Emerging Common Functional Principles in Intelligent Control Theory and the Vertebrate Brain: A Case Study with Autonomous Vehicle Control

Abstract

This paper describes emergent neurobiological characteristics of an intelligent multiple-controller that has been developed for simultaneously controlling the throttle, brake and steering subsystems of a validated vehicle model. Sample simulation results demonstrate the effectiveness of the proposed approach. While not explicitly designed for, the controller exhibits discrete behaviours, developed under adaptive tuning, governed by its two component controllers. These controllers are selected according to task demands evaluated by a fuzzy logic based supervisor. The system therefore displays ‘action selection’ under central switched control, as has been proposed to take place in the vertebrate brain. Further biological links are made by likening the supervisor and modular controllers to higher and lower levels of functionality, respectively, in a layered architecture similar to that found in the brain which provides successive levels of sophistication in sensorimotor control. The multiple-controller also uses an internal (predictive) model of the plant which is analogous to cognitive knowledge of action-outcome associations in the brain. Further, the plant model makes use of sensory prediction errors to shape the behaviours encoded in the modular controllers; this resonates with recently proposed ideas about the role of phasic dopamine in reinforcement learning in the brain. We conclude that advances in neuroscience and control theory have reached a critical mass which make it timely for a new rapprochement of these disciplines.

7 Figures and Tables

Cite this paper

@inproceedings{Hussain2008EmergingCF, title={Emerging Common Functional Principles in Intelligent Control Theory and the Vertebrate Brain: A Case Study with Autonomous Vehicle Control}, author={Amir Hussain and Kevin N. Gurney and Rudwan Abdullah and Jon Chambers and Pete Redgrave}, year={2008} }