Where is the multi object tracking code for the RC?

[EHaskins]

Dave said the multi object code was available online, but I can't find it. Does anyone know where it is?

Thanks,
Eric Haskins

[Ctrl Alt Delete]

Does kevin's code work for this?
http://kevin.org/frc/frc_camera_2.zip

[Jeff Rodriguez]

http://www.intelitekdownloads.com/easyCPRO/

[DanDon]

Originally Posted by Ogre

http://www.intelitekdownloads.com/easyCPRO/

Do we use last year's unlock code for this?

[Alexa Stott]

Originally Posted by Ogre

http://www.intelitekdownloads.com/easyCPRO/

So wait, are you only able to use that code with EasyC?

[Kingofl337]

Currenty only easyC and WPILib have code written for Multi-Target.
Brad Miller will be making a project on Monday for WPILIB.
The reason being the GDC used easyC for FRC to program
the Demo and most of the test robots.

easyC PRO has a full IDE builtin now so you can write C code
as much as your heart desires. Also, easyC PRO has a graphics
display window that works similar to the way the demo did.

Team contacts will recieve a CD-KEY Monday.
You can run in evaluation mode till then.

[Kevin Watson]

Originally Posted by Kingofl337

...Only easyC and WPILib have code written for Multi-Target....

Really?

-Kevin

[chris31]

Originally Posted by Kevin Watson

Really?

-Kevin

I canr wait.

Originally Posted by JBotAlan

I am confused; I didn't even know that the *hardware* supported multiple objects. I'll have to have a look-see at that code to figure out how multi-tracking works.

JBot

If course it could. Its all about the software implementation.

[JBotAlan]

Originally Posted by chris31

If course it could. Its all about the software implementation.

Yes, but if you have 2 targets in one field-of-view, you will get a wide box and a low confidence value. That's at the camera's firmware level. So just by looking at the tracking packets, you wouldn't be able to see 2 targets. And I know we couldn't dump the contents of the frame into the RC; we don't have that kind of bandwidth.

However, if you only had one target in the field-of-view, and panned until you saw another, this would work fine. I'm guessing that's how the software works.

JBot

[maniac_2040]

Originally Posted by Kevin Watson

Really?

-Kevin

So...IS there going to be multi-object tracking code available?(Not the EasyC) Or do we have to modify the existing camera code ourselves?

[EHaskins]

Originally Posted by maniac_2040

So...IS there going to be multi-object tracking code available?(Not the EasyC) Or do we have to modify the existing camera code ourselves?

If you look at the EasyC code it is very simple. All the information from the camera is available through Kevin's code. I think it took me about 2-3 to get it working.

[Kevin Watson]

Originally Posted by maniac_2040

...Or do we have to modify the existing camera code ourselves?

I've been making suggestions on how this could be done in the hope that someone would modify my code (tracking.c) and discuss their modifications here. Given the lack of actual discussion taking place, I'm not sure if teams just went into their "Skunk Works" mode or if teams are just throwing-up their collective hands and not working on it.

Anyway, I have an idea for an algorithm that I'll have time to test out over the next few evenings. If it works well, I'll post it on my website. As all good engineers keep a backup plan in their pocket, you might consider implementing one of the other algorithms discussed in the forums, or invent your own. Either way, your time won't be wasted thinking about the problem.

-Kevin

[Donut]

Originally Posted by Kevin Watson

I've been making suggestions on how this could be done in the hope that someone would modify my code (tracking.c) and discuss their modifications here. Given the lack of actual discussion taking place, I'm not sure if teams just went into their "Skunk Works" mode or if teams are just throwing-up their collective hands and not working on it.

Anyway, I have an idea for an algorithm that I'll have time to test out over the next few evenings. If it works well, I'll post it on my website. As all good engineers keep a backup plan in their pocket, you might consider implementing one of the other algorithms discussed in the forums, or invent your own. Either way, your time won't be wasted thinking about the problem.

-Kevin

I'm not sure how many of the teams have gotten to the camera yet. Our team does mainly design and field elements the first week, so we haven't touched the camera yet.

Sounds like we're all thinking along the same track.

Lets say the rack is turned so an unlighted sided is directly in front of the robot. The two lights will be on the opposite sides of the tracking box and the centroid about in the middle. So keep the centroid in the middle and march to the unlighted goal. Too Simple??? perhaps.

BC

[Donut]

Another method for approaching the spider leg head on is to know your position on the field (meaning you have to get a gyro and accelerometer/encoder combo working in addition to the camera). If you know your position and the angle of the light you're tracking relative to you, you can determine what angle the light is facing and thus what angle you need to go in at.