Learning Thermal-Aware Locomotion Policies for an Electrically-Actuated Quadruped Robot

TL;DR

This paper introduces a thermal-aware reinforcement learning control method for quadruped robots that prevents motor overheating, enabling over three times longer continuous operation without thermal shutdowns while maintaining effective locomotion.

Contribution

The paper presents a novel thermal-aware control approach that integrates motor temperature into reinforcement learning policies with thermal-constraint rewards, improving sustainable robot operation.

Findings

Baseline policy triggers overheat protection in ~7 minutes.

Proposed method operates over 27 minutes without thermal issues.

Maintains comparable command-tracking performance.

Abstract

Electrically-actuated quadrupedal robots possess high mobility on complex terrains, but their motors tend to accumulate heat under high-torque cyclic loads, potentially triggering overheat protection and limiting long-duration tasks. This work proposes a thermal-aware control method that incorporates motor temperatures into reinforcement learning locomotion policies and introduces thermal-constraint rewards to prevent temperature exceedance. Real-world experiments on the Unitree A1 demonstrate that, under a fixed 3 kg payload, the baseline policy triggers overheat protection and stops within approximately 7 minutes, whereas the proposed method can operate continuously for over 27 minutes without thermal interruptions while maintaining comparable command-tracking performance, thereby enhancing sustainable operational capability.