Skip to main content

Research Repository

Advanced Search

An adaptive modular recurrent cerebellum-inspired controller

Maheri, Kiyan; Lenz, Alex; Pearson, Martin J


Kiyan Maheri

Alex Lenz


Michael Mangan

Cutkosky Mark

Anna Mura

Paul Verschure

Tony Prescott

Lepora Nathan


Animals and robots face the common challenge of interacting with an unstructured environment. While animals excel and thrive in such environments, modern robotics struggles to effectively execute simple tasks. To help improve performance in the face of frequent changes in the mapping between action and outcome (change in context) we propose the Modular-RDC controller, a bio-inspired controller based on the Recurrent Decorrelation Control (RDC) architecture. The proposed controller consists of multiple modules, each containing a forward and inverse model pair. The combined output of all inverse models is used to control the plant, with the contribution of each inverse model determined by a responsibility factor. The controller is able to correctly identify the best module for the current context, enabling a significant reduction of 70.9% in control error for a context-switching plant. It is also shown that the controller results in a degree of generalization in control.


Maheri, K., Lenz, A., & Pearson, M. J. (2017). An adaptive modular recurrent cerebellum-inspired controller. In M. Mangan, C. Mark, A. Mura, P. Verschure, T. Prescott, & L. Nathan (Eds.),

Conference Name Biomimetic and Biohybrid Systems, Living Machines 2017
Start Date Jul 26, 2017
End Date Jul 28, 2017
Acceptance Date Jul 16, 2017
Publication Date Jul 1, 2017
Deposit Date Aug 21, 2017
Journal Biomimetic and Biohybrid systems
Peer Reviewed Peer Reviewed
Pages 267-278
Series Title Lecture Notes in Computer Science
Series Number 10384
Series ISSN 0302-9743
ISBN 9783319635361
Keywords adaptic control, cerebellar inspired
Public URL


You might also like

Downloadable Citations