Learning Dynamical System for Grasping Motion

TL;DR

This paper introduces a novel dynamical system framework that couples position and orientation for robotic grasping, demonstrating improved synchronization and adaptability in online experiments.

Contribution

It proposes a new method to learn coupled position-orientation dynamical systems using diffeomorphisms, enhancing grasping motion coordination.

Findings

Effective synchronization of position and orientation during grasping.

Demonstrated online adaptability in real-time experiments.

Improved robustness to environmental perturbations.

Abstract

Dynamical System has been widely used for encoding trajectories from human demonstration, which has the inherent adaptability to dynamically changing environments and robustness to perturbations. In this paper we propose a framework to learn a dynamical system that couples position and orientation based on a diffeomorphism. Different from other methods, it can realise the synchronization between positon and orientation during the whole trajectory. Online grasping experiments are carried out to prove its effectiveness and online adaptability.