Abstract
In this paper, we investigate the influence of an agent’s morphology on its neural controller. Our model consists of a number of identical modules, each of which comprises two half-wheels for movement and a central pattern generator (CPG) as its own neural control. Based on a series of simulation experiments, we conclude that one single type of CPG can adapt well to different types of morphologies, and that there seems to be a suitable or optimal morphology depending on the environmental givens.