Proprioceptive Force Estimation for Quadruped Locomotion and Human-Robot Interaction

2026-09-28

AI Summary

This study addresses the challenge of external force perception in quadruped robots during load-carrying and towing interactions by proposing a unified estimation and control framework that infers three-dimensional forces from proprioceptive history. Methodologically, the force estimator and locomotion policy are jointly trained, leveraging latent context modeling and analytical velocity mapping to achieve cross-task shared representations through a unified interface for both mobility and interaction. Experimental results demonstrate that the proposed approach reduces force estimation error to 1.44 N and improves task survival rate to 94.6%. Furthermore, real-world deployment successfully validates the robot’s capabilities in overloaded transport and guided towing. These findings indicate that the framework significantly enhances environmental adaptability and the robustness of physical interactions for legged robots.

Abstract

Payload forces must be accommodated during locomotion, while leash forces can specify desired motion. We investigate whether a shared three-dimensional force estimate in newtons, inferred from proprioceptive history under sustained loading, can support both tasks. An estimator and locomotion policy are jointly trained with supervised force and velocity outputs and learned latent context. The estimated force conditions locomotion and additionally generates planar-velocity and yaw-rate commands for leash guidance through an analytical map. In sustained-force simulation sweeps, temporal means of componentwise force root mean square error range from 1.44 to 2.83\,N. Compared with a domain-randomized baseline, the framework reduces velocity-tracking and base-orientation error scores by 21.6\% and 46.5\%, respectively, and increases mean survival from 68.29\% to 94.60\% in separate sustained-force tests. Unitree Go1 experiments demonstrate stationary vertical and horizontal force estimation, locomotion with an 8.5\,kg payload whose weight exceeds the 70\,N training force limit, and leash guidance using the same force-estimation interface.

Problem

• Proprioceptive Force Estimation
• Quadruped Locomotion
• Human-Robot Interaction
• Payload Adaptation
• Leash Guidance

Innovation

• Proprioceptive Force Estimation
• Quadruped Locomotion
• Human-Robot Interaction
• Joint Policy Training
• Leash Guidance