Abstract
This article presents a connectionist model of a central pattern generator for salamander locomotion. A 3D biomechanical simulation of the salamander's body is developed whose muscle con traction the locomotion controller simulated as a leaky-integrator neural network determines. While the connectivity of the neural circuitry underlying locomotion in the salamander has not been decoded for the moment, the general organization of the designed neural circuit corresponds to that hypothesized by neurobiologists for the real animal. In particular, the locomotion controller is based on a body central pattern generator (CPG) corresponding to a lamprey-like swimming controller, and is extended with a limb CPG for controlling the salamander's limbs. A genetic algorithm is used to instantiate synaptic weights of the connections within the limb CPG, and from the limb CPG to the body CPG, given a high level description of the desired gaits. A controller is thus developed which can produce a neural activity and locomotion gaits very similar to those observed in the real salamander. By varying the tonic excitation applied to the network, the speed, direction and type of gait can be varied.