Control of Legged Robots using Model Predictive Optimized Path Integral

TL;DR

This paper introduces MPOPI, a novel sampling-based predictive control method combining MPPI, CE, and CMA to improve real-time, efficient locomotion control for legged robots in complex environments.

Contribution

It presents a new control strategy that enhances sample efficiency and locomotion performance for legged robots by integrating MPPI with CE and CMA methods.

Findings

MPOPI outperforms traditional MPPI in sample efficiency.

MPOPI enables real-time, adaptable control in complex scenarios.

Simulation results show improved locomotion capabilities.

Abstract

Legged robots possess a unique ability to traverse rough terrains and navigate cluttered environments, making them well-suited for complex, real-world unstructured scenarios. However, such robots have not yet achieved the same level as seen in natural systems. Recently, sampling-based predictive controllers have demonstrated particularly promising results. This paper investigates a sampling-based model predictive strategy combining model predictive path integral (MPPI) with cross-entropy (CE) and covariance matrix adaptation (CMA) methods to generate real-time whole-body motions for legged robots across multiple scenarios. The results show that combining the benefits of MPPI, CE and CMA, namely using model predictive optimized path integral (MPOPI), demonstrates greater sample efficiency, enabling robots to attain superior locomotion results using fewer samples when compared to typical MPPI algorithms. Extensive simulation experiments in multiple scenarios on a quadruped robot show that MPOPI can be used as an anytime control strategy, increasing locomotion capabilities at each iteration.