Optimal robust sliding mode tracking control of a biped robot based on ingenious multi-objective PSO

The aim of this paper is to present novel Multi-objective Particle Swarm Optimization (MOPSO) called Ingenious-MOPSO and compare its capability with three well-known multi-objective optimization algorithms, modified NSGAII, Sigma method, and MOGA. The application of this investigation is on an intellectual challenge in robotics, that is, a biped robot walking in the lateral plane on slope. Recently, a number of researches have been done on the walking of biped robots in the sagittal plane; however, biped robots require the ability to step purely in the lateral plane in facing obstruction, such as a wall. Hence, this paper introduces an optimal robust sliding tracking controller tuned by Ingenious-MOPSO to address the problem of heavy nonlinear dynamics and tracking systems of the biped robots which walk in the lateral plane on slope. Two phases of a biped robot, single support phase and double support phase; and also impact are regarded to control the robot. In the sliding mode controller, the heuristic parameters are usually determined by a tedious and repetitive trial-and-error process. By using Ingenious-MOPSO, the trial-and-error process is eliminated and the optimal parameters are chosen based on the design criteria. In the proposed algorithm, Ingenious-MOPSO, the rate of convergence and diversity of solutions are enhanced simultaneously, and innovative methods are proposed to select the global and personal best positions for each particle. Moreover, a new fuzzy elimination technique is suggested for shrinking the archive which promotes the diversity of solutions. A turbulence operator is utilized to evade local optima, for further improving the search ability. Numerical results and analysis demonstrate the superiority of Ingenious-MOPSO over three effectual multi-objective optimization algorithms.