Learning Deep Energy Shaping Policies for Stability-Guaranteed Manipulation
This addresses stability issues in DRL for robotic manipulation, providing a model-free framework for contact-rich tasks, though it is incremental by integrating control theory principles into DRL.
The paper tackled the problem of ensuring control stability in deep reinforcement learning for robotic manipulation by deriving an interpretable deep policy structure based on energy shaping and passivity, achieving the first stability-guaranteed DRL on a real robotic manipulator in a peg-in-hole task.
Deep reinforcement learning (DRL) has been successfully used to solve various robotic manipulation tasks. However, most of the existing works do not address the issue of control stability. This is in sharp contrast to the control theory community where the well-established norm is to prove stability whenever a control law is synthesized. What makes traditional stability analysis difficult for DRL are the uninterpretable nature of the neural network policies and unknown system dynamics. In this work, stability is obtained by deriving an interpretable deep policy structure based on the $\textit{energy shaping}$ control of Lagrangian systems. Then, stability during physical interaction with an unknown environment is established based on $\textit{passivity}$. The result is a stability guaranteeing DRL in a model-free framework that is general enough for contact-rich manipulation tasks. With an experiment on a peg-in-hole task, we demonstrate, to the best of our knowledge, the first DRL with stability guarantee on a real robotic manipulator.