paper: TrackerBox: Object Tracking and Autonomous Robot Steering using a Raspberry Pi and a webcam

Thread created automatically to discuss a document in CD-Media.

TrackerBox: Object Tracking and Autonomous Robot Steering using a Raspberry Pi and a webcam
by: Mike Ounsworth

Want to track and follow an object but don’t have strong programmers? We provide a Raspberry Pi image that lets you train a target,and then it sends steering commands to the cRIO telling you where in the image the target is.

Our goal is to provide easy access to visual object tracking to teams who do not have sufficient programming knowledge to implement computer vision. We created a software and hardware system that allows teams to train a vision target, say an orange basketball on the ground or a green square on the wall, and have the robot autonomously steer to keep the target in the centre of the image. The system we
built should be simple enough that even teams with little programming knowledge can assemble, download, and use it, and cheap and light enough that any team can use one or multiple of them on their robot.

To download our custom Raspberry Pi image
http://isohunt.to/torrent_details/10051646/TrackerBoxRPi-img-zip

paper_RSS_format.pdf (2.89 MB)

I’m dying to see how this works. From personal experience, the best way to get stared seeding is to copy the file to a few other machines on other networks and use them as additional seeds. Once a few real people acquire the file it should be self-sustaining. In the meantime, could you describe a little about how the actual code works? Thanks!

Mike, very nice work on the paper! I’m a mentor and just purchased a Raspberry Pi. I would be very interested in trying out your software to see how well I can get it to work.

Unfortunately the torrent doesn’t seem to work but if you could send me the files via email that would be great, PM me please, thanks!

Ryan

If you received it could you pm me the files as well?