Re: Programming Issues with Crab Drive System
 

Our programmers have told me that both sides have independant PID control so they are already seperated. When the wheels are off the ground, the modules seem mostly in sync. When they are on the ground, thats when they sometimes get out of sync. Also, there is a strange situation when reversing. The two sides turn around 180 but go in opposite directions. Looks cool, but not a good thing. This does not happen while the wheels are off the ground.

I know that if they request the machine to turn more than 90 degrees, they make the wheels reverse instead assuming that it is quicker than trying to fully rotate the wheels with the wheels driving forward.

I was just asked another question by one of our mentors. Should the PWMs be in brake mode for steering or in coast. Our programmers think it should be set in brake mode so that's what its currently set for. Off the top of my head though, I would think that brake mode might interfere with the PID algorithm, but thats just my guess.

I suppose this should have been an off season project, but we knew it might be tough! It does go over the bump quite nicely though.

Alan

Re: Programming Issues with Crab Drive System
 

I do not recommend brake mode for steering or driving. The brake mode will add unexpected damping to the system when no throttle command is being sent to the controllers. This damping is variable dependent on the speed of the motor when the zero throttle is sent. Essentially it puts a short on the motor. Your motor temperature rise may be partially caused by this mode. Current in the motor creates heat regardless of where it comes from, the battery or motion.
It would seem that the condition you describe with the modules turning requires some really high level programming. It also will make the drivers confused. Once you get the crab functioning it still requires a lot of driver practice. Take baby steps, get the modules to function as a normal tank drive then start moving up to crab.

Re: Programming Issues with Crab Drive System
 

Al,

You were right, taking the pwms off of brake mode improved things quite a bit. The machine is much smoother now.

We also figured out why the modules rotated in two directions when reversing. The students attempted to use a single stick to control both direction and throttle. Which appears to work fine until you try to back up by pulling the stick in the direct opposite direction. Thinking they should probably split direction and throttle to two seperate sticks. Is that what everyone else does?

We remeasured everything with a fluke 337 DC clamp on true RMS meter with min and max peak memory. Window motors are drawing between 4 and 7 amps max now. As Raul suggested, we've decided to gear down the window motors and are going from 1.78 to 3.17 which should drop our current draw to a peak of about 4 amps or so. That means they will only spin at 25 RPM hopefully that will be fast enough. Of course, that means more weight.

Thanks for all the suggestions!

Alan

Re: Programming Issues with Crab Drive System
 

It sounds like you are beginning to see the light at the end of the tunnel. Remember the key from now on is practice and feedback from drivers to programmers. Start with driving forward and from side to side. This will give your students a lot of trouble at first but they can do it. Have them practice with obstacles on the field. You can then move to carts being pushed by other students that act like other robots.

Re: Programming Issues with Crab Drive System
 

We used a joystick that has a twist in it, so you would use the X and Y axis of the joystick to indicate speed in that direction relative to the robot, and then when you twist the joystick it would rotate the robot. It was reasonably intuitive.