Evolving neural behaviour control for autonomous robots
Martin Hülse, Bruno Lara, Frank Pasemann, and Ulrich Steinmetz
Contact the author: Please use for correspondence this email.
Submission date: 04. Apr. 2001
published in: Artificial neural networks / G. Dorffner ... (eds.)
Berlin : Springer, 2001. - P. 957 - 962
(Lecture notes in computer science ; 2130)
DOI number (of the published article): 10.1007/3-540-44668-0_132
Download full preprint: PDF (322 kB), PS ziped (344 kB)
An evolutionary algorithm for the creation of recurrent network structures is presented. The aim is to develop neural networks controlling the behaviour of miniature robots. Two different tasks are solved with this approach. For the first, the agents are required to move within an environment without colliding with obstacles. In the second task, the agents are required to move towards a light source. The evolution process is carried out in a simulated environment and individuals with high performance are also tested on a physical environment with the use of Khepera robots.