Inverse Dynamics Trajectory Optimization for Contact-Implicit Model Predictive Control

TL;DR

This paper presents a real-time contact-implicit model predictive control method using inverse dynamics trajectory optimization, enabling fast and effective manipulation and locomotion tasks on robots, including hardware demonstrations at over 100 Hz.

Contribution

It introduces a series of innovations to inverse dynamics trajectory optimization that enable real-time contact-implicit model predictive control for complex robotic tasks.

Findings

Real-time control achieved at over 100 Hz on hardware

Effective manipulation and locomotion demonstrated in simulation

Open-source solver implementation supports various tasks

Abstract

Robots must make and break contact with the environment to perform useful tasks, but planning and control through contact remains a formidable challenge. In this work, we achieve real-time contact-implicit model predictive control with a surprisingly simple method: inverse dynamics trajectory optimization. While trajectory optimization with inverse dynamics is not new, we introduce a series of incremental innovations that collectively enable fast model predictive control on a variety of challenging manipulation and locomotion tasks. We implement these innovations in an open-source solver and present simulation examples to support the effectiveness of the proposed approach. Additionally, we demonstrate contact-implicit model predictive control on hardware at over 100 Hz for a 20-degree-of-freedom bi-manual manipulation task. Video and code are available at https://idto.github.io.