Bio-inspired adaptive feedback error learning architecture for motor control

Abstract

This study proposes an adaptive control architecture based on an accurate regression method called Locally Weighted Projection Regression (LWPR) and on a bio-inspired module, such as a cerebellar-like engine. This hybrid architecture takes full advantage of the machine learning module (LWPR kernel) to abstract an optimized representation of the sensorimotor space while the cerebellar component integrates this to generate corrective terms in the framework of a control task. Furthermore, we illustrate how the use of a simple adaptive error feedback term allows to use the proposed architecture even in the absence of an accurate analytic reference model. The presented approach achieves an accurate control with low gain corrective terms (for compliant control schemes). We evaluate the contribution of the different components of the proposed scheme comparing the obtained performance with alternative approaches. Then, we show that the presented architecture can be used for accurate manipulation of different objects when their physical properties are not directly known by the controller. We evaluate how the scheme scales for simulated plants of high Degrees of Freedom (7-DOFs).

Publication
Biological Cybernetics
Niceto Luque
Niceto Luque
Associate Professor

Associate Professor at the Department of Computer Engineering, Automation and Robotics and Principal Investigator at the Applied Computational Neuroscience Group.

Jesús Garrido
Jesús Garrido
Associate Professor

Associate professor in Computation technology, senior researcher at the Computational Neuroscience and Neurorobotics Lab and principal investigator of the VALERIA lab of the University of Granada.

Eduardo Ros
Eduardo Ros
Full Professor

Full professor in computer architecture, principal investigator at the Computational Neuroscience and Neurorobotics Lab and principal investigator of the VALERIA lab of the University of Granada.