Multi-Objective Optimization of Parallel Tracking Mechanism Considering Parameter Uncertainty

Abstract

Multi-objective optimization of a typical parallel tracking mechanism considering parameter uncertainty is carried out in this paper. Both dimensional and sectional parameters are regarded as design variables. Workspace, kinematic, stiffness, and dynamic performances are simultaneously considered in formulating optimal objectives and constraint conditions. Considering manufacturing and assembling errors, parameter uncertainty is modeled and evaluated to minimize their effects on the optimized performances. Analytical models between objectives and design variables are established to improve the efficiency of optimization while its accuracy is assured. The study of parameter uncertainty and analytical mapping model is incorporated in the optimization of the parallel tracking mechanism. With the aid of particle swarm algorithm, a cluster of solutions, called Pareto frontier, are obtained. By proposing an index, a cooperative equilibrium point representing the balance among objectives is selected and the optimized parameters are determined. The present study is expected to help designers build optimized parallel tracking mechanism in an effective and efficient manner.