Humanoid robot gait planning resorting to an adaptive simulated annealing algorithm

This paper addresses a new approach to optimize gait parameter sets using an Adaptive Simulated Annealing optimization algorithm. It presents the issue of reducing energy consumption with gait planning optimization. The developed realistic simulator, based on a real and commercial robot, emulates the mechanical and electrical aspects of the humanoid robot and reduces the software and prototyping production time. It also allows to develop and test different algorithms and optimization procedures. The developed simulator is used to optimize the energy consumption and the distance travelled by the robot in each step based on a Adaptive Simulated annealing optimization approach. Final results prove the benefits of the presented approach