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

Going upstairs is a common humanoid robot task. In this paper, a genetic al gorithm (GA) gait synthesis method for going upstairs and a radial basis fu nction neural network (RBFNN) implementation, are considered. The gait synt hesis is analyzed based on the minimum consumed energy and minimum torque c hange. The proposed method can easily be applied to generate the angle traj ectories for going downstairs, overcoming obstacles, etc. In our work, the stability is verified through the ZMP concept. For the real time implementa tion, a RBFNN which is taught based on the GA results, is considered. The R BFNN generates the optimal gait in a very short time, where the input varia bles are the step length, step height and step time. Simulations are realiz ed based on the parameters of the "Bonten-Maru I" humanoid robot.