Camera Looks away from light???? Help?

[coldfusion1279]

We downloaded the easy c demo code from the kickoff, and downloaded it to the bot. When we shine the light, the camera actually looks away from the green light...

we have tilt in pwm 10, pan in 11, and the power in 12. We have tried switching them, but that just switched the axes of course.

I must mention we are inexperienced programmers and prefer easy c since the source code on kevin's website confuses all of the students.

also, is it already programmed in the kickoff code to have the robot drive to the light once it is tracked? if so does anybody know what function it is defined as?

Thanks in advance to anyone who can help!

[JamesByrne]

we had this problem for a little while, we found that floresent lights being to close to wehre we hold the light like 3 feet or so above the light screwed it up. And also the camera piking up reflections off other things.

Let me know if this does not work.

[coldfusion1279]

well we tried different distances..

from about a foot away, the camera just went crazy trying to stay away from the light

from about 6 feet away, the camera didnt even recognize the light at all, so we didnt even hold it any further..

any other ideas?

[Michael DiRamio]

I haven't used the EasyC code for the camera, but I know that with the tracking code in C you have to set the direction of the pan/tilt.

If a team was to mount the camera "upside down", then the tracking code would run away from the light instead of locking onto it.

I'm sure it's something like changing one parameter of the tracking to -1 instead of 1.

If that doesn't do it let me know and I can install EasyC and check it out.

[Terry Yiu]

Our team is having the exact same problem as you. This isn't an isolated incident either, there was another thread somewhere regarding this problem.

I'm thinking it has to do with a buggy code within the provided EasyC demo program. We want to modify the provided code, but EasyC won't allow us access into the source files. Or of course, it could be that the demo code works perfectly and we're doing something wrong... which is a possibility.

We have also tried mounting it on reverse with no luck, actually makes it worse.

[JamesByrne]

this is adresed in a diferent thred, i do not have a link. but if you are trying to use a robot batery to power the light. make shure the voltage is above 12.

[Swampdude]

We had this problem last year and found that we had the pan linkage attached wrong. Flip the pan link 180deg around that attaches to the servo. That fixed it for us.

[sniggel]

This is happening to us also The wierd thing it was working /tracking the green light, and now the code is running away from it. I thought we messed up the code and re downloaded the 2007 demo and it still runs away from the light. The camera says
found two targets, gives a readout then the camera moves above the target. We are doing this on a test bed, no motors attached. Am I missing something about how the camera will act on a test bed? Is it because the code thinks the robot has moved, and the camera than trys to adjust higher as it thinks it getss close? We have tried placing the light in mutiple positions but it still seems to run away from the light. Any suggestions on what might be going on here?