Safe Deep Policy Adaptation

arXiv:2310.08602v3 Announce Type: replace-cross
Abstract: A critical goal of autonomy and artificial intelligence is enabling autonomous robots to rapidly adapt in dynamic and uncertain environments. Classic adaptive control and safe control provide stability and safety guarantees but are limited to specific system classes. In contrast, policy adaptation based on reinforcement learning (RL) offers versatility and generalizability but presents safety and robustness challenges. We propose SafeDPA, a novel RL and control framework that simultaneously tackles the problems of policy adaptation and …

abstract adapt artificial artificial intelligence arxiv autonomous autonomous robots autonomy contrast control cs.ai cs.lg cs.ro dynamic enabling environments intelligence policy reinforcement reinforcement learning robots safe safety stability type uncertain