I can see clearly now…
Feb 21 05 | 6:58 am
The camera system works.
Just meditate over that for a while. It’s so sweet. All the camera bits came together and we can reliably ID where the vision tetra is (read: in one of eight random places). It was so sweet. First we stomped some bugs in Jeff’s routine. Then we fixed the servo zeroing and all. Then we fixed a problem with what the horizon on the camera was. Then we fixed a problem where Jeff’s routine was adding x to y and y to x. Then we fixed a problem where the angle was correct only had the wrong sign (read: positive instead of negative). Key problem with camera = getting it to see the green properly.
However, autonomous isn’t together yet. For some reason the robot is jittering around instead of driving. First a busted encoder wheel (read: counts revolutions to measure distance) was suspect, but then we repaired it and still no success. Kim found out that the omni wheels (read:) we have have rubber pieces that deform under the robot’s weight, making them not slide well; this accounts for perhaps both the bouncing robot and autonomous problems.
Just hope we can work out the kinks before ship.