Building an energetic model to evaluate and optimize power consumption in walking robots

Abstract

An energetic model for walking robots based on both dynamic and actuator models is proposed in this paper. While applied to walking machines, this method allows the evaluation of the influence of leg configuration, body weight, and gait parameters on power consumption. The model is validated by using genetic algorithms to identify the unknown parameters, which enables it to be used as a tool to evaluate and optimize the performance of a legged robot configuration according to the power consumption. This method has been applied to find the optimum stride length for the minimum energy expenditure of a biped prototype depending on the speed and payload, considering both level and slope walking.