Link length determination method for the reduction of the performance deviation of the manipulator: Extension of the valid workspace

Abstract

In this study, a link length determination method for the reduction of the performance deviation in the whole workspace of the manipulator was proposed. The structural length index (SLI), condition number (CN), and global conditioning index (GCI) were selected as performance indices for link length determination. To evaluate deviation, the CN was calculated in the whole workspace, and then the data was used to calculate the coefficient of variation. The Taguchi method was used to decrease the time of optimization for the whole workspace, and Grey relational analysis was performed to convert the values for multiple objective functions to one value defined by the Grey relational grade. Furthermore, the effects of the design factors on the objective function were analyzed through analysis of variance. Optimization was performed twice for each objective function to compare the result of the considering the performance deviation and the result of the considering the performance mean. Each objective function consists of (A) the SLI and the coefficient of variation of the CN and (B) the SLI and GCI. The validity of the proposed method was verified by comparing the results of (A) and (B) for performance deviation, mean and the rate of valid workspace.