Biologically-based Control Of Omnidirectional Leg Coordination

Abstract

Animals have been engineered by evolution to adapt and survive in a dynamic environment. We are studying the neural control mechanisms of the American lobster and applying the resultant neurophysiological models to robot control. Our long term goal is to study the utility of biological models to robot control for an underwater walking machine. Our robot architecture is based on a well accepted biological model that includes command systems, coordinating systems, and central pattern generation. We focus, in this paper, on the central pattern generator and coordinating systems. The central pattern generator controls the motion of the leg appropriate to walking directions with and without sensory feedback. The coordinating systems control the gait pattern of the legs. We have built a simulator which models the central pattern generator and coordination and demonstrates basic walking motions of the leg in any direction and at multiple speeds. We also simulate the coordination of a stable gait. Our future research includes testing our simulated results on an eight legged walking machine based on the morphology of the lobster, and adding sensors, reflexes, and taxes to our robotic model.