An efficient technique for finding the desired global optimum of robotic joint displacement

For an industrial robot on a daily operation basis such as pick and place, it is desired to minimize the robotic joint displacements when moving the robot from one location to another. The objective of the optimization here is to simultaneously minimize a robot end effector's positional error and the robotic joint displacements. By modifying the searching algorithm in the existing complex optimization method, this article presents a technique for finding the desired global optimum solution more efficiently. To compare the optimum searching capability between the proposed and existing searching algorithms, a modified Himmelblau's function is used as an objective function. The presented technique is then applied to a spatial three-link robot manipulator for global minimization of the joint displacements. © 1992 John Wiley & Sons, Inc.