Hybrid design of PID controller for four DoF lower limb exoskeleton

Abstract

Listen

Translate article

• Mathematical model for Four DoF Lower Limb Exoskeleton was examined. • Hybrid design of PID controller is based on combination of Genetic Algorithm and Particle Swarm Optimization. • Numerical analysis of the combined algorithm indicates advantages of such controller. • The algorithm was applied to acquire the parameters of the controller for each joint for minimizing trajectories’ errors. In this paper, a method of tuning a proportional-integral-derivative controller for a four degree-of-freedom lower limb exoskeleton using hybrid of genetic algorithm and particle swarm optimization is presented. Transfer function of each link of the lower limb exoskeleton acquired from a pendulum model, was used in a closed-loop proportional-integral-derivative control system, while each link was assumed as one degree-of-freedom linkage. In the control system, the hybrid algorithm was applied to acquire the parameters of the controller for each joint for minimizing the error. The algorithm started with genetic algorithm and continued via particle swarm optimization. Furthermore, a 3-dimensional model of the lower limb exoskeleton was simulated to validate the proposed controller. The trajectory of the control system with optimized proportional-integral-derivative controller via hybrid precisely follows the input signal of the desired. The result of the hybrid optimized controller was compared with genetic algorithm and particle swarm optimization based on statistics. The average error of the proposed algorithm showed the optimized results in comparison with genetic algorithm and particle swarm optimization. Furthermore, the advantages of the hybrid algorithm have been indicated by numerical analysis.