Efficient Model Identification for Tensegrity Locomotion

TL;DR

This paper presents a practical approach to identify unknown physical parameters of a Tensegrity robot using a physics engine and Bayesian optimization, improving locomotion control accuracy.

Contribution

It introduces a method that projects the high-dimensional model identification problem into a lower-dimensional space for efficiency.

Findings

More accurate parameter identification within limited time.

Enhanced locomotion control precision.

Effective use of physics engine and Bayesian optimization.

Abstract

This paper aims to identify in a practical manner unknown physical parameters, such as mechanical models of actuated robot links, which are critical in dynamical robotic tasks. Key features include the use of an off-the-shelf physics engine and the Bayesian optimization framework. The task being considered is locomotion with a high-dimensional, compliant Tensegrity robot. A key insight, in this case, is the need to project the model identification challenge into an appropriate lower dimensional space for efficiency. Comparisons with alternatives indicate that the proposed method can identify the parameters more accurately within the given time budget, which also results in more precise locomotion control.