Skip to content

Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing

Authors: Haoru Xue, Chaoyi Pan, Zeji Yi, Guannan Qu, Guanya Shi

Published: 2024 (Conference Paper)

Source: IEEE International Conference on Robotics and Automation (ICRA)

Algorithm: DIAL-MPC

arXiv: 2409.15610

DOI: 10.1109/ICRA55743.2025.11127320

Summary

Takes a perspective of treating MPPI as a single step of a denoising diffusion process, and generalizes this process to a multi-step diffusion-style annealing process. DIAL-MPC starts optimizing the control sequence with smooth but inaccurate objectives and gradually shifts to more accurate local objectives.

Abstract

Due to high dimensionality and non-convexity, real-time optimal control using full-order dynamics models for legged robots is challenging. Therefore, Nonlinear Model Predictive Control (NMPC) approaches are often limited to reduced-order models. Sampling-based MPC has shown potential in nonconvex even discontinuous problems, but often yields suboptimal solutions with high variance, which limits its applications in high-dimensional locomotion. This work introduces DIAL-MPC (Diffusion-Inspired Annealing for Legged MPC), a sampling-based MPC framework with a novel diffusion-style annealing process. Such an annealing process is supported by the theoretical landscape analysis of Model Predictive Path Integral Control (MPPI) and the connection between MPPI and single-step diffusion. Algorithmically, DIAL-MPC iteratively refines solutions online and achieves both global coverage and local convergence. In quadrupedal torque-level control tasks, DIAL-MPC reduces the tracking error of standard MPPI by 13.4 times and outperforms reinforcement learning (RL) policies by 50% in challenging climbing tasks without any training. In particular, DIAL-MPC enables precise real-world quadrupedal jumping with payload. To the best of our knowledge, DIAL-MPC is the first training-free method that optimizes over full-order quadruped dynamics in real-time.

Primary

Paper arxiv.org

Standard

arXiv Abstract 2409.15610 arXiv PDF 2409.15610 arXiv HTML 2409.15610 DOI 10.1109/ICRA55743.2025.11127320

Alternate

ieeexplore.ieee.org ieeexplore.ieee.org lecar-lab.github.io lecar-lab.github.io GitHub: LeCAR-Lab/dial-mpc github.com

Tags

  • Trajectory optimization

  • MPPI

  • Diffusion

  • Locomotion

  • Model predictive control

  • Annealing