This paper presents an approach to incrementally learn a reshaping term that modifies the trajectories of an autonomous dynamical system without affecting its stability properties. The reshaping term is considered as an additive control input and it is incrementally learned from human demonstrations using Gaussian process regression. We propose a novel parametrization of this control input that preserves the time-independence and the stability of the reshaped system, as analytically proved in the performed Lyapunov stability analysis. The effectiveness of the proposed approach is demonstrated with simulations and experiments on a real robot.
Incremental Motion Reshaping of Autonomous Dynamical Systems
Springer Proceedings in Advanced Robotics
International Workshop on Human-Friendly Robotics ; 2019 ; Reggio Emilia, Italy October 24, 2019 - October 25, 2019
2020-02-21
15 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Incremental Motion Reshaping of Autonomous Dynamical Systems
TIBKAT | 2020
|Incremental Skill Learning of Stable Dynamical Systems
Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2018
|An Incremental Stochastic Motion Planning Technique for Autonomous Underwater Vehicles
British Library Conference Proceedings | 2005
|Motion control of mobile autonomous robots using non-linear dynamical systems approach
BASE | 2017
|