hey Pololu, I just joined in the hopes of learning more about computer vision robots. I want to know how autonomous motion tracking robots like zero-op.com/mercenary and paintballsentry.com/ do it? I’ve read what they say on their sites, but how can someone else get at least a similar result with open source products?
After hours of research, I have narrowed my choices to these:
We do not have any experience with any of the things you linked to, but it is possible that some people from the community might have comments. Could you introduce yourself a little more? For example, do you have experience with programming? What languages and systems are you familiar with?
By the way, if you are serious about this, you are going to have to do a lot more than “hours of research” to get started with computer vision and robotics! For example, it might take you a couple of hours just to verify that a particular model of camera is compatible with a particular embedded computer, which will get you maybe 1% of the way there. Also, making it “fast” is probably way too high of a bar for a first try! I recommend breaking up the project into individual pieces that you can work on separately, like acquiring images, image processing, and motion control; each one of those parts could easily take many weeks of work. Some of the sites you linked to have guides that should help you with the individual steps, but you will still need to do a lot of research and experimentation on your own.
If I were doing this, I would try to learn how to acquire images before anything else, because, with my particular experience, that would be the most challenging part of the project, and I expect that many constraints (processing power, frame rate, etc.) will result from decisions made at this step.
although I do find it strange that you would say that you don’t have any experience with this. Since Zero Op sentry uses your servo controllers…which is why I posted here to see if you had information on how it used your technology to make his project work.
