Robot control and the evolution of modular neurodynamics
Frank Pasemann, Ulrich Steinmetz, Martin Hülse, and Bruno Lara
Contact the author: Please use for correspondence this email.
Submission date: 12. Jul. 2001
published in: Theory in biosciences, 3 (2001), p. 311-326
DOI number (of the published article): 10.1078/1431-7613-00047
Download full preprint: PDF (375 kB), PS ziped (223 kB)
A modular approach to neural behavior control of autonomous robots is presented. It is based on the assumption that complex internal dynamics of recurrent neural networks can efficiently solve complex behavior tasks. For the development of appropriate neural control structures an evolutionary algorithm is introduced, which is able to generate neuromodules with specific functional properties, as well as the connectivity structure for a modular synthesis of such modules. This so called ENS3-algorithm does not use genetic coding. It is primarily designed to develop size and connectivity structure of neuro-controllers. But at the same time it optimizes also parameters of individual networks like synaptic weights and bias terms. For demonstration, evolved networks for the control of miniature Khepera robots are presented. The aim is to develop robust controllers in the sense that neuro-controllers evolved in a simulator show comparably good behavior when loaded to a real robot acting in a physical environment. Discussed examples of such controllers generate obstacle avoidance and phototropic behaviors in non-trivial environments.