26057, including the following plugins:
         o MythBrowser
         o MythGallery
         o MythGame (not integrated with MiniMyth configuration)
         o MythMusic
         o MythNews
         o MythVideo
         o MythWeather
         o MythZoneMinder
   * MythStream mythtv-r21640
   * MPlayer SVN 32050 and FFmpeg SVN 25042
   * MPlayer 1.0rc2 with patches to add
         o XvMC-VLD hardware acceleration
         o DVD navigation fixes
   * VLC 1.1.4 and FFmpeg SVN 25042
   * xine-ui Pre 0.99.7 (HG 3056), xine-lib Pre 1.2 (HG 11577) and FFmpeg SVN 25042
   * Hulu Desktop support (MiniMyth does not include the Hulu Desktop binary)
   * Adobe Flash Player support (MiniMyth does not include the Adobe Flash Player binary)
   * FCE Ultra 0.98.13-pre for Nintendo Entertainment System emulation
   * jzIntv 1.0-beta3 for Intellivision emulation
   * Mednafen 0.8.D.2 for Game Boy, Game Boy Color, Game Boy Advance and Nintendo Entertainment System emulation
   * Stella 3.2.1 for Atari 2600 emulation
   * VisualBoyAdvance 1.7.2 for Game Boy, Game Boy Color and Game Boy Advance emulation
   * ZSNES 1.51 for Super Nintendo Entertainment System emulation (32-bit builds only)
   * Xorg 7.4+ (Xorg Server 1.6.5 and Xorg packages with which it works)
   * openChrome Xorg graphics driver SVN 856
   * NVIDIA Xorg graphics driver 256.53
   * LIRC 0.8.6
   * LCDproc 0.5.3
   * Linux 2.6.34.3 with patches to add
         o ALSA 1.0.23
         o Unionfs 2.5.5

Project/URL: http://www.minimyth.org/

Expected results:

Compiler configuration for MiniMyth Linux and graphics drivers
Hardware interface for infrared remote control
Interface configuration for LCDProc (alphanumeric LCD)
Optional: Xorg running on attached LCD

Background & work by projects submitter/s:

OpenSource enthusiast
Linux follower since 1991
Some 30 years of experience in developing hard/soft/firmware projects
Compiled and adapted MiniMyth for various Notebook configurations

Contact email:


Flight Simulator Glass Cockpit controller

Title: Flight Simulator Glass Cockpit controller

Description: There are various projects to setup glass cockpits (EFIS and Garmin 1000 simulator) for MS Flight Simulator or for X-Plane. However, they usually involve various controllers, DisplayLink devices, LCD screens, and so on.

It would be very nice to have a single device that handles Button- and LED/Lamp I/O as well as screen output and that could be connected to MS Flight simulator by a single USB cable.

One of the challenges would be to provide a second monitor output through a driver that implements DisplayLink-like features (additional monitor integrated in Windows-Desktop, transmitting compressed video data through USB). Actually, that is the main concern right now. Any solution?

Another project would be a Garmin 1000 simulator for X-Plane.

Project/URL (if exists): None yet

Expected results:

One or two LCD screens that work as separate monitors in MS Windows
Interfaces for buttons and rotary switches that show up as HID devices in Windows
Interfaces for digital lamp- and LED-output addressable from MS Flight simulator
G1000 simulator for X-Plane

Background & work by projects submitter/s:

Private Pilot
OpenSource enthusiast
Linux follower since 1991
Work with MS Flight Simulator since it was first released on Apple ][. Back in the eighties. Way back.
Some 30 years of experience in developing hard/soft/firmware projects

Contact email:


Convert PanadaBoard Schematics and PCB to Altium

Title: Convert PandaBoard Schematics and PCB to Altium:

Description:

Convert PandaBoard Schematic and PCB to Altium and add accurate 3D parts for all connectors.
There are several reasons why users prefer to use Altium but probably the most important is the close integration between the Schematic and PCB parts. This ensures that there are no discrepancies between the footprints defined in the schematic and those used in the PCB and those defined by the part number. Also, Altium parts database is closely linked to Digikey, Mouser, Newark and others so that BOM can reflect the current available inventory and volume pricing. Finally, Altium is probably the least expensive full featured PCB CAD application available.

Expected results:

A completely synchronized PandaBoard Schematic and PCB in Altium together with accurate 3D parts for all connectors. The Gerber output from Altium will match the Gerber output from Allegro.

Background & work by projects submitter/s:

Previously I converted the Beagle-xM Schematics and PCB to Altium. I also added 3D parts to Beagle-xM board and the complete board with 3D parts was imported into Solidworks. This work was submitted to Gerald Coley.

Contact email:

ARToolkit PandaBoard

Title: ARToolkit PandaBoard

Description:

The goal of this project is to develop an Augmented Reality (AR) platform on the PandaBoard.
Augmented Reality technology has seen widespread interest from industry and the public as a new approach to merge real and virtual worlds, aiding a large range of applications (including gaming, tourism/cultural heritage, architecture, engineering). This technology is increasingly turning to embedded AR solutions, including personal mobile devices, home consumer products, car navigation systems, etc.
We're very excited about using the PandaBoard as a state-of-the-art system for the design and development of embedded AR applications. It's open software stack, small form factor, and wide range of connectors (camera, USB, DVI), creates an ideal platform for AR research. More importantly it's fast multi-core OMAP4 processor and powerful GPU, will allow us to push the boundary of what is traditionally considered possible in portable AR.
Through this project we aim to create a first Augmented Reality platform on the PandaBoard offering an innovative and advanced open-source solution for the AR community. Our work will be based on ARToolKit library (GPL, artoolkit.sourceforge.net), which has been partially developed at our lab. ARToolKit is an Augmented Reality library which supports video acquisition, computer vision processing, tracking, and 3D Rendering (OpenGL). We aim to utilise the multiple cores of the OMAP4, it's NEON instruction set, and hand optimisation, to maximise processing speed, creating impressively smooth and responsive AR applications.
The code and documentation produced will benefit others considering processor intensive projects using the OMAP4. It will demonstrate the power of the underlying chip, and supply sample code and advice on how best to maximise throughput on the OMAP4.
This work will provide a foundation for further long term research & development, optimizing non-marker based tracking (NFT using SURF for example), and other computer vision recognition (OCR, face, etc) as part of an advanced, integrated embedded platform.

Project/URL (if exists): http://www.hitlabnz.org/wiki/EmbeddedAR

Expected results:

* a port of the ARToolKit Library to the PandaBoard
* an optimised version of the library leveraging the possibilities of the PandaBoard (multi-core, NEON code)
* documentation for those wishing to optimise CPU heavy applications for the PandaBoard
* augmented reality samples demonstrating how to develop AR applications on the PandaBoard

Background & work by projects submitter/s:

Through our existing EmbeddedAR project (see previous url), we have ported ARToolKit to the BeagleBoard and other OMAP3 devices (N900, Gumstix), and this project will be a natural extension of this work. Our knowledge and skills in Augmented Reality development is extensive, including proficiency in developing AR applications, software, tools and libraries. We are a world-recognised public AR research lab; the developers of the OSGART library (www.osgart.org), heavily involved in supporting and contributing to the ARToolKit library, with experience optimizing and porting OpenCV to the N900 as part of a high-level AR Framework. Our staff and students are highly skilled, ranging from researchers and software developers, to electronic engineers. We posses all of the skills required to successfully execute the AR PandaBoard project.

Contact email:

Handheld LCD & GPS & Navigation Expansion Board

Title: Openmoko Panda Hybrid

Description:

We are currently working on a new community driven open Smartphone "GTA04" (similar the Openmoko Freerunner GTA02). This is based on the OMAP 3 Beagleboard design and adds UMTS as well as improved navigation services (Gyroscope, Compass).

Since it is obvious that the open source community is greedy to get new OMAP 4 based devices as soon as possible, we would like to start early with an upgrade, even before the OMAP 3 based design is completed and available to the community.

To prepare this step, we have to study the OMAP 4 design early to better understand the differences to a OMAP 3 design.

As a design tool for the OMAP 3 based device, we have built a expansion board for the Beagle Board that provides

On the software side we have adapted U-Boot (http://projects.goldelico.com/p/gta04-uboot/) and the Kernel (http://projects.goldelico.com/p/gta04-kernel) so that the additional components are supported, e.g. for running LXDE on Debian Lenny.

If we get a board for our project, we want to adapt the hardware (if needed), U-Boot and the Kernel to provide a demonstrator for an open OMAP 4 based handheld. If that is successful and the OMAP 4 eventually becomes a catalog product (like the OMAP3530), we want to base an open handheld design project ("Openmoko GTA05") on it.

Project/URL (if exists):

Expected results:

The target is to demonstrate a future OMAP 4 based open smartphone. The long term goal is to provide a new Openmoko device with integrated OMAP 4 CPU.

Background & work by projects submitter/s: We have already designed and produced hardware (see project URLs) and adapted U-Boot and Kernel. We are also in contact with TI field application engineers in Germany and SMD production facilities for a future OMAP3530 based Openmoko device.

Contact email:


Bicycle all-purpose computer

Title: Bicycle all-purpose computer

Description:

The goal of this project is to develop a small piece of computer as a satisfying helping tool for bicycling.
The standard problems when using a bicycle are always the same. Several different (and expensive) devices are necessary to reach a same standard as in a car. Navigation, audio devices and standard informations like speed are off-the-shelf equipments. And nobody would think that it could be missing.
Through this project we create an onboard-unit which integrates several devices in one. On its small display a navigation software helps to reach the desired places. A multimedia part integrates an audio player and perhaps a video part for playing videos of a usb stick. The special hardware part gets into contact to the bicycle parts like speed, temperature and humidity sensor and others. Other prospectice aspects are connectivity per UMTS and a bluetooth connection to a mobil phone with all its functions as used in cars.
This project is a very special device but a breakthrough for millions of bicycle drivers. Several applications are on the market which need several optimizations and a corporate design at the front of the touch screen. This project fits all together and is optimal for this platform.

Expected results:

* touch screen usage
* navigation device (open streetmap based)
* bicycle specific applications (e.g. speed)
* audio / video entertainment
* Internet connectivity
* hands-free equipment
* battery and dynamo hub driven

Contact email:


OMAP4 support in FFmpeg

Title: OMAP4 support in FFmpeg

Description:

FFmpeg is a complete, cross-platform solution to record, convert and stream audio and video. It includes libavcodec - the leading audio/video codec library. It is used by many open source multimedia projects like VLC, mplayer, XBMC, MythTV, ... Although it has already NEON assembler for many codecs but 1080p H.264 (or similar complex codecs) videos real time playback won't be possible without using accelerating functions of the IVA-HD sub unit.

Project/URL (if exists): http://ffmpeg.org, http://www.mythtv.org

Expected results:

Primary Goal: Playback of H.264 1080p broadcast streams using a FFmpeg based decoder in MythTV or any other software using FFmpeg. Support for other codecs and encoding is secondary goal.

Background & work by projects submitter/s:

Contact email:

Fedora-ARM Project

Title: Fedora-ARM Project

Description: The Fedora-ARM project is a rejuvenated effort to bring Fedora to the ARM platform and achieve (perhaps slightly delayed) parity with the Fedora primary architectures. This will provide a large package base (>15,000 packages) upon which other projects can be based; the package mix includes server, graphical, end-user apps, tools and utilities, and many other categories.

Project/URL (if exists):

Expected results:

Background & work by projects submitter/s:

The Fedora-ARM subteam at Seneca College currently manages a 22+ system build farm that's working on the Fedora 13 build. The team includes includes:

Contact email: chris.tyler at senecac on ca

MeeGo-ARM Pandaboard hardware adaptation

Title: MeeGo-ARM Pandaboard hardware adaptation

Description: MeeGo is a Linux-based open source mobile operating system project that combines the efforts of Intel on Moblin and of Nokia on Maemo into one project. MeeGo is intended to run on a variety of hardware platforms including handhelds, in-car devices, netbooks and televisions.

Project/URL (if exists): http://wiki.meego.com/ARM/

Expected results: The plan for the project is to get the MeeGo Core operating system working on the Panda Board. This way we have a cheap/affordable platform to get new contributors to the distribution. The goal is to improve the build process for MeeGo and creating an (unofficial) hardware adaptation.

Background & work by projects submitter/s: Gerard Braad - Software/Hardware engineer, project-lead for the Fedora-MIPS port. Has been using and developing for Fedora and Maemo on ARM devices for a long time (Beagleboard, 770, N810, etc).

Contact email:

Sugar on a PandaBoard

Title: Sugar on a PandaBoard

Description:

Sugar on a Stick is a Live USB operating system featuring the award-winning Sugar Learning Platform and designed to fit on an ordinary USB thumbdrive ("stick"). The most recent version is version 3, code-named Mirabelle and was released on May 25, 2010. The next version, version 4, will be released in early November.

Project/URL:

Expected results:

The plans for the project is to get Sugar working on the Panda Board platform to allow for easy running on cheap, low powered OMAP4 processors which will allow more options for large deployments of computers in the developing world to improve children's educational opportunities.

Background & work by projects submitter/s:

Peter Robinson - I am a core maintainers of the Sugar on a Stick project and am also actively involved in the SugarLabs and OLPC projects.

Contact email:

James renderer

Title: James renderer

UPDATE
Since submitting this project a number of things have been achieved:

Note that this gui is only a concept, there will be a smaller and less complicated gui for e.g. SWAT (and everything will get the same look and feel).
Also due to some issues taking too much time, I was unable to add mediatomb and the webcam yet.
Will try to add the mythtv backend somewhere during this weekend

END OF UPDATE

Description: In the past I created the James project (acronym for Just A Miniature Entertainment System). Aim of this project is to create a system providing a home entertainment solution. However during the project it became clear that it would be a too big a load for a single system to acquire, serve and render content. Also the BeagleBoard used is not capable of doing 1080p.

Therefore it is decided to split off the rendering part. Content acquisition and serving will be done on one system (HawkBoard as it has SATA and ethernet). This system can e.g. be put in a utility closet or the attic.
Rendering and user interaction will be created as part of this project. The pandaboard will act as a renderer. It will get its content using wireless or wired ethernet. The pandaboard will also act as control center as it will render the (web-based) UI.

Initially (and target for this project) the medida and UI will be rendered on TV (connecting the PandaBoard using HDMI). In a later stage, a local touchscreen can be added allowing the UI to be run on the device itself.

With the James renderer the user can:

Output sources:

Control devices

Rationale for using a PandaBoard

Project/URL (if exists):

Expected results:

Background & work by projects submitter/s:

Contact email:

Panda Home

Title: Panda Home

Description: One device to rule your home...

The goal is to create a low cost open home automation using the Pandaboard and Android or iPhone devices as the master piece of the system.

Thanks to this we will:

Project/URL (if exists):

Not yet created but it will be done in the next few weeks.

Expected results: A low cost home automation system with:

Background & work by projects submitter/s:

Two students work on this project.

We are supported by our computer science professor, M. Sylvain Jubertie, who works on parrallel programming (SMP/multicore processors, clusters, parallel languages, algorithms, applications to Virtual Reality applications). For two years he has proposed several projects on smartphones to his students (ARM and Android/iPhone).

Adrien Carteron has been coding on Linux, Windows, and Apple platforms for five years. He has developed a network mobile application for a French company for just in time stock. He specialized in electronic cards development.

José Martins is a freelance 2D and 3D game developer (iPhone and Android OS). He is studying computer science at the University of Orleans. Last year, he completed a project on mobile device using the same native code on both Android Archos IT5 and Itouch 2nd generation.

Contact email:

Bluetooth Hands-free

Title: Bluetooth Hands-free

Description:

The Bluetooth Hands-free Profile (HFP) comprises voice-call control and full-duplex voice-call audio streaming over Bluetooth. The profile defines two roles, Audio Gateway (AG) and Hands-Free Unit (HF), and each of the peers communicating over this profile perform one of these two roles.

The AG is the peer which connects the voice-call, through a modem, VoIP network, etc and makes the voice-call audio available to the HF. It also notifies the HF on an incoming voice-call. The HF is the peer which commands the AG to start a new voice-call, answer an incoming call, hangup, etc. It's also the HF which has speakers and microphone to play / record the audio to / from the user.

This project aims to provide an out-of-the-box Bluetooth Hands-free solution based on the PandaBoard and open-source software. This will make easy for anyone deploing a solution based on the PandaBoard to add both the AG and/or the HF funcionality to her product, be it a multimedia car-kit, an internet tablet, a smartphone, a television, or whatever our user can imagine.

To accomplish this an embedded (meta-)distribution of Linux will be used (most likely OpenEmbedded), together with the BlueZ userspace libraries (for the bluetooth logic), oFono (for the AT-chat), PulseAudio (to handle the audio streams), and D-Bus (general IPC and fd-passing between BlueZ and oFono). A good amount of the logic necessary to make this work is already available upstream on these projects an even tested on the OMAP3 platform (N900). But there are some other parts of the code which will need some stabilization work and testing (BlueZ' HF Unit implementation, oFono HFP plugin, D-Bus fd-passing). Besides that, there is still some code lacking to enable an oFono modem to be controlled from BlueZ (which is, in turn, controlled by the remote HF peer).

As a final goal, this functionality will be integrated into one of the various free/open media-center softwares avaialble (enna, xbmc, canola2) making possible to use the PandaBoard as a HTPC with the ability to answer / make calls using the users mobile phone wirelessly or using a 3G modem attached to the board from the comfort of her Bluetooth headset. This will make Bluetooth HFP usage a lot easier and more user-friendly than it is right now.

Project/URL:

Status updates and information will be posted on my personal blog. Code will be published at some public git repository and submitted upstream when applicable. In the future a dedicated webpage for this project might be created.

Expected results:

By the end of the project an image ready to be used on the PandaBoard will be available together with booting / flashing instructions. Also, instructions to build all the required software on the chosen platform and its source code will be published, and the applicable packages / scripts / recipes will be generated.

A commented, fully-functional example code on how to integrate this on your application and some nice screenshots and videos will feature on the project webpage.

Background & work by project submiter:

Computer engineer since 2007, Linux user and programmer since 2002, now I work as a software developer for ProFUSION embedded systems, where I've gathered some experience with Freescale's iMX31/51 platforms. Contributor of several open-source projects, both on free and work time, most actively to BlueZ, oFono, and PulseAudio. I'm the author of the Bluetooth modules of PulseAudio and the Bluetooth A2DP Sink role on BlueZ, both as GSoC projects, and have helped on the HF Unit support for Linux. More info can be found on my Ohloh account, , and personal blog.

Contact email:

Plug'n'Play Mythtv

Title: Plug'n'Play Mythtv

Description:

Mythtv is a Open Source digital video recorder with a lot plugins. A Mythtv can have diffrent roles (Masterbackend (with DB and TV-Tuner Card/Stick),Slavebackend (with TV-Tuner Card/Stick), Frontend (with screen and remote control)). The Plug'n'Play Mythtv for the pandaboard will be a frontend with optional (master)backendrole. Mythtv is mainly used on x86 systems (there are also some beagleboard versions, but no Plug'n'Play one).

Project/URL (if exists): www.mythtv.org

Expected results:

Background & work by projects submitter/s:

Software developer since 2005, Linuxuser since 2004 on diffrent plattforms (x86, Maxtor shared storrage II (a NAS), Diffrent HTC Smartphones (Win Mobile ARM based)) . Mythtv user since 2007

Contact email:

NAS-Server (Ubuntu) with LAMP and video inspection using LAN-cameras

Description:

The idea is to work out a project that was formerly published by Andreas Krenn called NetHDD but had never been realized. Pandaboard should be the basis for a Linux/NAS Server including many more features as NAS alone, including LAMP. Additionally it should be able to interact with AXIS cameras (or others) in order to have video inspection including remote access.

Expected results:

A small box using external USB-drives for external storage and based OS on Ubuntulinuxserver probably resident on a SD/MMC Card (if bootable). Headless.

Background & work by projects submitter/s:

Years of experience in installing, configuring and using Linux-systems, especially Debian and Ubuntu. Working as system- and networkadministrator for many years.

Contact email:

FPGA based expansion board

Title: FPGA based expansion board

Description:

Goal of this project is to develop and build a expansion board for the Panda Board based on FPGA technology.
While the Beagle Board only allows for SPI style peripherals inside an FPGA extension, the Panda Board offers the possibility to extend the system bus into the configurable logic, giving the developer much more flexibility and freedom with respect to transfer speed, complexity and number peripherals as well as easier software access because of the direct memory mapping.
The expansion board will feature an FPGA connected to the 16 data/address lines exposed on the Panda Board expansion header.
The remaining pins of the FPGA will be routed to pin headers on the expansion board, to allow for maximum flexibility with the option of a third stacked board in mind, which will be specific to the targeted usage scenario (which is a car multimedia system in my case).

Expected results:

The result of this project will be an expansion board for the Panda Board featuring a system bus connection to the OMAP on one side and pin headers in the other. This way it will be a universal development platform for custom peripherals.
Given the short time span of two months, example designs will probably be developed after PEAP, but of course I will continue to work on this project to get a complete car multimedia system up and running.

Background & work by projects submitter/s:

I've got a Master's degree in Electrical Engineering (MSEE) since 2009.
Currently I'm working for a small startup company in the field of FPGA based prototyping platforms featuring a Linux operating system.
I'm interessted in both, hardware and software development, and especially their combination into complete systems.

Contact email:

Ptxdist for panda board

Title: Ptxdist with Qtcreator for the panda board

Description: ptxdist.org is a embedded platform builder (like OpenEmbedded or Buildroot). It can be used to build a full custom linux distribution for any hardware. Also ptxdist has a focus on "executable documentation". This means that e.g. some patches which are not upstream can be maintained within the ptxdist platform for the panda board. Since the panda board has enough processing power for graphical applications the QtCreator 2 with ptxdist makes development and deployment of new applications tailored for the panda board very easy.

Project/URL (Only the qtcreator part exists): qtcreator for ptxdist

Expected results: Expected results are a platform description for ptxdist which builds a root image that boots on the panda and has the qt development tools available. The qtcreator will be adapted to make use of the panda ptxdist environment for easy development and deployment on the panda board. Debian packages will be made available.

Background & work by projects submitter/s: I have successfully finished "Diplom Informatik" (comparable to Comp.Sci.Master) studies with focus on computer engineering. My first contact with linux where the 2.0pre series for which i wrote a small driver supporting my custom hand made ISA-SPDIF receiver. I am currently working with ptxdist for my employer. Some patches have been sent to upstream ptxdist. The qt-creator extension for ptxdist has also been published.

Contact email: [.]de

Mageia support

Title: Mageia support

Description:

Mageia is currently developing an ARM port of its Linux operating system, which can run on a wide range of platforms, desktops, notebooks, netbooks, servers.
The Mageia Linux distribution officially supports x86, and the ARMv5 port is basically ready, with support for Marvell Kirkwood platform and on-going support for Freescale i.MX51. We would like to bring TI OMAP4 as an officially supported SoC in the distribution, and extend it to platforms like tablets.
This effort will include merging TI OMAP4 hardware support in the Mageia distribution, and have graphical desktops available, to be used on smartbooks or tablets based on TI OMAP4.

Project/URL: http://mageia.org/en/

Expected results:

Background & work by projects submitter/s:

Our team has experience in managing ports to non-x86 architectures, like the currently supported ARM architectures (Marvell Kirkwood and Freescale i.MX51). The MIPS port for the Gdium netbook is another good example.

Contact email: ,

Classroom e-Learning System

Title: Classroom e-Learning System

Description:

An interactive system that provides a way for a student to interact in a multimedia environment with content provided by the teacher. This content could be on a CD or DVD, a file on a server in a local network or the Internet. It will consist of a multi-point interactive whiteboard using an infrared camera and a projection screen (or LCD screen) as an input device.
The project should strive for a low-cost solution for schools. The system should be able to interact with other devices, laptops, cellphones, etc. In order to provide a collective interactive environment in class. Also, the inclusion of a video camera will provide a way to bring telepresence from other groups or teachers that are located elsewhere.
There are many open source projects that deal with the problem of providing low-cost e-Learning software for schools and families (http://www.schoolforge.net/). The product should be easy to use as a stand-alone system, or be part of a more sophisticated system of networked computers.

Project/URL (if exists):NA

Expected results:

Wiimote software ported (http://johnnylee.net/projects/wii/) to the OMAP4, GCompris (http://gcompris.net/), TuxPaint (http://www.tuxpaint.org/), Celestia(http://www.shatters.net/celestia/), etc. Light pens and multitouch support.

Background & work by projects submitter/s:

10+ Years working experience with embedded Linux. Work with the Samsung S3C2442 on the Openmoko Freerunner, Samsung S3C2412, and the Beagleboard. Working experience with C/C++, Java, Android and x86 assembly. Software developer for the DARA debug board at Texas Instruments.

Contact email:

High Resolution Camera Board

Title: High Resolution Camera Board

Description:

A camera board using a 1/2.3 inch 10 or 14 Mega-pixel sensor to allow the pandaboard to capture high resolution images. The board will allow for the attachment of a lens mount and will also allow for additional expansion(ie. the i2c bus and power will be available via a expansion header).

Project/URL (if exists):NA

Expected results:

Produce and hopefully produce a camera board that will allow a high resolution image sensor(10-14MP) to be connected to the pandaboard.

Background & work by projects submitter/s:

I have previously designed camera boards using high resolution sensors(10MP and 14MP) for OMAP3 systems. I have 3 years hardware design experience,I have also written software and drivers for the ARM platform.

Contact email:

HK Workstation 1

Title: HK Workstation 1

Description:

A tool for the hardware developer and hacker enthusiast meant to replace the mess of wires and tools on his work bench with single unit capable of doing the following :
- provide a lab power source with variable voltage and amperage
- can measure characteristics of various electronic parts (act as a multimeter)
- provide a soldering iron mini station with variable tip temperature and sensors
- provide a touchscreen input device to view schematics and parts information
- provide a software development environment to go with the open source projects built
- have a small projector that can point to a wall for extra info space
- have a webcam and web access for social needs
- have inputs for extensions and other devices
- maintain a approximate small size so it can be mobile and transportable to hacker-spaces and conferences
The Pandaboard and the powerful capabilities that it offers will be the brain of this machine and will offer real time data on the project currently working on.

Project/URL : http://harkopen.com/projects/hk-workstation-1

Expected results:

Electronics and IT development went two separate ways but now with more and more open source hardware projects available and the power that the internet offers, a hardware developers work bench becomes more and more in need of a PC and added with his previous tools it creates a dangerous mix. HK Workstation 1 comes to provide a minimum of tools and features to build a project from scratch and maintain a small footprint and mobility. Think of it as Iron Man's lab on the go.

Background & work by projects submitter/s:

I am the founder of Harkopen.com , an open source hardware community, long time DIY electronics aficionado, passionate about open source hardware and with a few friends ready to get their hands dirty to build an awesome open source hardware project. The project will be shared with the community and hopefully attract contributors and testers along the way.

Contact email:

OE-lite PandaBoard port

Title: OE-lite PandaBoard port

Description:

OE-lite aims to improve certain use-cases which are not easily done using standard OpenEmbedded. One primary focus is on image generation in constrast to package feeds. When an image is created and deployed it is not possible to install packages. New features should only be deployed by building a new image from scratch. Extra care is taking to make sure that the same image build can _never_ result in two different results. Another primary focus is on making the smallest possible footprints. Images should be easily customizable onto the file level e.g. the user should be able to specify what "mount" binary should be used.
OE-lite aims at making it easier to develop your own custom applications to go with the image created. This is achieved by building an SDK that corresponds exactly to the created image. OE-lite also aims a creating a community around Git in a truly distributed manner. Only the most central components are kept in the "core" repository, while other parts will be kept in topic repositories e.g. an "X11" repository. New tools are developed to facilitate this kind of development, and to make the workflows easier.
The OE-lite project is supported by the company DoréDevelopment with both experience and equipment for hardware debugging.

Project/URL (if exists):

Currently no homepage is created, focus is still on rapid development of the main components. The project is hosted on Gitorious: http://gitorious.org/oe-lite

Expected results:

Make the PandaBoard one of the first well supported embedded boards of the OE-lite project. The PandaBoard will serve as a great way of attracting contributors with it's many nice features and use-cases.

Background & work by projects submitter/s:

The group behind the OE-lite project consist of developers working daily porting Linux to new embedded platforms, and long time users/contributors to the OpenEmbedded project and other Linux distribution efforts.


Contact email: ,

Program Trace Macrocell kernel driver port

Title: Program Trace Macrocell kernel driver port

Description:

Program Trace Macrocell (PTM) is a part of cortex-a9 core that does realtime tracing of whatever is being executed on the MPU. The trace can then be directed to the Embedded Trace Buffer (ETB) (which is a part of OMAP) for subsequent retrieval. This can be used for investigating problems in kernel (and sometimes userspace) code paths. PTM is a continuation of the Embedded Trace Macrocell (ETM), which was used in previous ARM cores and was present in previous versions of OMAP and is supported in the mainline kernel.
The trace stream, when retrieved, can be decoded by etm2human open source decoder (which also needs to be updated to support PTM's trace protocol format).

Project/URL: http://github.com/virtuoso/etm2human/wiki

Expected results:

Current ETM driver should be ported to fully support PTM; a guide should be written to make it easier for people to understand and use ETM/PTM in development; etm2human is updated to support PTM trace stream.

Background & work by projects submitter/s:

I have implemented the existing ETM driver for linux kernel and the etm2human trace stream decoder, which is the only open source ETMv3 decoder to the day; I am also a contributor to linux kernel and various other open source projects.

Contact email:

Distributed Mobile Object Tracking

Title:

A swarm of image based object detectors

Description:

The Pandaboard will be equipped with cameras and form intelligent nodes that build up a swarm of networked detectors for object tracking.
We have visions:
Focus: Object Tracking and statistical analysis Object Tracking and situation prediction
Goals: Increase throughput, better performance Forecast and avoidance of critical situations
Operating mode: Offline Analysis Real time Analysis
Vision: Rushing in some store, installation of the mobile detection system, gathering and analyzing real time data and finding hot spots, typical paths and customer behavior at the end of one single day. Distribute mobile equipment at event site, analyzing data in real time and prediction of critical situations in near future depending on models and gathered real time data. For instance early warning systems for human crushes, prediction of heavy traffic jam or avoiding waiting times at queues.

Key Features:

Additional Key Features:

Examples of Fields of Application:

Necessary hardware:

Necessary knowledge:


Project homepage:

Embedded in the bachelor programme (B.sc. Scientific Programming): http://www.rz.rwth-aachen.de/matse

Expected results:

Visions become reality.
First prototype system within one year.
Detailed project timeline and requirements specification within one month.
Different libraries will be implemented such as Network Meshing, Image Recognition / Object Tracking , Statistical Math and Machine Learning.

Background & work by projects submitter/s:

Two Project directors with strong personal experience in embedded programming and image analysis currently working as professors in Software Engineering at the RWTH Aachen University and FH Aachen University of Applied Sciences offer highly scalable software development teams drawing programmers from more than 300 students.

Contact email:

[.]de

Multitouch gesture controller

Title: Multitouch gesture controller

Description:

Our aim is to create a multitouch-controller based on open source and hardware. The controller could be used to enable tracking of gestures or touches of users standing in front of a shop windows. Behind the shop window (inside) we will place a beamer, a camera and some IR-diodes - and a pandaboard, evaluating the touch events and forwarding them to a standard PC, using the output from the pandaboard as "mouse input". This separation of the multitouch systems into a dedicated touch-controller and a standard PC allows us to use the same system out-of-the-box in several contexts.

Project homepage:

not yet

Expected results:

A working first version, installed in a real location.

Background & work by projects submitter/s:

We are two ICT engineers who used to work in a telco R&D department in Switzerland and have just created our startup some days ago. We have backgrounds in SW-&HW-dev, electrical engineering and innovation. Out main goal (besides having fun) is to work on low-cost hard-&software prototypes for special events.

Contact email:


PandaBox - Hybrid HD digital TV receiver/streamer for terrestrial, cable and satellite broadcasts

Title : Hybrid HD digital TV receiver/streamer for terrestrial, cable and satellite broadcasts ( DVB-T(2) / C / S(2) )

Description :

The objective is to implement a hybrid HD digital receiver that can receive, display and stream digital terrestrial, cable or satellite broadcasts.
This relies on the panda board itself plus a few (cheap) usb tuners. With these simple tuners we're able to add as many as we want, for different sources and/or record parallel streams.
There will be two distinct layers in the system architecture: first there's the backend, which will handle the data/video streams, recordings, schedules, requests from the user, and so on. The second layer will serve as the main User interface, display videos/info onscreen, and communicate with the backend via tcp/ip, in a simple protocol yet to be defined.
What's particular with this implementation is that everything will be seen in a browser window, i.e. the User interface will rely entirely in HTML5, CSS and Javascript, meaning the pandaboard will act as the backend and also, when connected to a tv, as the main user interface unit. Any device on the network that can handle a webpage, will be able to act as the 2nd layer and control the backend and (when possible) stream video. It will not matter if the remote device is a computer, a tablet or a phone (any brand, any OS, as long as it has a webbrowser).
With the advent of cheap internet tablets and network connected media players, it should be possible to create a plugin for the backend (pandaboard) to handle requests and serve data to other media players. Assuming there's enough bandwidth, it should be possible to access streams from anywhere on the internet.
This implementation will attempt to bring the best features of VDR, MythTV, Freevo, Dreambox and others, and break the boundaries between the receiving hardware and the "look" of the final user interface, as it will be html5 based.
By using a backend + web based front end, it's possible to split the "problem" in smaller blocks, and create an open framework, that can evolve… where a skilled designer can create an awesome new UI, or a hw coder can add support for a new receiver/tuner without the need to change the whole system. This also means that one system will be able to receive data anywhere in the world, as long as we add specific external tuners.


Project/URL :

http://linux-hacker.net/blog/

Expected results:

I plan to release instructions and software that implements a hybrid (digital satellite and digital terrestrial broadcasts) set-top-box receiver/streamer.
The User interface will be webbased (HTML5, CSS and Javascript), so all the video and OSD will be sent to a webbrowser.
The pandaboard will act as both backend and main user interface unit (when connected to a TV).
By using an open framework, I hope to create a base for the project to grow, as users can add/tweak it as they want.
Ideally this work will either evolve into (or be merged with) a media center implementation, providing a universal set top box/media center with the ability to stream to/from the internet.

Background & work by projects submitter/s :

I have a masters degree in electrical Engineering and Computers, and have been working in R&D, mostly building prototypes based on embedded systems with linux and designing webbased UI's.
I have done little bits of most of what I describe above in other projects, so now it's a matter of joining all the pieces of the puzzle, build the missing ones and make them all "fit".
As a hobby I like to "hack" electronic stuff/gadgets, as the RM4100 (http://www.settoplinux.org/index.php?title=RCA_RM4100) and zenega S100 (http://forum.zenega-user.de/) and turn them into useable machines, in this case, media-centers. Progress on these was a real challenge, as some of the hardware components required real hacking, either via hardware or software, in order to enable "simple" things like video out, audio out, and so on...
I have a working DVB-T streaming "box" using debian on the RM4100 plus a cheap ebay dvbT tunner, but right now, control is done via the command line, so it's not yet ready for the end user. Neither of these 2 boxes has enough processing power to h264 streams or even HD, so the Pandaboard seems like the ideal candidate to replace them.

Contact email : :

OMAP4430 Bootloader + OS + drivers and app frameworks

Title:

OMAP4430 Bootloader + OS + drivers and app frameworks

Description:

Given that many embedded OSes already exists this may not look very feasible at first, but the goal of the project is to design and implement an operating system for the PandaBoard, and do everything is necessary to create a big community in developing apps for it.
Note however that the OS is defined to be more controller-specific, and not board-specific. Portability of the OS to other controllers is a bonus, not a requirement in this project.
A few self-imposed requirements of the OS as defined at this time are

Project/URL (if exists): TBD if selected

Expected results:

For expected results see above.
Milestones TBD. Amount of work estimated is 1 man-years. Self imposed deadline is 2 years from the first 'hello world' program debugged.
Moreover, the desired achievement is forming a big community for developing apps for the platform. Target of apps are from internet-tv, car media players or game consoles to mobile devices or automotive clusters.
One side achievement would be to gather a team of OS developers to finish the OS in under 2 years, but only after the basics of the OS have been coded.

Background & work by projects submitter/s:

I am a passionate programmer for 16 years, when I was first beginning with a Z80 platform. From then, my main occupation was programming, even if the fields of programming were different.. I have coded many applications in many languages: from simple apps for Z80 to complex games for PC's, and from assembly and C to Java, HTML, Javascript, SQL and PHP and so on. I must say I liked them all.
It is 3 years now from when I first began to change my passion to profitability as I was in need of a place to work.
I now work as an embedded developer in a big multi-national automotive corporation, where my main work is building drivers and applications for embedded systems.
My best project relevant to the OMAP OS is this (I will not point out the confidential things, just the main idea):
At my workplace we are building a test system. Basically a bootloader loads an image(comprised of multiple modules compiled together) over CAN network and it waits for commands. The image contains payloads for various commands, from big LCD display drivers, to I2C communications or whatever it is on the hardware board. When a command is issued over CAN, the system lunches it. The image is tightly coupled with the system, in a relationship of a service provider -> service consumer, but the system is built in such a way that both the image and the system are executing everything in sequential order, and NO hardware interrupts are allowed. What I have done to overcome this is another module that enables the system to work in parallel. No change to the binary of the system or image is necessary. Everything is essentially the same except there is this new module which handles everything by stealing pointers and intercepting calls and instead makes context switches, allocates timers dynamically where there were static, handles hardware interaction and communication and so on, while maintaining the size of it to half of a medium sized normal module, all this in 120h of work. Of course the gain is quite big, as now the runtime of all the tests is half what it was.
All this made me think to build an OS, and from some time now I am documenting myself and it seems that while it is quite a difficult task, it is not impossible, and I have the advantage of being passionate about this topic..

Contact email:

OMAP Programming Book for Application Developers

Title: OMAP Programming Book for Application Developers

Description

'OMAP Programming for Application Developers' will be a book written by an application developer for programmers who want to get an insight into embedded development using OMAP3. As an application developer who spends a lot of time optimizing user interfaces and multimedia apis for mobiles, I find myself often wanting detailed explanations on how modern hardware works. I find existing books too abstract since they don't focus on any particular hardware. Hardware manuals make sense only if you some prior experience with the technology. I want to write a book that explains how the hardware works and then goes ahead to explain how to program the hardware using simple programs.

Some examples of questions that the book will answer:

Project/URL (if exists):

TBD

Expected results:

The book will plain HTML/PDF available online for anyone. It won't be a wiki. I expect the project to take 6 months for the initial version.
At the end of the book, the reader should be able to figure how to best write his applications for OMAP and the best tools/libraries for the job. Say, the best way to write a image manipulator, the best way to display camera output say without using XVideo.

Background & work by projects submitter/s:

I am an ex-Trolltech employee where I was a senior Qt developer and release manager. I have contributed to various parts of Qt and am the author of Qt Style sheets, Autocompletion framework and prototyper of Widgets inside QGraphicsView. I continue contributing to Qt through the WebKit project where I have Comitter status. I enjoy writting small programs that help me understand the system better (http://git.forwardbias.in/).
I have taken an interest to writing for the past year for my own good because it helps me understand concepts better. I reviewed the Style Sheet chapter of the C++ GUI Programming with Qt4 book. I am the author of the Qt internals articles at http://developer.qt.nokia.com/wiki/Category:QtInternals.

Contact email:

Ad-blocker for TV

Title: Ad-blocker for TV

Description:

Perform real-time video processing using the OMAP chipset. The goal is to create a device that interfaces between the TV signal coming from a set-top box and the TV, monitoring the video. When the commercials start playing, the device
  1. mutes the sound
  2. shows pre-determined user-specific content (slideshow/video of user's kids?)
  3. or re-shows content from just before the commercial started

The following user-controls will be provided:

The device will take these inputs using either a remote control connecting through IR or a mobile app connecting through wifi.

Project/URL (if exists): Not yet

Expected results:

Working device as described above. It involves the following:

Parameters need to be tweaked based on field tests (like how many blank frames, what percentage of pixels change).

References: 1 2 3

Additional hardware:

Background & work by projects submitter/s:

I worked for Texas Instruments India (ASIC CAD) 2001-07. I now work on Qt for desktop/mobile as part of a startup. I have contributed code to Qt Creator and .

Contact email:

Multimedia Support

Title: Multimedia Support

Description:

I just wanted to add Enna support as my project and realized that a lot of proposed project deal with multimedia (FFmpeg, XBMC, MythTV, VDR, …).
This project proposes to subsume all these projects into a big project to benefit from each other. For example, if I am not mistaken FFmpeg support is needed to play video content and the other projects could test FFmpeg support and provide feedback.
I do not know how successful the projects of the BeagleBoard early adopters program were, but the success rate probably increases if more people work together.
Of course all participants need a PandaBoard to be able test quickly. ;-)
Since non-big-cooperation developers do not have access to OMAP4 devices yet, this would assure that OMAP4 gets a lot of testing and is usable for normal users.
My proposal to choose a development framework in the beginning (for example OpenEmbedded) to be able to develop and test the software easily.
If you would be willing to join this group of developers to get the PandaBoard supported, please add a comment or add your address to the submitter’s section.

Project/URL (if exists): not yet

Expected results:

All major multimedia applications will be supported and run on the PandaBoard. Instructions will be provided on how to get these working. Also a pandaboard-demo-image, similar to the beagleboard-demo-image, will be provided being to show case the features of the PandaBoard.

Background & work by projects submitter/s:

A year ago I got a BeagleBoard and got involved with the OpenEmbedded framework. Already before that I have been using and I have been interested in VDR and Enna.

Contact email: [.]net

PandaBoard Demo Image

Title: PandaBoard Demo Image

Description:

(If my »Multimedia Support« proposal is not acceptable or applicable to the PEAP project rules, I am proposing something different.)
People love to see examples to be able to judge the capabilities of a device. This project wants to provide a demo image, similar to the BeagleBoard demo image, to show case the features of the PandaBoard.
One option for example is to use the PandaBoard as a device to control other devices using Bluetooth, to show interesting animations or to play a HD video (probably Big Buck Bunny).

Project/URL (if exists): not yet

Expected results:

The demo image will be build using OpenEmbedded so the first result will be to get the PandaBoard supported in OpenEmbedded.
Ȧngström – which is also used for the BeagleBoard – will be used as the distribution.

Background & work by projects submitter/s:

A year ago I got a BeagleBoard and got involved with the OpenEmbedded framework.

Contact email: [.]net

Universal TFT LCD Interface Board

Title: Universal TFT LCD Interface Board

Description:

This expansion board would enable an easy solution to add a LCD panel of almost any size and interface.
Adding a LCD to an embedded project is a challenge not only on the physical interface (different FFC styles and pinouts) but on the software side as well (different timings). Wouldn't it be nice to have an interface board that easily plugs into the Panda and then offers some of the more common FFC connectors used on LCD panels and then has a configurable method to easily map the correct pinouts from the Panda to the LCD. No more custom adapter boards for every different LCD panel your project/product may require.

Project/URL (if exists): None at this time.

Expected results:

This board should provide an easy way to not only mechanically connect a LCD panel to the Panda but also help with configuring the kernel to output to the panel.

Background & work by projects submitter/s:

Been in the electronics industry for 25 years designing and developing custom products. Recently been working with the BeagleBoard and "porting" different OS' to run on it.

Contact email:

In-Classroom Expo-Computer

Title: In-Classroom Expo-Computer

Description:

This proyect would help the expositions in the classroom. There's always problems when some person want to expose to the class some media files and he or she have to bring the computer, connect the VGA cable, search the media to show and wait to charge the program that display it.

PandaBoard is very useful in this matter because is small, fast, low cost and has all the HW needed to finish this proyect. We would use open source OS, like Meego, Fedora or Ubuntu to help with the core part and make our own file manager to browse through the external media drive, like usb-drives, external HD, or make a conection by Bluetooth to my computer and share the files to display. Project/URL (if exists): None

Expected results:

In the end of the proyect we expect to test this in our university, Universidad Tecnica Federico Santa Maria, Chile. And make easier to the students to expose and share media with the class.

Background & work by projects submitter/s:

We are a group of student that investigates and develop for ARM arquitecture. In our proyect, we have a media-center and a micro-server in beagleboard.

Contact email: [.]com

HandHeld Media-Server

Title: HandHeld Media-Server

Description:

There's always the need of a portable, low consuming, high procesing unit that help storing movies, photos, music and will connect with TV, Monitors, Computers and that would be portable, easy to plug in and browse the media, this would with a touch-screen that would select the media to show and the way to show it(HDMI, VGA, Screen). The PandaBoard would be the perfect board to do this because has a lot of processing power and support 1080p.

Project/URL (if exists): None

Expected results:

We expect to create an open source portable media server with a touch screen to browse trough the files and display them on a LCD, TV. With the Bluetooth we can take media from other laptops or celphones and be a by-pass to show them in the display.

Background & work by projects submitter/s:

We are a group of student that investigates and develop for ARM arquitecture. In our proyect, we have a media-center and a micro-server in beagleboard.

Contact email: [.]com

BlankOn Srimubil/ARM

Title:

BlankOn Srimubil/ARM

Description:

BlankOn Srimubil is an effort to make BlankOn for mobile devices.
BlankOn itself is an Indonesian Linux distribution which was originally derived from Ubuntu and now slowly maintains its own packages. It is a very popular distribution in the region and has a big userbase there.
The BlankOn Srimubil project aims to fill the gaps of operating system for mobile devices. Many generic computers come from China and they are not really suitable for running the desktop version of BlankOn because of screen resolutions. Most of them are using Atom based or ARM based CPUs. And we want those products to include BlankOn as the OS. During last BlankOn developers and users yearly conference, BlankOn Srimubil chose MeeGo UX as it's user interface as it was considered suitable for the use cases.
The common use cases that we are trying to solve are:
Wiki:

Expected results:

With this program we would like to solve the second use case (and maybe the third). What we would achieve:

Background & work by projects submitter/s:

Give details of your background and all relevant information that will help the judges determine you have the necessary skills, interest & background needed to execute your proposal
BlankOn started on 2004 and we have reached our sixth release this year, and expecting the seventh next year (yearly release). We have a handful of developers have experience in ARM environment but we believe for those who don't will get more new experience with this project. Two of the ARM developers are professionals in this area, and one of them which is the contact person below is a principal engineer in MeeGo Computers and have 5 years experience of getting the OMAP based Nokia devices out with Maemo.

Contact email: Mohammad Anwari

Amateur Radio Experiment Platform

Title: Amateur Radio digital modulation and Software Defined Radio experiment platform based on OMAP4/PandaBoard

Description:

The aim is to create an flexible platform for amateur radio experiments with both digital modulation schemes and software defined radio / zero-IF. This can leverage the DSP and EGL capabilities of the PandaBoard to filter, further process the data stream and finally present it to the user e.g. in the form of an 3D FFT of an received signal in real time.
The simplest setup can be achieved by adding a minimal RF frontend consisting of an fixed frequency mixer to transform a part of the spectrum (e.g. around 7 MHz) to 0-96kHz (or less depending on ADC) and sampling it with the ADC of a sound interface (e.g. TWL6040).
Though the plan is to attach an more capable ADC to directly sample a larger part of the RF spectrum (ideally at least 0-2MHz). Possible ADC's would be e.g. AD7266 and AD9649.
The sampled signal spectrum will then be processed which will include, filtering, downconversion and demodulation. Ideally large parts of this will be moved to the DSP core.
To give the user an visual feedback one or more FFT's of the spectrum (e.g. raw input, filtered, downconverted) will be rendered by an GUI implemented in Qt and using EGL for 3D visualisation. Also decoder states can be visualized (QAM, PSK, …).
Besides simple modulation schemes like PSK31, RTTY or SSTV highly advanced modulation schemes like e.g. DRM (Digitale Radio Mondiale) should be possible due to the available computational power on the PandaBoard. DRM uses COFDM combined with QAM up to QAM-64 and the audio payload is e.g. AAC-HE. An Linux implementation is available (DReaM by Alexander Kurpiers, DL8AAU and Volker Fischer, http://drm.sf.net ) and could be adapted to benefit from the DSP core. Also derived modulations schemes for amateur radio applications could be tested (e.g. signal reduced to ~2,1kHz bandwith) as the software is flexible and also allows transmission.
A further benefit is the low power consumption when compared to PCs or laptops. This is important as radio amateurs will occasionally operate from remote parts of the world with no or unreliable mains power. For this also an touch-screen LCD interface could be implemented as such hardware is readily available to us.

Expected results:

A solid, fully FOSS, base for further amateur radio usage and experiments.

Background & work by projects submitter/s:

The project team comprises of radio amateurs with advanced OMAP based embedded device experience - Antti Seppänen, OH3HMI; Thomas B. Rücker, DM8TBR and other members of the OH3TR amateur radio club of the Tampere University of Technology.

Contact email:

DIY Ambilight

Title: DIY Ambilight

Description:

Ambilight (Ambient Lighting Technology) is a technology present on a few Philips TV's that projects light from the back of the TV/monitor onto the wall.
There are a few RGB leds on the back which vary intensity and color based on the actual image beying displayed.
There's an incredibly simple way to achieve this with any TV/Monitor... ;)
All you need is an embedded system, like the pandaboard, a webcam and a few RGB leds and controllers (for example a few GPIO's on the board, or an external PIC or Atmel microcontroler)
After aligning the webcam to "see" the TV without obstacles, we simply need to define the area of the captured image that matches our tv screen, and start sampling images as fast as possible... for each image, we analyse the main colour and intensity of each side/corner, and adjust the backlight accordingly.
This means it would work with any source, either analog, digital, SD or HD... anything.. as long as you don't stand in front of the camera.. :p

Project/URL (if exists): http://www.linux-hacker.net/blog/

Expected results:

Write the software and a tutorial on how to achieve a simple and cheap "Ambylight" feature with any TV/monitor + a pandaboard, a webcam and a few leds.

Background & work by projects submitter/s:

I have a masters degree in electrical Engineering and Computers, and have been working in R&D, mostly building prototypes based on embedded systems with linux and designing webbased UI's. As a hobby I like to "hack" electronic stuff/gadgets. :)

Contact email:

x86 Firmware/BIOS debugging and recovery tool

Title: x86 Firmware/BIOS debugging and recovery tool

Description:

For hobbyists and professionals working on x86 firmware/BIOS/EFI/coreboot development as well as Windows driver developement and Linux kernel development, no affordable debugging and recovery tools exist. Especially on modern legacy-free x86 boards there is no way at all to trace code execution from the outside unless you're willing to sign an NDA and shell out $100,000 or more for some in-circuit emulators or JTAG eqivalents. In the past, serial ports were available on most mainboards and those serial ports could be used for debugging as well as a limited degree of remote control. Modern mainboards don't have serial ports anymore, and the often-suggested serial port replacement, an EHCI Debug device, is pretty hard to get (read: not manufactured anymore). Besides that, classic EHCI Debug devices are not really usable with USB 3.0 anymore.
Given that the Pandaboard supports a programmable USB 2.0 OTG interface, it can be turned into an EHCI Debug device (doesn't require EHCI on the Pandaboard, the name is misleading). For all recent x86 boards (even those without serial and with USB 3.0), it is possible to attach a LPC bus client (the Intel LPC bus exists on every x86 board manufactured in the last 8 years or so, and it is unlikely to go away) and such a LPC bus client can act as virtual high speed serial port without the need for extra drivers. Besides that, very recent boards have their firmware flash chips attached via the SPI bus, so it is possible to trace execution by following the reads issued over SPI. With a Pandaboard supporting EHCI Debug, virtual serial port and SPI tracing, debugging firmware/BIOS/EFI/coreboot and Linux/Windows kernels becomes a snap.

Project/URL (if exists): No URL yet.

Expected results:

The Pandaboard will get a driver which allows it to emulate an EHCI debug device. The Pandaboard will also get a LPC client driver which allows it to attach to a LPC bus and offer a virtual serial port. In addition, the Pandaboard will also get the ability to trace firmware execution in an x86 board with a SPI command decoder for common SPI flash read commands. A Pandaboard with such drivers will be a universal debugging tool and allow hobbyists to work in x86 lowlevel development/debugging again, and professionals will love having an all-in-one device for all their connection/debugging needs.

Background & work by projects submitter/s:

I have 4 years of x86 firmware development experience, a few more years of Linux kernel development experience, and I am the lead developer of the open source flashrom project which allows in-system BIOS updates, has a common interface for external flash programmers and DIY hacks for BIOS recovery, as well as the ability to flash the firmware/ROM of dozens of SATA/IDE controllers, network cards, graphics cards and other devices.

Contact email:

ZiPanda Home Automation

Title: ZiPanda Home Automation

Description:

A system to control electronic home devices (interconnected through Zigbee) where the 'coordinator' will be the PandaBoard.
The system will have the ability to be controlled with an app in a cell phone (using Bluetooth or WiFi) or the Internet (from an Applet)
The status of lights will be displayed on the TV

Project/URL (if exists):NA, one will be made once approved

Expected results:

Control lights on/off from a cellphone and having UI on a TV

Background & work by projects submitter/s:

I'm a recent EE graduate, currently working as Embedded Software Developer. During school I made a project where I could control light dimmers from a cellphone using bluetooth. The project was presented at a Congress (CONIELECOMP) "Controlling digital dimmer through mobile phone" sponsored by IEEE.

Contact email:

A mobile visual programming platform on ARM

Title:

A mobile visual programming platform on ARM

Description:

This open source project is trying to introduce the visual programming technology directly into mobile platforms(Meego/OMAP4 firstly),
which aims at provide a more user-friendly programming environment not only for the professional developers, but also for the
non-professional developers or the man even know little about how to write a program for mobile devices.
In a word,let us enjoy the fun of seeing a program running on the mobile device just written by ourselves via a much
easier and intuitiver way than the traditional mobile developement methods.

Project/URL (if exists):

TBD

Expected results:

1) implement a well designed Python-based portable visual programming platform for mobile platforms base on ARM processor
2) provide a on-device IDE like the Labview(www.ni.com/labview), which allow even the non-professional
developers can write mobile programs directly on mobile device by just draging and connecting in a visual environment
3) provide a corresponding host SDK and the interactive remote control methods to the target devices

Background & work by projects submitter/s:

I have nearly 7 years development experience on mobile platforms such like BREW and Android, the experienced projects
cover from App to the underlying framework. Recently, I am very interested in mobile Python related technologies, and
have developed some Python-based development tools for Android.

Contact email:

Autonomous Flying with Mikrokopter

Title:

Autonomous Flying with Mikrokopter

Description:

This Open Source project will combine the powers of a Mikrokopter with the powers of a mobile HD device.
The Pandaboard will do the processing of images. (using OpenCV and other vision tools) and send commands based on that info trough serial to the Mikrokopter.
We would like to use the HD capabilities and fast processing on the PandaBoard for this, as other boards are often to slow, or to heavy.

Project/URL (if exists):

TBD

Expected results:

1) Have working OpenCV libraries on the Pandaboard.
2) Combining Camera systems with the Pandaboard.
3) Doing fast image processing (if possible by using the dsp) on the board.
4) Seriously, a Mikrokopter driven by visual processing on the Pandaboard... videos of this will be all over the place, how awesome would that be.
5) Making publicity for the awesome OMAP boards/

Background & work by projects submitter/s:

I have done visual processing on the IGEPv2 (Beagleboard clone)
Our team has experience with DSP programming
Our company is specialised in visual processing in General. (+10 years of experience)

Contact email:

Connected Navigation

Project Connected Navigation

Title:Open Sorce Connected Navigation

Description:

We plan an open source navigation platform that uses the oss navit and
http://openstreetmap.org/ open streetmap data.
the navigation is enhanced by collecting life-data using camera and gps. GPS is used for tracking as usual with open streetmap.
Here we can use bluetooth-gps mounted to the roof of the car to have good satelite sight.
The new features are:
  • car integration:
    • support for ibus based BMW multifunction steering wheel in order to use the steering wheel buttons to manually control the navigation and to
tag things like speed traps and "bad guys".
see ibus (rs232 variant) here: e46-ibus
[4] IBUS-Analyser (to be modified to fit here)
IBUS-project
-read speed from ibus to navigate without satelite view in tunnel
ibus bmw e46: "80 05 bf 18 xx yy chk - xx = velocity/2, yy = rpm/100, message every 2 sec.
-other car interfaces like can using open source can2usb-dongle (if not there, we make one),
if there is a community who hacked the interessting data (velocity, distance, buttons for that car).
  • camera:
    • recognize traffic signs (speed limits and other regulations, parking signs)
    • recognize traffic density by counting cars in front.
    • warn if we drive to close (distance by triangulation of image size (we know the width of an automobile)) and speed by gps
    • try to mimic advanced cruise control by giving advice to speed-up or slow down to hold a distance (just a gimmick to see quality of distance measurement)
    • read numberplates and make still picture if ordered in order to oficially blacklist well known speed trap-cars.
    • warn if bad-guy (=speed trap, hidden police car) number plate has been seen.
    • use vision lib openCV (will be ported, where needed)
  • WLAN:
    • have car2car communication:
      • using wlan ad hock and some store and forward: good old internet news protocol.
      • we collect data (speed traps, traffic density and speeds at certain times, fined parking zones) etc for every road segment held by open street map.
      • we collect who is where (pseudonymously if user wants to be seen)
      • data will be synchronized with internet if a special wlan-access-point has been passed (a modified fonera for car syncing)
Project/URL (if exists):
www.openhardware.de (will be hosted here soon)
additional hosting here:
[5]
Slides
[6]
[7]
Expected results:
The result will be a standard platform using OMAP4 for the opensource connected navigation system in order to ease the design of the
open source connected navigation, which can be build laterly as a custom board for the masses.
  • The gap addressed is the need for
    • a vendor independend open source navigation system, that has a rich community and collects geodata at ease so everybody becomes a gps
    • scout and the street data will become better then commercial, because if the critical mass reached there are enough cars to report about traffic jams, construction sites dangerous spots and speed traps in near real time, automatically through 'magic' of crowd sourced data and the internet
    • a development platform for different car applications like mp3 player, documentation tool of car usage for book keeping and tax reasons.
    • a basis for telematic applications like pizza car job management, taxi routing and the like
    • other nice things like
    • car2car chat and "who is that girl I saw at the traffic lights in the green chevy van? :-)"
Background & work by projects submitter/s:
The maintainer luja has linux experience since 1994, and studied electrical engineering. He was employed as mainboard designer for telecommunication
products, he then worked many years as freelance engineer consulting in embedded systems, many of them embedded linux/arm.
Some friends will support this project e.g. hsank who is an asic engineer and has a strong linux and computer vision background
Contact email:

24/7 torrent-emule Server , NAS , 24/7 Voip , tv media player

Title: Integrated : p2p Server, NAS , voip , tv media player

Description:

This project aims to use the powerful potential of the low power consumption and powerful ARM9 PandaBoard
to me this kind of hardware open the possibility for the home user to have a computer powered and connected
to the internet 24h a day, everyday.
This mean you can for the first time have a voip solution (even with video) completely alternative to traditionalule
telephone .
you can also have a Low power consumption torrent and amule Server and NAS
at last i aim to add the sw to create a media player to connect to tv.
Whith this project i want to install OS : and add all the packages needed to the functions described
I aim to explore various solution in the OS and the packages and find the best solution . I also aim to create
a detailed description of the installation and compilation procedures to aim the PandaBoard Users to get
the same results as easily as possible.

Project/URL (if exists): i will develop during the project

Expected results:

to Get a working integrated NAS - voip - media player - p2p Server and
to precisly document all the step to get to this solution

Background & work by projects submitter/s:

I an electronic engeneer working in the embadded system programming (in the sector of electronic controlled battery charger)
i have also multiple years background in the system administration of my company net (it integrates linux server with win :pc's) . I am linux user and experimenter since fifteen years

Contact email:

Automotive RT-Linux (with GPMC extension to FPGA)

Title:

Automotive RT-Linux (with GPMC extension to FPGA)

Description:

I want to build a rapid-prototyping platform for automotive ECU-developing. Many algorithm are written in Matlab, hence they really need a decent system for real-time processing (OMAP4 is perfect for this task). Furthermore the camera-link plus the DSP will make many automotive-developer happy :-), video-processing is gaining more and more attention these days. The project will provide a basis for automotive testing/developing. In the first step, Linux with RT-patches will be ported to the pandaboard. The first step is closed after some timing-tests are full filled (preempting, maximum latency, ...). The first step of the project is not only related to automotive development (automation-systems, applied-science will have fun using the platform). The second step will make a Xilinx FPGA (Spartan6) available on the GPMC. Cars do not (yet) communicate over Ethernet and USB but LIN, CAN, FlexRay and Most150. Many controllers are implemented as soft-cores (FPGA), even if its a hard-core as a MFR4310 (FlexRay controller) the FPGA can pass-through to that controller. I will develop a simple Xilinx FPGA based extension board (in help with a friend), which is connected to the GPMC. Furthermore the project will provide a simple UIO-driver how to communicate with the FPGA.

Project/URL (if exists):

If accepted, the project will be hosted on Google Code.

Expected results:

A RT-patched Linux-Kernel with a user-space application, that communicates over GPMC with a FPGA.

Background & work by projects submitter/s:

I am currently finishing my diploma in electrical engineering at the university of Karlsruhe while working at a project at the BMW Group. During the years of study, I worked for the Robert Bosch GmbH, Siemens AG and a small company X2E (developing embedded-systems for the automotive market). Most embedded-systems, I worked with are based on the POWER-architecture, but ARM should not be a big step, thou.

Contact email:

Embedded Open Source Platform for Sail Boats

Title: EOSPSB ?

Description: Today every computer aids for sailing are really expensive, and totally closed. You can just buy electronic devices, and you can't even learn how they works.

My plan would be to start a whole range of devices for helping sailors. A good start point would be to create a charter, a low-powered device which can display nautical charts (like with openCPN for example) with connections to many other devices, like GPS, anemometer and the like... Also some kind of energy management is mandatory for that type of environment.

Afterwards, a platform like PandaBoard enable many things, like radio communications to get latest weather forecasts, auto-pilot, etc. Many commercially available functions have to be duplicated in the free software world.

Project/URL (if exists): not yet, but soon !

Expected results: At least a chart reader, with easy connections to multiple devices...

Background & work by projects submitter/s: I have a sail boat. Some experience with physical programming (Arduino). Deep-knowledge in web and linux stuff...

Contact email:

Distributed native GCC compilation based on Fedora project

Title:

Distributed native GCC compilation based on Fedora project

Description:

The project aims to create a system to build native ARM applications from x86 Host to multi-target OMAP boards.

Instead of using a cross toolchain to build the code, the system will build the code natively on the targets, through DISTCC, a working network (not necessarily local) and native GCC running on target boards.

The target will be based on Fedora project, the host will utilize a GCC compatible frontend for handling distributed compilation.

Project/URL (if exists):

Not yet.

Expected results:

We expect to have a package ready to be distributed on Fedora distribution for PandaBoard and an addon for the IDE.

Background & work by projects submitter/s:

We have several years experience on embedded developing. We have worked on core like ARM7, ARM9, ARM11 and PowerPC architecture, with OS like Linux and different types of RTOS.

Contact email:

Android Multi-User Media Sharing Table

Title:

Android Multi-User Media Sharing Table

Description:

The goal of this project is to allow users of Android phones to share media in an easy, fun, and eye pleasing way. Users will connect to a coffee table device with their android phones. This table has a screen in the center that has multi-touch abilities (calculated by the Pandaboard). A representation of each users phone is displayed and media from each phone can be viewed and sent to other phones using a natural user interface.

Expected results:

tBeta running on the PandaBoard [8]
Server that connects to android clients, and combines that data with the multi-touch info creating a great user experience.
Client software for android phones that passes media info to server

Background & work by projects submitter/s:

Working on CS and EE degree at George Mason University
AVR and PIC experience
Constructed a demo multi-touch table running on a Mac Mini
Expertise in programming android apps

Contact email:

Maintaining Root File Systems

Title:

PandaBoard Root File Systems

Description:

Any open board needs a root file system, kernel, initrd and a set of Uboot parameters to maintain. This project would provide board adopters with easy to use instructions on building their own kernel, initrd, and root filesystem, on their Linux OS of choice: Ubuntu, Fedora or Debian. It will also provide ready-to-go RFS images on these OS. These may be tarballs for NFS, or squashfs+aufs2 images.
In our experience with other open hardware projects like HawkBoard, the board developers are quick to deliver a kernel tarball. But everything else needs to be "borrowed" from somewhere else like Angstrom. This is fine, but often board adopters have no prior experience working with Angstrom etc. With Ubuntu, Debian and even Fedora providing fantastic support for ARM, there should be a way for users to use their favourite OS on their board. This way, they can just apt-get or yum in the packages they need, and focus on what they want to do - their projects. They shouldn't have to bother with building toolchains and learning a new package management tool.
We even propose to provide users with downloadable virtual machines where the entire kernel build toolchain is ready-to-use in their favourite OS - Ubuntu, Debian or Fedora. This way, they can get up and ready ASAP.
Finally, we find that for the first few months that a board is around, users must use the kernel source tarball or git put out by developers. Linux folks like to work with patches on a kernel.org kernel, not tarballs of some guys pull from the kernel. Our objective is to work with the board bring-up team, and as soon as possible, start putting out patches for folks. Again, these will be against both kernel.org kernels, and kernel sources from people's favourite OS: Ubuntu, Debian or Fedora.

Project/URL (if exists):

Expected results:

We expect to provide users with a way to have a clean Linux experience on their PandaBoard, within days of the board being out in the open. Folks should be able to get up and running with their favourite distro and focus on their projects, not on filesystems.

Background & work by projects submitter/s:

I've been building Linux root file systems for embedded targets for 10+ years, starting with NFS booting x86 diskless clients. I got Debian working on the VT8500 ARM SoC within days of receiving it. Getting Debian and Ubuntu working on Hawkboard took only hours. I also do kernel development, helping developers test their drivers and putting out patches. I've brought up numerous embedded targets in the past, especially on ARM.

Contact email:

Building a Chassis for the PandaBoard

Title:

PandaBoard Chassis

Description:

This project will design a sheet metal chassis for the PandaBoard. The skin will be designed to fit the board snugly, providing mounting holes for the board. All the CAD will be done using open source software such as QCAD and Blender. The design files will be open sourced.
I've personally blown up two HawkBoards while building ancillary circuits for them. Everyone likes to protect their target boards, but that's easier said than done. This project will provide folks with a nice skin for their babies. Another advantage of a chassis is that after folks are done with their R&D and are ready to leave the board running 24x7, they'll need to protect it from dust and water spills.
We'll even do a manufacturing run in India, and take pre-orders. The set up cost of the sheet metal machining system will get amortised over the number of orders we get, so the more the merrier.

Project/URL (if exists):

Expected results:

The expected outcomes are (a) complete design files, (b) sample production to test the design, (c) community runs of the design so folks can actually have the chassis, (d) assisting other hobbyists in modifying the chassis per their needs.

Background & work by projects submitter/s:

I've earlier designed a chassis for the Hawkboard using QCAD, and prototyped it by hand using a thin steel sheet. We didn't do a production run though. I regularly do mechanical design work in my day job (www.thinvent.in), and have a good understanding of how sheet metal processing and milling works.

Contact email:

Watching YouTube videos on TV

Title:

Watching YouTube videos on TV

Description:

The aim of this project is to enable a user to watch YouTube videos on TV set. PandaBoard will be connected with internet source like DSL router through WiFi interface on one end and with TV set through DVI interface on the other end. A USB mouse and keyboard will also be connected with board for user input.
An application running on PandaBoard will enable user to search / play videos from YouTube. Output of application will be displayed on TV set. User will be able to interact with application using attached USB mouse and keyboard. Application will present a similar interface like YouTube applications on iPOD/iPhone or high end android phones. Ideally an existing Linux / Android application can be used but general idea is the environment which enables viewing a live video stream on TV.
Initially this will only be targetted for YouTube videos, later it can be extended to general internet browsing as well.

Project/URL (if exists):

Expected results:

Expected outcome is a Linux based application running on PandaBoard enabling user to watch YouTube videso on TV set.

Background & work by projects submitter/s:

I have around 10 years of experience with embedded systems in general with a range of experience from application development, connectivity stacks like USB, PCI, SPI etc. and low level drivers. I am relatively new in Linux / Android domain but I have sufficient knowledge and experience to carry this project to completion.

Contact email:

QEMU on OMAP4

Title:

QEMU on OMAP4

Description:

This project aims providing multiple, concurrent guest OS environment using QEMU and Xen-ARM. QEMU is open-sourced hardware emulation and virtualization software. Xen is hypervisor(a.k.a. VMM, Virtual Machine Monitor) software to support hardware virtualization for x86, x86_64, IA64, ARM, and other CPU architectures. Current Xen for ARM has not been evolved since 2008.
Successful result will provide an environment to run multiple linux-based mobile OSes such as Android, Meego, Maemo and etc on OMAP4 development platform and/or OMAP4 based devices. This can allow personal developers to have flexible development environment with a single device. i.e. Running two Android environment and testing simultaneously with only one device. Running a test software which can cause serious environment damage without actual damage.
This project originally conceived to build a hardware virtualization environment for OMAP3430(or 34xx) and Nokia N900 but with Panda board, this can move on to next generation OMAP and be flexible for more developers.

Expected results:

Successful result will provide a full software distribution and documents of hardware virtualization environment for OMAP4.

Background & work by projects submitter/s:

I am a mobile OS developer working on SymbianOS and expecting to extend knowledge and experience to Linux based environment, too. I originally have a degree on embedded system development. I also have full-cycle development experience on mobile applications and games for J2ME and Symbian based mobiles.

Contact email:

Home Server

Title:

Mini-Server for home automation and multimedia

Description:

The target of this project is to use the Panda Board as base for a energy-efficient and quiet home automation server. It should make it possible to control things like relais (For light, multimedia system etc.), radiator and so on. This can be done manually by the user via a web interface (Optionally accessible via IPv6 from anywhere) or an infrared remote control (The IR-receiver is already existing, see Background) or automatically depending on values from various sensors (Temperature, time, light etc.). It's planned to use the OpenBus-Project as base for the interface between the Panda Board and the sensors and actors.
Since the OMAP 4 on the Panda Board got really good multimedia-features it's planned to use it as multimedia-server too (Would be a shame not to do so). One thing I would be specially interested in is to port the OGG/Vorbis and FLAC decoders to the IVA's DSP.

Expected results:

A fully working home automation system that controls the up to now manually operated things in my apartment automatically.
When this is working the next project would be porting the OGG/Vorbis and FLAC decoders to the DSP (Couldn't find much details about the used DSP on the TI-website though... maybe someone can give me a link to the documentation?).

Background & work by projects submitter/s:

I'm studying electrotechnics and informatics and will begin working on my final thesis soon. I've worked on some smaller electronic-projects the last 5 years, including an infrared-receiver that's registering as HID USB-device on the host and can be used like a normal USB-keyboard (As noted above I'm planning to integrate it into the described project too). I'm also using linux intensively for more than 5 years now and got a lot of programming experience in C/C++, Python and other languages. The knowledge I have should be sufficient to realize the described home automation project. At the moment I'm doing an internship where I'm writing software for a TI DSP, so I think it shouldn't be too hard for me to port the OGG/Vorbis and FLAC decoders to the DSP on the OMAP 4 (Would be great to have accelerated decoders for these formats... really missing that on my Nokia N900).

Contact email:

Communication Center for Hospitals

Title:

Communication Center designed for children who live long periods in Hospital Rooms

Description:

The main idea is to provide a communication center on the tv of each room of an hospital, specially those where there are children with difficult diseases, providing them an :entertainment and a way of connecting and improving their every day life. It would be a small box connected to the TV of each room, and it would offer several control methods.
The idea of this project would be using Meego as OS, as it offer a very easy UI and it will be a good OS for this kind of systems.
The communication center would offer to these special children, the possibility of connecting to the Internet, access to many multimedia contents, doing some e-learning activities, :courses and games, to connect with social networks and to have videoconferencing / VoIP telephony to keep in touch with their family, friends, and other children of the hospital in :their situation.
It would be personalized for the child, with a welcome dashboard, and a calendar/timetable with alarms for medicine schedule or any other important information.

Project/URL (if exists):

N/A

Expected results:

The expected result will be a small box, that would offer many possibilites of interaction that may help or almost entertain those children with deseases. The core functions of the :system would be:
- Music Centre
- Access to centralized content server with movies and series
- Access to internet contents (Videos, Images, Music)
- Music Streaming (with possibility of using bluetooth headphones)
- Dashboard Welcome Screen (iGoogle style, with info about weather,jokes, news, People connected or social networks updates)
- On Screen alerts with the time of necessary medicines.
- VoIP telephony/videoconferencing with bluetooth headset & webcam (family, friends, contacts)
- Very simple UI to access contacts, friends.
- Network with other children in similar situation.
- e-Learning:
Integration of e-learning educational activities as curiosities, games, experiments,
Courses about different subjects (Computers, individual skills, languages)
- Control possibilities: Wireless keyboard, remote control (wiimote), Bluetooth headset,

Background & work by projects submitter/s:

I'm a computer engineer and actually I'm studying a Master Degree in Embedded Systems and communication networks. This project would be my final work for the master degree (thesis), so I will work hard to make it real.
I'm an experienced linux user since many years and also got experience in C programming. I always enjoyed using linux and creating new systems and now I'm very interested in embedding :linux in this kind of boards. Some months ago I did a project of installing debian on an IGEP v2 board, with an ARM processor, and creating a low power frontend for mythtv. The system :connected to my mythtv backend, and you can control your mythtv and watch films, images, music, all the functionalities of mythtv but with a small/low power box near connected to the :tv. ()
Right now I'm installing Meego on this IGEP v2 board, and playing with it also in my netbook.
I think I have all necessary skills for this project, and if I may need some help, I would be happy to work in team.

Contact email:

A Modular Sound Synthesizer and Effects Processor with Many Ins and Outs aka The Magical Audio Octopus

Title:A Modular Sound Synthesizer and Effects Processor with Many Ins and Outs aka The Magical Audio Octopus

Description:

Once upon a time audio synthesis and effects processing was done by little boxes with circuits inside them. These were analog synthesizers. Modular synthesizers were the most powerful of all as they allowed you to connect many little boxes in many different ways to get many different kinds of sounds. All-In-One-Box synthesizers were easier to use and created most of the sounds most of the people wanted and so they were popular (Mini-Moog etc..) Modular synthesizers are difficult to recreate the same settings on time-after-time, however they made it easy to use other analog voltages/signals to control oscillators and make synthesized sounds that were driven by guitars, ribbon controllers, or anything that emitted a signal. Modular Synthesizers allowed you to connect the various Oscillators, Filters, LFO's, Envelopes etc. together in the order YOU wanted to connect them. Many hobbyists built their own modules and there were common standards for control voltage ranges. One commonly saw famous musicians in rock and fusion bands in the 1970's using massive walls of modular synthesizers and programming sounds while playing live.
Then things went digital. DSP power drives most of the modern hardware synthesizers of today. (Roland, Korg, Nord Modular and Nord Lead, Access Virus, Waldorf Microwave and Blofeld, etc.) Most digital hardware synthesizers are hand-coded masterpieces of assembly language DSP. This is beyond the programming capabilities of most audio synthesists.
Various high-level audio programming languages exist, such as C sound. Graphical Flow languages like MAX/Msp and the open-source PD (http://puredata.info) allow the programmer to create objects, messages, functions, (boxes) and connect them with signal or data level "cables". This resonates very nicely with the concept of "Modular Synthesizers" and indeed, it is possible to construct the modules of a synthesizer in PD and many have done so. PD is an excellent tool to construct many things and has been extended into the Visual realm with the GEM package. PD is used as the basis of several popular video-game sound-engines and is easily extended due to its open-source nature. The PD language elements themselves are extendable. It is also possible to code (C/C++) additional elements called "externals" and the PD development community is constantly doing this.
Today it is possible to run PD on an OMAP using operating systems like Android, Maemo, and various Linux-based systems. We can use the full-power of the DSP running DSP/Bios with the DSP-Link on the ARM side offloading our complex audio calculations to the DSP. The true impact of OMAP technology lies in harnessing the power of the DSP core in the context of a fully functioning development stack on a modern multitasking operating system running on the ARM core.
This project implements a Modular Synthesizer and Effects Processor in a box with many inputs and outputs for both control-voltages and audio signals. The device can implement MIDI and OSC over USB to communicate with your Favorite Sequencer or Digital Audio Workstation (Cubase, Logic, Ableton, Garage Band, Audacity, Ardour...) running on your home computer. The elements of The Magical Audio Octopus, or MAO, are sorts of "macros" developed in PD and easily extended by both the developer community and the end user. The Magical Audio Octopus has a simple user-interface of touchscreen, knobs, and buttons for the end user. The real-time programming will allow the connecting of modules using only this UI. Further programming will be possible with the use of a USB keyboard and mouse for deeper PD functionality. This MAO will feature 8 ins and outs for both Audio Signals and Control Voltages. There will be a preset system with many pre-made constructions that the beginner can find useful in making music or sound design. There will doubtless be those who only buy the box for the use of preset settings to play, create, and process musical sound, and there will be those who wish to further extend the synthesizer and effects modules, as well as folks interested in deeper development in terms of writing PD "externals" (additional PD elements compiled from C or C++ and useable in the creation of the modules of the synthesizer and effects unit)
Go ahead! plug in your guitar and play some low string notes to drive a deep synth baseline. Plug in your microphone and hear your singing voice transformed into a chorus of robotic t-pains chopped into granular clouds that swarm around every woo-woo woo you croon. Connect your Midi keyboard and play monstrous modular patches with lush effects. Use your Midi DrumPads to trigger the perfect analog drum emulator. The Sky is not even the limit.
This is intended to be an open hardware and software platform, although there is the intention of eventually producing the MAO (Magical Audio Octopus) and selling it pre-installed and running out of the box with presets etc…


Expected results:

I am interested to implement a Modular Synthesizer and Effects Processor in a box with many inputs and outputs for both control voltages, and audio signals, and controllable via MIDI or OSC over USB. It will feature a touchscreen interface for synth and effects programming allowing the simple drag and drop connection of modules and the alteration of their parameters via the touchscreen and the integrated knobs (rotary controllers) and buttons. Deeper programming of the modules themselves, the PD layer below, the operating system and device itself, etc. would require the connection of a standard USB mouse and keyboard. An external monitor connected to the pandaboard hdmi could of course be used to improve the ability to develop on the device.
-Panda or similarly equipped OMAP4 board
-Ideally 8 channel, but some amount of multichannel high-resolution audio I/O (24bit 96khz adc/dac >95db snr?)
-MidSized Touchscreen
-USB Mouse/Keyboard support for programming patches and/or the device
-Buttons and Knobs on the top deck!
-Full DSP optimization via DSP/BIOS DSP/LINK and any additional OS-based libraries available
-Open Source Linux-based OS optimized for running PD
-PD based Modular Synthesizer and Effects Processor
-Many modules including: Oscillators, FM, RingMod, Filters, Envelopes, Arpeggiators, Delays, Phasers, Analog Audio Inputs and Outputs, Mixers, etc...
-Touchscreen Optimized UI allowing full real-time construction using Synthesizer and Effects Modules
-Preset system with pre-installed and user-created presets for easily recalling Modular Synthesizer and Effects Processor settings.
Phases of Implementation
-choice of OS/Kernel stuff (RT options, ways to optimize for PD) as well as core libraries for DSP and UI of the MAO itself (PD ui used only for development and deep editing purposes)
-Implementation of basic MAO in PD using existing audio hardware I/O (2 channels) on touch-screen equipped Pandaboard
-selection of Multichannel High-Resolution Audio I/O solution and driver integration (I2C? McBSP? will i need to multiplex i.e. TDM?)
-Integration of Knobs, Buttons, and other low-bit serial encoders
-Design and integration of enclosure in conjunction with the UI and I/O
-Extension of Basic MAO modules and programming of user presets.
-Further Development...

Background & work by projects submitter/s:

Audio Engineer, Musical Composer/ Sound Designer, Electronic Music Producer (http://www.ocelotmusicshop.com http://siteholder.net/?p=945) working with synthesis, computers, and recording of music since 1993. Currently living from music alone since 6 years (Featuring the Perpetual World Tour- (nearly every weekend it seems!) over 50 countries so far for many thousands of people from Siberia to Brazil to Japan and beyond…)
BA Integrative Multimedia, 10+ years IT experience integrating Open Source solutions in the corporate environment. *nix system programming (shell-scripting, C-programming, compiling/making/building etc.) Hobby electronics projects- a midi monophonic analog synth, an 8 band vocoder, and a signal-driven octave doubler. Arduino midi and simple pcm projects. Hobby musical instrument maker of odd electric bass/guitar-like creatures.

PD user and fan. Highly self-interested with regards to HAVING a Magical Audio Octopus in working form to program and play with in my musical compositions and electronic music production.

Contact email:

MAO-Based Guitar Synthesizer aka The Embedded Guitar Synth

Title: MAO-Based Guitar Synthesizer aka The Embedded Guitar Synth

Description:

This project is based on my previous proposal of The Magical Audio Octopus.
Essentially, this is the Magical Audio Octopus embedded in a handcrafted wooden electric guitar.
1) hand-craft an electric guitar from wood
2) embed a Pandaboard and a 6-channel ADC codec
3) use the signal from each string driving a pickup (piezo or electromagnetic) to feed the ADC
4) expose the 6 channels of data streams in a way that PD can use them to control sound synthesis algorithms.
5) integrate the 6 channels into the modular synthesizer and effects processor architecture of the Magical Audio Octopus
6) create many easy to use presets that allow full playing of the guitar and further performance and patching via the touchscreen, knobs, and buttons.

Expected results:

I will create a musical instrument based on the humble guitar driving the massive modular synthesis and effects processing possibilities of the Magical Audio Octopus. Each string can control independent lines of signal processing for the creation of very complex and organic patches. The touchscreen can be utilized for creating modular synthesis and effects processing patches as in the original Magical Audio Octopus.

Background & work by projects submitter/s:

Audio Engineer, Musical Composer/ Sound Designer, Electronic Music Producer (http://www.ocelotmusicshop.com http://siteholder.net/?p=945) working with synthesis, computers, and recording of music since 1993. Currently living from music alone since 6 years (Featuring the Perpetual World Tour- (nearly every weekend it seems!) over 50 countries so far for many thousands of people from Siberia to Brazil to Japan and beyond…)
BA Integrative Multimedia, 10+ years IT experience integrating Open Source solutions in the corporate environment. Self-taught *nix system programming (shell-scripting, C-programming, compiling/making/building etc.)Hobby electronics buff- built various simple projects including a midi monophonic analog synth, an 8 band vocoder, and a signal-driven octave doubler. Hobby musical instrument maker of odd electric bass/guitar-like creatures. PD user. Highly self-interested with regards to HAVING a Magical Audio Octopus in working form to program and play with in my musical compositions and electronic music production.

Contact email:

Synthetic Programming on ARM

Title:

Synthetic Programming on ARM

Description:

This open source project is trying to introduce the synthetic programming technology to ARM via porting CorePy(www.corepy.org)
to ARM v7/v7A, together with a IPython-based host SDK to better support corresponding development.This project is aimed at
providing ARM developers with the ability to direct access the processor from high-level scripting-liked language--Python,
which has to be done by native ARM assembly language before, and make full use of the powerful productivity of synthetic
programming against multi-core/SMP/SIMD/HPC/ related computing.

Project/URL (if exists):

TBD

Expected results:

1) porting CorePy to ARM v7/v7A
2) provide an IPython-based host Software Development Kit for synthetic programming on ARM
3) research and practice the assembly-level programming on NEON SIMD instruction set via Python for

optimizing multimedia and crypto related programs on ARM processor

Background & work by projects submitter/s:

I have nearly 7 years development experience on mobile platforms such like BREW and Android, the experienced projects
cover from App to the underlying framework. Recently, I am very interested in mobile Python related technologies and
Python-based reverse engineering, and have developed some Python-based development tools for Android.

Contact email:

Open Source Android Tablet E-Reader

Title:

Open Source Android Tablet/E-Reader

Description:

We will build a completely open source Android tablet E-Reader hybrid using the new low-powered Pixel Qi Adaptive Screen and a 3M Micro Touch capacitive touch layer.
The tablet will support HD video playback and 3D gaming taking advantage of the Pandaboard's 1080p and 3D graphics support.
We intend to release all relevant source code and schematics under open licenses.


Project/URL (if exists):

TBD

Expected results:

A functional Open Source Android tablet E-Reader hybrid within the next 6 months with tablet rotation, camera, and additional preipheral device support within the next 12 months.

Background & work by projects submitter/s:

Our team is composed of two two Computer Science and Math undergrads at North Carolina State University, with a strong background in Linux and software development. Both are currently working on an open source, Arduino-based, mesh networked badge that allows for communication and gaming (similar to the Ninja Badge, found at http://ninjas.org/badges/defcon18.html).
Carson Holgate is an officer in the NCSU Linux User Group, a part-time developer working Linux applications, and an open source advocate. She also has previous hardware development experience. Her open-source projects can be found at http://github.com/clholgat
Michael Wright is a researcher with Systems Security at NCSU. His current research involves building a more secure javascript environment, but he's also working with the Android security team and intends to transition there full-time after his current project is complete. He has previous experience as a Linux application developer and system administrator, and has contributed to open source projects. His open-source contributions and projects can be found at http://github.com/mdwrigh2

Contact email: ,

Interactive Display and Instruction Unit

Title:

Interactive Display and Instruction Unit

Description:

The goal of this project is a proof of concept for an augmented reality application, providing an interactive projector display and instructional device for educational and training purposes. The ultimate goal is portable device incorporating a visual display, speaker, camera and DLP picoprojector (pico-DLP), which can provide instruction on the use of appliances, musical instruments, electronic devices, etc. The appliance of interest is selected and predefined operational instructions are downloaded from the internet. The camera will locate the appliance in front of the handheld device and the pico-DLP will highlight areas of interest on the appliance while instructions are provided on the display and/or speaker. For example, the projector can highlight buttons on the appliance in the order they need to be pressed while instructions are read. Similarly, the projector can highlight keys on a piano keyboard in the order to play them.
The PandaBoard offers access to the necessary peripherals through USB inputs for connecting to a camera, and HDMI and DVI outputs for connecting to a display and a TI pico-DLP. The OMAP4 processor offers computational capabilities for optimized image processing and augmented reality applications. WLAN access is critical for access to online instructional sets. The footprint of the Panda will allow this to be handheld and portable.

Project/URL (if exists):

Currently no homepage exists. The proposers’ research group website can be found at http://ecs.utdallas.edu/research/researchlabs/service-lab/index.html

Expected results:

The apparatus consisting of Panda Board, camera, display, and pico-DLP as well as the following algorithms implemented in OMAP4 will realize an augmented reality application. Proof of concept will consist of a live feed from a digital camera to a display monitor. A user can select a region of interest on the display, and the pico-DLP will highlight the selected area in the scene in front of the camera with a beam of light. Steps to reach the proof of concept include:
1. Methods to calibrate the camera and pico-DLP in order to ascertain their relative location and operational parameters
2. Matching projector beam location to the desired pixel location in the camera. There will be a line of possible matching points, across which the pico-DLP must sweep a beam of light. This line will depend on the calibration settings found in step 1.
3. Using the camera to detect when the beam has intersected the location of interest in the scene.
4. Storing the location of points of interest for future playback

Background & work by projects submitter/s:

Wen Yu worked as a co-op intern in the Texas Instrument, Dallas, for the team of OMAP4 Hardware Reference Platform. During that time, she helped validate the OMAP4 (Blaze) using the DARA debug board. She was included in discussion of Panda board design.
Nick Gans is a professor at the University of Texas at Dallas. He has over ten years of experience in field of machine vision. Biographical information and a research overview can be found at http://www.utdallas.edu/~ngans/
This project is supported by Roman Staszewski and Veera Raju from the Texas Instruments Natural User Interface (NUI) team.


Contact email: ,

Naarada Muni

Title: GNU radio on Pandaboard

Description:

PHASE 1: Interface the existing USRP2 infrastructure to the Panda baord via USB and port GNU Radio code to the Panda board

PHASE 2: Interface the USRP2 to the Pandabaord trying to remove the USB bandwidth bottleneck. The fundamental limitation the Bandwidth of the current GNU radio implementation with USRP2 is the USB link to the x86 based PC.The FPGA and AD9862 chips give us a 64 MS/s i.e. an effective bandwidth of 32 MHz The USB 2.0 connection has a theoretic data rate limited to 480 MBit/s. In practice it is worse. A benchmark program included in GNU Radio shows, that the USB 2.0 bus can sustain about 32 MByte/s of continuous data throughput, thus limiting the transfer to a maximum of 8 MS/s of complex signals (16-bit I and Q channel).["T. Schmid, O. Sekkat, M. B. Srivastava, "An Experimental Study of Network Performance Impact of Increased Latency in Software Defined Radios][9] The ideal aim is to just connect the FPGA directly to the pins of OMAP4 chip.

PHASE 3: The ultimate aim of this project is actually to make a board on which the FPGA is right on the same board as the OMAP4 chip but then this is, I guess, outside the purview of the Panda Board. Basically I want to create an all-contained Software Defined Radio with its own stripped down Linux kernel and UI.

This project will create a portable opensource Software Defined Computer, a handheld software defined radio device. And the Panda board is perfect as a beginning point for that. It has Video outs, audio outs, USB functionality and most importantly the OMAP4 IC that is required to begin with.

Project/URL (if exists): The current website for the project, is under development at

Expected results:

Initially the aim is to get the GNU radio working on the Panda board. After that we will optimize the code to work for the OMAP architecture. Finally the integration of the FPGA USRP2 to the Pandaboard directly to the OMAP4 chip.

Background & work by projects submitter/s:

Koustubh Sinkar:-
Background: Completed Bachelor of Engineering from Mumbai University majoring in Electronics and Tele-communication
Project: Completed his Final year project using OpenCV. The project was about segmentation of high resolution satellite imagery with watershed segmentation helped by edge detection markers after mean shift filtering of them image.


Contact email:

Koustubh Sinkar:

Solar Storm Radio Telescope

Title: Solar Storm Radio Telescope

Description:

The goal is to create Embedded Linux based Solar Storm Radio Telescope which will monitor solar storms indirectly by using radio waves to sense sudden disturbances in ionosphere (SID’s).Earth's ionosphere reacts to the intense X-ray and ultraviolet radiation released during a solar flare. The ionospheric disturbance enhances VLF radio propagation. By monitoring the signal strength of a distant VLF transmitter, sudden ionospheric disturbances (SIDs) are recorded and indicate a recent solar flare event.
This Embedded Solar Storm Radio Telescope for armature radio astronomers will replace PCs or laptops with sound-card which are dedicated for continues monitoring. This will have low power consumption, which will help in remote parts where power mains is not available.This will use audio and DSP capabilities of the PandaBoard. This includes porting of applications and libraries like FFTw to get FFT of an received signal in real time. Data can be logged on SD/MMc card, and use GNU Octave for numerical computations. WLAN/Bluetooth will provide quick and easy access to logs.

Project/URL (if exists): NA

Expected results:

Low cost sudden ionospheric disturbances monitor stations for armature radio astronomers and for science education in universities.

Background & work by projects submitter/s:

Have over 5+ Years Experience in Embedded Linux field.
Embedded Linux toolchain and distribution build experience for x86, ARM and MIPS architecture.
Hobby projects on robotics using embedded Linux on hawkboard.
Armature astronomer for over 10 years
Armature radio astronomy for 1 year

Contact email:

Autonomous MobiCopter

Title: Autonomous QuadCopter with Mobile Phone control

Description:

Developer a QuadroCopter vehicle (with stabilization system) capable of both autonomous flying and manual control through a mobile phone.
The vehicle will also send video stream to the mobile phone and will use a couple cameras to detect obstacles and avoid it.
The user can choose to control the QuadroCopter manually or set the "auto-pilot" mode. In "auto-pilot" mode the user will set the destination point and the beahavior when it reaches the destination (eg. land, wait, return, etc.).
The vehicle will be equipped with a gps to guide the navigation, a camera to video stream and a couple cameras to detect obstacles (thanks to PandaBoard proccessing power).
The connection between the quadrocopter and the mobile phone will be made trough wi-fi and in case of lost of signal the vehicle will go back to the last point where the signal was present.

Project/URL (if exists): None yet

Expected results:

Is expected to have a stable quadrocopter able to detect obstacles, fly automatically between two pre determined gps coordinates and take actions (land, go back and other user's defined actions).

Background & work by projects submitter/s:

As computer science student I have knowledgement in model driven app, computer vision, advanced algorithms, software engineering and some other areas.
As software developer I've worked with Symbian, iPhone, Android, Qt, digital tv, pervasive systems and plenty others frameworks and technologies
As Hobbyst I've worked a lot with microcontrolers, electronics, Robots and embedded systems. Also I'm a member of the Embedded Systems and Pervasive Computing Lab, hosted in Federal University of Campina Grande (Brazil)

Contact email:

RoboRescuer

Title: RoboRescuer autonomous rescue robot

Description:

Robocup Rescue competition (http://www.robocuprescue.org/), of Robocup fame (http://www.robocup.org/robocup-rescue/) is a well known and highly published contest employing Urban Search & Rescue (USAR) scenaria, where mobile robots compete in special arenas (see http://www.isd.mel.nist.gov/projects/USAR/), modeled from buildings in various stages of collapse, for objective performance evaluation of robots as they perform a variety of urban search & rescue tasks.

Expecting our RoboRescuer project to inspire and promote exploitation of the rich-connectivity OMAP4 platform at prototype robots participating in Robocup Rescue and related competitions, and also at commercial models serving people in critical situations throughout the world. It could serve as a starting point for other researchers to expand upon and potentially even as a reference design & implementation of a succesful rescue robot.

Project/URL (if exists): http://robotics.mech.upatras.gr/forum

Expected results:

RoboRescuer, an autonomous mobile robot for rescue in USAR scenaria (Urban Search & Rescue) will be designed, implemented and fine-tuned for competing in RoboRescue competition of Robocup. It will feature all typical capabilities of USAR robots, like mobility, sensory perception, planning, mapping, coupled with a practical operator interface (as stated in http://www.robocuprescue.org/rescuerobots.html).

RoboRescuer robot will make use of OMAP4 and the PandBoard, exploiting as much as possible of the OMAP4 platform potential, to be positioned as an able contester of the Robocup Rescue competition in the various available USAR arenas.

All mechanical and electronics design, together with implemented software will be released with an OpenSource license, together with documentation (including videos) of the build process and FAQ, guidelines, tutorial documents and a final case-study.

Disemination of our work via publications in conferences and research journals and spreading-the-word in social networks, traditional and web media will also be sought.

Background & work by projects submitter/s:

Team is comprised of PhD researchers at Robotics Research Group of Mechanical Engineering & Aeronautics Department at University of Patras (http://robotics.mech.upatras.gr) and Electrical Engineering & Computer Technology Dept. (http://www.ece.upatras.gr/en_index.php), in collaboration with university students from the two departments. Work & lab equipment images/video samples at http://robotics.mech.upatras.gr/portal/gallery.aspx

Proposal led and submitted by George Birbilis, PhD researcher (Multi-Agent Robotics), Msc Computer Engineer, Microsoft MVP J#-R&D 2004-2010, Borland Spirit of Delphi 2001 (http://www.zoomicon.com), with more than 20 years of experience in hardware and software engineering of complex systems.

Contact email:

Pandaboard as a Heterogenous Multiprocessing Platform (Or, Running Linux on all the ARMs)

Title: Pandaboard as a Heterogeneous Multiprocessing Platform (Or, Running Linux on all the ARMs)

Description:

There has been lots of recent research into using heterogeneous processors with the same ISA to improve power efficiency of multiprocessors (just google for "big core small core heterogeneous"). However, to date these have been limited to hypothetical platforms, or standard multicore CPUs with some cores artificially handicapped. The OMAP4 processor presents a great opportunity to implement this on real heterogenous hardware (Cortex-A9 and Cortex-M3) and examine its effects on power consumption and efficiency (and just do it because it's cool).
The Cortex-M3s are designed for running SYS/BIOS and interfacing with the imaging subsystems, however it appears from the TRM that there is no reason why it cannot run the same Linux instance as on the A9s and execute standard Linux user processes.
It would be really awesome to have non-CPU intensive workloads able to run on the M3, leaving the A9 in a deep sleep state. It may also pave the way to requiring one less OS/firmware blob running on an OMAP4.

Expected results:

The ability to schedule Linux tasks to run on either the A9 CPUs or the M3s.

Background & work by projects submitter/s:

We are a bunch of OS hackers from the Embedded Real Time Operating Systems group (ERTOS) at UNSW. We have students and researchers working on power management, virtualisation, reliable device drivers and more. Several of us have been hacking on the Linux kernel, including having virtualised UP and SMP Linux in numerous different ways.

Contact email: [.]com[.]au

Search and Rescue body and limb detection

Description:

Our focus is on providing digital video solutions to companies that are entering the arena.  We would like to port one of our existing search and rescue solutions to this platform, in order to take advantage of the advanced features. 
Our solution presently supports:
 - Video record
 - Video playback
 - Image capture
 - Infrared analysis
 - Basic image analysis for search and rescue.

We would like to demonstrate these features:

 - Human detection (head, body or limbs)
 - Position estimation.
 - Outlining of the detected body on the LCD with GPS coordinates estimated and displayed.

The idea is that the a person would carry around a unit that is using stereoscopic capture to alert the person that a) a body is present and b) the location of the body.  The location would use GPS as well as distance estimation in order to obtain a GPS estimation as to where the body is located.

We are also considering the implementation of "threat detection" for security uses.  Using the OMAP4, a "new" object that enters a busy area and doesn't leave can be detected and highlighted.  For example, a carried briefcase enters the field of vision and is placed on the ground and then remains there for many seconds. This would be detected within a crowded area, even when only glimpses of the item are seen for short periods.

Project/URL (if exists):

Company

    http://www.css-design.com     http://www.css-design.com/digital-media-solution.html

Sample Product

    http://www.con-space.com/searchcam/product/searchcam-3000-0

Expected results:

- Algorithms / algorithm additions to support the search and rescue operations.
 - BSP improvements/additions to support these features.
 - Demonstration to show these operations in a practical real-time application.

Background & work by projects submitter/s:

Our company has designed many products based on Davinci and OMAP and continue to lead development efforts on these platforms.  We are a recognized leader in this area, with regard to video development and have helped many customers succeed.
Our lead engineers have over 10 years of experience each, in the field of embedded systems, Linux, video, and DSP.  We have been optimizing ARM and DSP code for many years and have developed many advanced video and audio algorithms.

Contact email:

Retrieved from "https://omappedia.com/wiki/PEAP_Projects"
Personal tools
Namespaces
Variants
    Views
    • Read
    • View source
    • View history
    Actions
      Navigation
      Toolbox
      • What links here
      • Related changes
      • Special pages
      • Printable version