Learning Near-global-optimal Strategies for Hybrid Non-convex Model Predictive Control of Single Rigid Body Locomotion

TL;DR

This paper introduces a learning-based approach to solve complex non-convex model predictive control problems for single rigid body locomotion, enabling near-global optimal solutions in real-time for diverse gaits.

Contribution

It removes assumptions of convex MPCs by solving mixed-integer non-convex problems and learns a fast problem-solution map for real-time control.

Findings

Achieved online gait generation at over 50 Hz.

Demonstrated adaptation and diverse gait behaviors.

Close to global optimality with warm-starts.

Abstract

Convex model predictive controls (MPCs) with a single rigid body model have demonstrated strong performance on real legged robots. However, convex MPCs are limited by their assumptions such as small rotation angle and pre-defined gait, limiting the richness of potential solutions. We remove those assumptions and solve the complete mixed-integer non-convex programming with single rigid body model. We first collect datasets of pre-solved problems offline, then learn the problem-solution map to solve this optimization fast for MPC. If warm-starts can be found, offline problems can be solved close to the global optimality. The proposed controller is tested by generating various gaits and behaviors depending on the initial conditions. Hardware test demonstrates online gait generation and adaptation running at more than 50 Hz based on sensor feedback.