Learning Differentiable Tensegrity Dynamics using Graph Neural Networks

TL;DR

This paper introduces a graph neural network-based simulator for tensegrity robot dynamics, offering improved accuracy and efficiency over previous methods, and demonstrating successful simulation and real-world application.

Contribution

It proposes a novel GNN-based contact dynamics model for tensegrity robots, outperforming prior differentiable physics engines and mesh-based GNN simulators in accuracy and computational efficiency.

Findings

Accurately models 3-bar and 6-bar tensegrity robot dynamics in simulation.

Achieves higher accuracy than previous differentiable engines on real robot data.

More computationally efficient than mesh-based GNN simulators.

Abstract

Tensegrity robots are composed of rigid struts and flexible cables. They constitute an emerging class of hybrid rigid-soft robotic systems and are promising systems for a wide array of applications, ranging from locomotion to assembly. They are difficult to control and model accurately, however, due to their compliance and high number of degrees of freedom. To address this issue, prior work has introduced a differentiable physics engine designed for tensegrity robots based on first principles. In contrast, this work proposes the use of graph neural networks to model contact dynamics over a graph representation of tensegrity robots, which leverages their natural graph-like cable connectivity between end caps of rigid rods. This learned simulator can accurately model 3-bar and 6-bar tensegrity robot dynamics in simulation-to-simulation experiments where MuJoCo is used as the ground truth. It can also achieve higher accuracy than the previous differentiable engine for a real 3-bar tensegrity robot, for which the robot state is only partially observable. When compared against direct applications of recent mesh-based graph neural network simulators, the proposed approach is computationally more efficient, both for training and inference, while achieving higher accuracy. Code and data are available at https://github.com/nchen9191/tensegrity_gnn_simulator_public