Learning the Geometric Mechanics of Robot Motion Using Gaussian Mixtures

TL;DR

This paper introduces a Gaussian Mixture Model approach to learn the geometric mechanics of robot motion, improving prediction accuracy and enabling application to diverse datasets beyond periodic gaits.

Contribution

The paper presents a novel GMM-based method for manifold learning of geometric mechanics models, enhancing prediction and applicability to various robot motion data.

Findings

Significant improvement in motion prediction accuracy.

Applicable to non-periodic and diverse motion datasets.

Pre-processing enhances extrapolation in linear regions.

Abstract

Data-driven models of robot motion constructed using principles from Geometric Mechanics have been shown to produce useful predictions of robot motion for a variety of robots. For robots with a useful number of DoF, these geometric mechanics models can only be constructed in the neighborhood of a gait. Here we show how Gaussian Mixture Models (GMM) can be used as a form of manifold learning that learns the structure of the Geometric Mechanics "motility map" and demonstrate: [i] a sizable improvement in prediction quality when compared to the previously published methods; [ii] a method that can be applied to any motion dataset and not only periodic gait data; [iii] a way to pre-process the data-set to facilitate extrapolation in places where the motility map is known to be linear. Our results can be applied anywhere a data-driven geometric motion model might be useful.