Whole-Body Constrained Learning for Legged Locomotion via Hierarchical Optimization

TL;DR

This paper introduces a hierarchical optimization-based reinforcement learning framework for legged robots that incorporates constraints to enhance safety and robustness in complex, real-world environments.

Contribution

It presents a novel constrained RL approach with hierarchical optimization that improves safety and transferability for legged locomotion tasks.

Findings

Successfully deployed on a hexapod robot in outdoor environments

Demonstrated improved safety with reduced joint collisions and slippage

Enhanced robustness in complex terrains like snow and stairs

Abstract

Reinforcement learning (RL) has demonstrated impressive performance in legged locomotion over various challenging environments. However, due to the sim-to-real gap and lack of explainability, unconstrained RL policies deployed in the real world still suffer from inevitable safety issues, such as joint collisions, excessive torque, or foot slippage in low-friction environments. These problems limit its usage in missions with strict safety requirements, such as planetary exploration, nuclear facility inspection, and deep-sea operations. In this paper, we design a hierarchical optimization-based whole-body follower, which integrates both hard and soft constraints into RL framework to make the robot move with better safety guarantees. Leveraging the advantages of model-based control, our approach allows for the definition of various types of hard and soft constraints during training or deployment, which allows for policy fine-tuning and mitigates the challenges of sim-to-real transfer. Meanwhile, it preserves the robustness of RL when dealing with locomotion in complex unstructured environments. The trained policy with introduced constraints was deployed in a hexapod robot and tested in various outdoor environments, including snow-covered slopes and stairs, demonstrating the great traversability and safety of our approach.