Abstract
Recently, Evolutionary Robotics (ER) approach has been attracting a lot of concern in the field of robotics and artificial life. In this approach, neural networks are widely used to construct controllers for autonomous mobile robots, since they intrinsically have generalization, noise-tolerant abilities and so on. However, the followings are still open questions: 1) the gap between simulated and real environments, 2) the evolutionary and learning phase are completely separated, and 3) the conflict between stability and evolvability/adaptability. In this paper, we particularly focus on the evolvability. In order to overcome this problem, we propose a concept of dynamically rearranging function of neural networks by neuromodulators. We apply this concept to construct a gait controller of a six-legged robot by carrying out simulations.
