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

2026-03-02

AI Summary

This work addresses the challenge of motor overheating in quadrupedal robots under high-torque cyclic loads, which often triggers thermal protection mechanisms and limits sustained operation. To mitigate this issue, the study introduces, for the first time, an explicit thermal-aware mechanism within a reinforcement learning–based locomotion control policy. By integrating a motor thermal model into the reward function as a thermal constraint, the approach simultaneously preserves locomotion performance and prevents excessive temperature rise. Evaluated on the Unitree A1 platform, the proposed method enables continuous operation for over 27 minutes—significantly outperforming the baseline approach, which lasts only about 7 minutes—while maintaining accurate command tracking throughout the extended duration.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Reinforcement Learning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

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.

Problem

• thermal-aware
• quadruped robot
• motor overheating
• locomotion policy
• overheat protection

Innovation

• thermal-aware control
• reinforcement learning
• quadruped robot
• motor thermal management
• locomotion policy