Safe Reinforcement Learning with Dead-Ends Avoidance and Recovery

TL;DR

This paper introduces a novel safe reinforcement learning method that constructs a safety boundary to avoid dead-ends, enabling safer exploration and improved task performance in continuous control tasks.

Contribution

The paper proposes a boundary-based safety mechanism and a decoupled RL framework with offline safety critic pretraining for enhanced safe exploration.

Findings

Better task performance than state-of-the-art methods.

Fewer safety violations during training.

Effective dead-end state discrimination.

Abstract

Safety is one of the main challenges in applying reinforcement learning to realistic environmental tasks. To ensure safety during and after training process, existing methods tend to adopt overly conservative policy to avoid unsafe situations. However, overly conservative policy severely hinders the exploration, and makes the algorithms substantially less rewarding. In this paper, we propose a method to construct a boundary that discriminates safe and unsafe states. The boundary we construct is equivalent to distinguishing dead-end states, indicating the maximum extent to which safe exploration is guaranteed, and thus has minimum limitation on exploration. Similar to Recovery Reinforcement Learning, we utilize a decoupled RL framework to learn two policies, (1) a task policy that only considers improving the task performance, and (2) a recovery policy that maximizes safety. The recovery policy and a corresponding safety critic are pretrained on an offline dataset, in which the safety critic evaluates upper bound of safety in each state as awareness of environmental safety for the agent. During online training, a behavior correction mechanism is adopted, ensuring the agent to interact with the environment using safe actions only. Finally, experiments of continuous control tasks demonstrate that our approach has better task performance with less safety violations than state-of-the-art algorithms.