Algodoo

It fail sometimes doeing frequant direction switch

I'm glad you brought that up. I'm not sure how to fix it. I've noticed that if you bring the wagon front wheel to the base of the inclined plane and start pulling, no amount of amplification will work. It only works if you give the wagon a running start. I think that what might be happening is that the inclined plane causes a step change in the wagon velocity which requires infinite acceleration (or change in velocity times sim.frequency in Algodoo). I'm interested in anybody's thoughts on what's going on or how to fix it.

I noticed that the drive wheel sits in one spot and jitters even when no torque is applied. I think there's a problem in your control algorithm, although I don't know what it is. If I were doing a similar scene, I would use a simple proportional control scheme (the "P" in "PID"). I might add a bit of derivative ("D") if it needed to be stabilized during accel/decal. And in place of the load cell, I would use a spring to monitor the pulling force from the hand. Just some ideas for ya!

Thanks for the ideas. I changed the hingeConstant from "NaN" t0 1000 on the load sensor hinge and that seemed to help (especially at the transition from flat to ramp). I also did some smoothing of the motor torque.

Oh yeah, the old NAN (Not A Number) issue! If Emil were still updating Algodoo I would suggest to him to preload those variables with sensible values instead of garbage (NAN). But unfortunately, it doesn't look like Algodoo will ever get updated again.

Well, since you said that, I went back to NaN because a hingeConstant of 1000000 was too springy.