Home > Published Issues > 2020 > Volume 9, No. 12, December 2020 >

Multi-Objective Optimization Algorithm of Humanoid Robot Walking on a Narrow Beam

Kittisak Sanprasit
Department of Electrical and Electronic Engineering, Faculty of Industrial Technology, Loei Rajabhat University, Loei, Thailand

Abstract— The humanoid robots associate and work with human, they should be able to access anywhere people can even in complex and harsh environments. This research proposes an optimal path design for humanoid robots to walk on a narrow beam. The experiments and simulations of a commercial humanoid robot (Bioloid Premium Type A), with 18 Degree of Freedoms (DOFs) were conducted. The multi-objective optimization was used to design the walking path of the robots on the beam by comparing four algorithms: MOWOA, MOGWO, MOHS, and MOGA. The performance comparison was made based on the hypervolume (HV) indicator. The optimal points were chosen from non-dominated solutions by MCDM method and minimization weighted sum method (WSM). There were two objective functions: 1) maximum postural stability of a humanoid robot walk and 2) minimal jerk. 

Index Terms— humanoid robot, MOWOA, MOGWO, MOHS, MOGA, weighted sum method, hypervolume, Pareto front

Cite: Kittisak Sanprasit, "Multi-Objective Optimization Algorithm of Humanoid Robot Walking on a Narrow Beam," International Journal of Mechanical Engineering and Robotics Research, Vol. 9, No. 12, pp. 1548-1559, December 2020. DOI: 10.18178/ijmerr.9.12.1548-1559

Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.