Design optimization of a robot arm using Kriging modelling and genetic algorithm

Abstract

In this work, the design optimization of a robot arm was conducted by following the Kriging-genetic algorithm method. The method followed has three main steps which are conducting design of experiment, building a response surface, and implementing design optimization.  In this method, Latin Hypercube Sampling is used for conducting the design of experiments. Kriging modelling is used for constructing a response surface. As regard with the optimization technique, the multi-objective genetic algorithm is used to govern the optimization process.  When the results of the method followed was compared with those of the FEM simulation, it was concluded that the method followed can give accurate and precise results for the design optimization. Moreover, at the result of the optimization process, a new robot arm having more strength and less weight was achieved. Also, the case study points out that the method of Kriging-Genetic algorithm can be effectively used in design optimization of mechanical components