Using Taguchi methods to determine/optimize robot process capability for path following

A methodology based on Taguchi methods was implemented to determined/optimize robot process capability (RPC) for path following. The methodology consists of the characterization of the robot path data, the experimental design, the data analysis procedure, and the verification of the results. Four figures of merit (FOMs) for accuracy and four FOMs for repeatability were used to characterize a robot's controlled path capability. The experiment considered seven controllable factors (i.e. load, speed, shape, size, direction, orientation, and height) and used an L₂₇(3¹³) orthogonal array. From the results of the data analysis, the optimal factor-level combinations for the best path capability and minimum process variation were determined. The experimental runs were verified. The recommended methodology has the advantages of cost effectiveness, efficiency, data compression, and flexibility.