Imitation Learning for High Precision Peg-in-Hole Tasks

TL;DR

This paper demonstrates that generative adversarial imitation learning can enable industrial robots to perform high-precision peg-in-hole tasks efficiently, achieving sub-10 micrometer accuracy with minimal demonstrations.

Contribution

It introduces a GAIL-based imitation learning approach for precise peg-in-hole insertion on industrial robots, requiring few demonstrations and significantly reducing insertion time.

Findings

Achieved sub-10 micrometer insertion accuracy.

Reduced insertion time from over 20 seconds to under 15 seconds.

Learned effective policies within 20 episodes from fewer than 10 demonstrations.

Abstract

Industrial robot manipulators are not able to match the precision and speed with which humans are able to execute contact rich tasks even to this day. Therefore, as a means overcome this gap, we demonstrate generative methods for imitating a peg-in-hole insertion task in a 6-DOF robot manipulator. In particular, generative adversarial imitation learning (GAIL) is used to successfully achieve this task with a 10 um, and a 6 um peg-hole clearance on the Yaskawa GP8 industrial robot. Experimental results show that the policy successfully learns within 20 episodes from a handful of human expert demonstrations on the robot (i.e., < 10 tele-operated robot demonstrations). The insertion time improves from > 20 seconds (which also includes failed insertions) to < 15 seconds, thereby validating the effectiveness of this approach.