Hysteresis-Based RL: Robustifying Reinforcement Learning-based Control Policies via Hybrid Control

Code

• github.com/hybridsystemslab/obstacleavoidancehyrl
  OfficialIn paperpytorch★ 1
• github.com/hybridsystemslab/unitcirclehyrl
  OfficialIn paperpytorch★ 1

Abstract

Reinforcement learning (RL) is a promising approach for deriving control policies for complex systems. As we show in two control problems, the derived policies from using the Proximal Policy Optimization (PPO) and Deep Q-Network (DQN) algorithms may lack robustness guarantees. Motivated by these issues, we propose a new hybrid algorithm, which we call Hysteresis-Based RL (HyRL), augmenting an existing RL algorithm with hysteresis switching and two stages of learning. We illustrate its properties in two examples for which PPO and DQN fail.

Tasks

Reproductions