Multi-objective optimization for a humanoid robot climbing stairs based on Genetic Algorithms

Abstract

A new method of gait optimization for a humanoid robot climbing stairs based on multi-objective Genetic algorithm(GA) was proposed in this paper. Based on the humanoid robot model and the staircase model, the complicated process of climbing stairs was parameterized, and a climbing stairs mode was built. A math function, which takes the stability judgment based on zero moment point (ZMP) and energy minimization as targets, was established and solved with Genetic Algorithm, and the optimum solution was used as the basis of gait planning. The simulation results showed the feasibility and effectiveness of the method.