Genetically optimized parameter estimation of mathematical model for multi-joints hip–knee exoskeleton

Abstract

Achieving precise parameters of multi-joints actuators for Hip–Knee Exoskeleton (HKE) is a crucial process due to its non-linear characteristics. In this paper, a Genetic Algorithm (GA) based optimization is used for parameter estimation of the mathematical model for a four-Degree of Freedom (DoF) multi-joint HKE, which is a type of Lower Limb Exoskeleton (LLE). Mathematical model for electro-mechanical, mechanical, and electrical components of the HKE has been formulated, and its parameters are estimated using GA and experimental method. An objective function is determined based on the difference between the simulated and actual angular trajectory for each joint. The performance of the mathematical model is examined with different voltages under the range of 4 V to 8 V for hip and knee, respectively. Furthermore, the performance of the estimated model is compared with Particle Swarm Optimization (PSO). The results and numerical analysis demonstrated that the estimated model by GA and PSO with varying voltages predicted the actual angular trajectory with acceptable error, while GA provides the more accurate model. It can be ascertained that the proposed method of estimation for mathematical model of the HKE is applicable to identify its parameters, and useful for designing a control system.