Residual Learning from Demonstrations

Contacts and friction are inherent to nearly all robotic manipulation tasks. Through the motor skill of insertion, we study how robots can learn to cope when these attributes play a salient role. In this work we study ways for adapting dynamic movement primitives (DMP) to improve their performance in the context of contact rich insertion. We propose a framework we refer to as residual learning from demonstration (rLfD) that combines dynamic movement primitives (DMP) that rely on behavioural cloning with a reinforcement learning (RL) based residual correction policy. Our evaluation suggests that applying residual learning directly in task space and operating on the full pose of the robot can significantly improve the overall performance of DMPs. We show that rLfD outperforms alternatives and improves the generalisation abilities of DMPs. We evaluate this approach by training an agent to successfully perform both simulated and real world insertions of pegs, gears and plugs into respective sockets.