Skip to main navigation Skip to search Skip to main content

RL-Compensated Model Predictive Control for Quadruped Robot Locomotion on Challenging Terrains

  • Zhefeng Xiao
  • , Dongdong Zheng*
  • , Zeyuan Sun
  • , Yi Zeng
  • *Corresponding author for this work
  • Beijing Institute of Technology
  • Ltd.
  • China North Artificial Intelligence & Innovation Research Institute
  • Collective Intelligence & Collaboration Laboratory

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Model Predictive Control (MPC) has been widely applied in quadruped robot locomotion. However, the method relies heavily on accuracy of the system model and the feasibility of the pre-planned desired trajectory. Reinforcement learning (RL) improves the performance of the target policy through continuous interaction between the robot and the environment. However, it cannot guarantee the safety of the robot’s actions, and the design of the reward function is a cumbersome process. In this paper, An RL-compensated MPC algorithm framework is proposed. We simplify the quadruped robot into a single rigid body (SRB) dynamic model and train an RL policy to compensate for the linear acceleration, angular acceleration, gait frequency and foothold location of the robot. Comparative experiments against the MPC algorithm in simulation verify that the proposed framework can improve the robot’s locomotion performance on irregular terrains.

Original languageEnglish
Title of host publicationNeuromorphic Computing - 4th International Conference, ICNC 2025, Revised Selected Papers
EditorsChuandong Li, Qi Zhou, Hongjing Liang, Jingang Lai, Bin Li, Kaibo Shi
PublisherSpringer Science and Business Media Deutschland GmbH
Pages209-219
Number of pages11
ISBN (Print)9789819215980
DOIs
Publication statusPublished - 2026
Externally publishedYes
Event4th International Conference on Neuromorphic Computing, ICNC 2025 - Chengdu, China
Duration: 12 Dec 202514 Dec 2025

Publication series

NameCommunications in Computer and Information Science
Volume2946 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference4th International Conference on Neuromorphic Computing, ICNC 2025
Country/TerritoryChina
CityChengdu
Period12/12/2514/12/25

Keywords

  • Locomotion control
  • Model predictive control
  • Quadruped robots
  • Reinforcement learning

Fingerprint

Dive into the research topics of 'RL-Compensated Model Predictive Control for Quadruped Robot Locomotion on Challenging Terrains'. Together they form a unique fingerprint.

Cite this