https://www.cc.gatech.edu/~turk/my_papers/stable_pd.pdf). This paper says as long as kd >= kp*h, stability is guaranteed. However, my observation with pybullet is that if I have a really large kp, I will still have oscillation problem even if the condition mentioned in the paper was met. Can anyone help me explain why stable PD control doesn’t work as what was described in the paper? At the moment, what I can think of is that maybe oscillation doesn’t count as instability?Statistics: Posted by r9dyt — Tue May 19, 2020 6:17 pm
]]>