Feeling Through the Load: Compliant Quadruped Locomotion under Payload Interactions

2026-10-06

AI Summary

This study addresses the challenge of quadrupedal robots carrying unsecured payloads, where distinguishing external interactions from self-motion remains difficult and balance instability poses significant risks. To overcome these limitations, this work proposes a force-aware locomotion framework that decouples force estimation from compliant control policy learning. A novel closed-loop data aggregation method for the estimator is introduced, enabling causal force inference using exclusively onboard sensors to transform payload interactions into guidance commands. By integrating reinforcement learning with closed-loop fine-tuning, the proposed approach is validated through both simulation and real-world experiments. Results demonstrate that the system maintains stable payload transport while remaining compliant to external interactions, supporting real-time trajectory adjustments via human guidance and achieving robust physical human-robot collaboration.

Abstract

Quadruped robots are increasingly expected to carry objects while moving through human environments. But what happens when a person interacts directly with the payload rather than with the robot? If the payload is unrestrained, the robot must distinguish intentional external interactions from ordinary payload motion, while still keeping the load balanced and maintaining stable locomotion. How can a quadruped infer and compliantly respond to such interactions using only onboard measurements? In this work, we develop a force-aware locomotion framework that treats payload interactions as commands that shape the motion of the combined robot-payload system. Our approach separates the learning of force-aware locomotion and force estimation on an unrestrained payload. We combine a compliant load-carrying policy with a causal force estimator, trained through estimator-in-the-loop data aggregation and finetuning, to predict interactions from onboard robot measurements. Our simulations and real-world experiments show that the resulting controller can maintain stable payload-carrying locomotion, yield compliantly to external interactions, and use the inferred force to support human-guided changes in the robot's trajectory.

Problem

• quadruped locomotion
• payload interaction
• force estimation
• compliant control
• unrestrained payload

Innovation

• quadruped locomotion
• payload interaction
• force estimation
• compliant control
• estimator-in-the-loop