Closed yijiangh closed 2 years ago
@yck011522 I'd recommend turning the reorientation movements into linear movements, and we can see if it works out-of-the-box or not. Opinion?
Worth giving it a try. @yijiangh Can do you a quick test form your side?
My quick test shows that it's definitely worth a try: https://user-images.githubusercontent.com/11905844/135121353-9334e49e-8c08-49df-a153-3c8a031ab910.mp4
For 180-deg, it's possible but the z coordinate of the orient pose needs to be higher:
This is A59_M1
of hypar:
This is A135_M1
of hypar:
After smoothing, the reorientation movement still looks like this:
I think here it's just that the end robot IK is sampled, and it's too far away from the start conf. We can either try turning this movement into a linear movement or be smarter about the sampling order.