Whole-Body Nonlinear Model Predictive Control Through Contacts for Quadrupeds

TL;DR

This paper introduces a real-time whole-body nonlinear model predictive control method for quadruped robots that optimizes contact locations and timings dynamically, demonstrating high-speed computation and robust hardware performance.

Contribution

It presents a novel nonlinear MPC framework that includes explicit contact dynamics and runs at 190 Hz, significantly faster than existing methods.

Findings

Achieved real-time control at 190 Hz on quadruped robots.

Successfully handled both periodic and non-periodic tasks.

Demonstrated robustness and transferability across different hardware.

Abstract

In this work we present a whole-body Nonlinear Model Predictive Control approach for Rigid Body Systems subject to contacts. We use a full dynamic system model which also includes explicit contact dynamics. Therefore, contact locations, sequences and timings are not prespecified but optimized by the solver. Yet, thorough numerical and software engineering allows for running the nonlinear Optimal Control solver at rates up to 190 Hz on a quadruped for a time horizon of half a second. This outperforms the state of the art by at least one order of magnitude. Hardware experiments in form of periodic and non-periodic tasks are applied to two quadrupeds with different actuation systems. The obtained results underline the performance, transferability and robustness of the approach.