Multiple CPGs with Local Feed- back Mechanisms for Locomo- tor Adaptation of Hexapod Robots

Walking animals show versatile locomotion. They can also adapt their movement according to changes in their morphology and environmental conditions. These emergent properties are realized by biomechanics, distributed central pattern generators (CPGs), local sensory feedback, and their interactions during body and leg movements through the environment. Based on this concept, a decentralized control system with different local sensory feedback mechanisms is presented here. In this system, the intra-limb coordination is formed by multiple single CPG-based controls, while the inter-limb coordination is attained by the interactions between body dynamics, neural control, and the environment through local sensory feedback of each leg. Simulation results show that these feedback control approaches can generate self-organizing emergent locomotion allowing the robot to adaptively form regular patterns, to deal with morphological change and special functionalities of the front legs, to maintain its body position while traversing irregular terrains, to achieve high walking speeds, and to synchronize its movement with another robot during a collaborative task. keywords: Adaptive behavior, Hexapod locomotion, Brain-body-environment interaction, Autonomous robots, Neural networks.