Beyond Robustness: Learning Unknown Dynamic Load Adaptation for Quadruped Locomotion on Rough Terrain

2025-07-10

AI Summary

To address poor locomotion stability of quadrupedal robots under unknown dynamic payloads and rough terrain, this paper proposes an external-perception-free load-adaptive control framework. First, we formulate a generic implicit dynamical model of payload effects, enabling the robot to infer payload characteristics solely from proprioceptive signals. Second, we design an end-to-end reinforcement learning policy network that jointly optimizes payload state estimation, motion control, and payload stabilization. This unified approach tightly integrates dynamic payload modeling, online proprioceptive inference, and closed-loop locomotion control. In both simulation and real-robot experiments, the method demonstrates superior disturbance rejection, payload stability, and cross-domain generalization—outperforming existing baseline methods significantly under diverse sudden payload changes and highly uneven terrains.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Unknown dynamic load carrying is one important practical application for quadruped robots. Such a problem is non-trivial, posing three major challenges in quadruped locomotion control. First, how to model or represent the dynamics of the load in a generic manner. Second, how to make the robot capture the dynamics without any external sensing. Third, how to enable the robot to interact with load handling the mutual effect and stabilizing the load. In this work, we propose a general load modeling approach called load characteristics modeling to capture the dynamics of the load. We integrate this proposed modeling technique and leverage recent advances in Reinforcement Learning (RL) based locomotion control to enable the robot to infer the dynamics of load movement and interact with the load indirectly to stabilize it and realize the sim-to-real deployment to verify its effectiveness in real scenarios. We conduct extensive comparative simulation experiments to validate the effectiveness and superiority of our proposed method. Results show that our method outperforms other methods in sudden load resistance, load stabilizing and locomotion with heavy load on rough terrain. href{https://leixinjonaschang.github.io/leggedloadadapt.github.io/}{Project Page}.

Problem

• Model unknown dynamic load for quadruped robots
• Capture load dynamics without external sensing
• Stabilize load interaction during rough terrain locomotion

Innovation

• General load modeling captures unknown dynamics
• RL-based control infers load movement indirectly
• Sim-to-real deployment validates rough terrain performance