Real time generation of humanoid robot optimal gait for going upstairs using intelligent algorithms

Abstract

Going upstairs is a common humanoid robot task. In this paper, a genetic algorithm (GA) gait synthesis method for going upstairs and a radial basis function neural network (RBFNN) implementation, are considered. The gait synthesis is analyzed based on the minimum consumed energy and minimum torque change. The proposed method can easily be applied to generate the angle trajectories for going downstairs, overcoming obstacles, etc. In our work, the stability is verified through the ZMP concept. For the real time implementation, a RBFNN which is taught based on the GA results, is considered. The RBFNN generates the optimal gait in a very short time, where the input variables are the step length, step height and step time. Simulations are realized based on the parameters of the “Bonten‐Maru I” humanoid robot.