Meta Reinforcement Learning for Sim-to-real Domain Adaptation
This addresses the low sample efficiency and unsafe exploration issues in reinforcement learning for robotic policy training, though it appears incremental as it builds on existing sim-to-real methods.
The paper tackles the problem of sim-to-real domain transfer in robotics by using meta learning and a task-specific trajectory generation model to train adaptable policies, resulting in more consistent and stable adaptation and better overall performance on a KUKA robot task.
Modern reinforcement learning methods suffer from low sample efficiency and unsafe exploration, making it infeasible to train robotic policies entirely on real hardware. In this work, we propose to address the problem of sim-to-real domain transfer by using meta learning to train a policy that can adapt to a variety of dynamic conditions, and using a task-specific trajectory generation model to provide an action space that facilitates quick exploration. We evaluate the method by performing domain adaptation in simulation and analyzing the structure of the latent space during adaptation. We then deploy this policy on a KUKA LBR 4+ robot and evaluate its performance on a task of hitting a hockey puck to a target. Our method shows more consistent and stable domain adaptation than the baseline, resulting in better overall performance.