Evolving Recurrent Dynarnical Networks for Robot Control
Dave Cliff, Phil Husbands, Inman R. Harvey · 1993
This paper describes aspects of our research into the development of artificial evolution techniques for the creation of control systems for autonomous mobile robots operating in complex, noisy and generally hos tile environments. At the heart of our method is an extended genetic algorithm which allows the open ended formation of control architectures based on ar tificial neural networks. After outlining the ratio nale for our work, and giving the background to our techniques and experimental method, results are pre sented from experiments in which we contrast the be haviours of robots evolved under the same evaluation function but with different sensory capabilities. One set of robots have tactile sensing only, whereas the other have both tactile and primitive visual sensors. Although the evaluation task does not explicitly rely on vision, results show conclusively that evolution is able to exploit visual input to produce very successful controllers, far better than those for robots without VISIOn.