Back
Distributed Online Learning of Central Pattern Generators in Modular Robots
In this paper we study distributed online learning of locomotion gaits for modular robots. The learning is based on a stochastic ap- proximation method, SPSA, which optimizes the parameters of coupled oscillators used to generate periodic actuation patterns. The strategy is implemented in a distributed fashion, based on a globally shared reward signal, but otherwise utilizing local communication only. In a physics-based simulation of modular Roombots robots we experiment with online learn- ing of gaits and study the effects of: module failures, different robot morphologies, and rough terrains. The experiments demonstrate fast online learning, typically 5-30 min. for convergence to high performing gaits (≈ 30 cm/sec), despite high numbers of open parameters (45-54). We conclude that the proposed approach is efficient, effective and a promising candidate for online learning on many other robotic platforms.
@incollection{escidoc:2316385, title = {Distributed Online Learning of Central Pattern Generators in Modular Robots}, booktitle = {From Animals to Animats 11}, abstract = {In this paper we study distributed online learning of locomotion gaits for modular robots. The learning is based on a stochastic ap- proximation method, SPSA, which optimizes the parameters of coupled oscillators used to generate periodic actuation patterns. The strategy is implemented in a distributed fashion, based on a globally shared reward signal, but otherwise utilizing local communication only. In a physics-based simulation of modular Roombots robots we experiment with online learn- ing of gaits and study the effects of: module failures, different robot morphologies, and rough terrains. The experiments demonstrate fast online learning, typically 5-30 min. for convergence to high performing gaits (≈ 30 cm/sec), despite high numbers of open parameters (45-54). We conclude that the proposed approach is efficient, effective and a promising candidate for online learning on many other robotic platforms.}, volume = {6226}, pages = {402--412}, series = {{Lecture Notes in Computer Science}}, publisher = {Springer}, address = {Berlin}, year = {2010}, note = {author: Doncieux, Stéphan}, slug = {escidoc-2316385}, author = {Christensen, David Johan and Spr{\"o}witz, Alexander and Ijspeert, Auke Jan} }
More information