The Comparison of Two Workspace-Density-Driven Inverse Kinematics Methods for Hyper Redundant Manipulators

Abstract

Motivated by the cascading of workspace densities, we have developed two new workspace-density-driven inverse kinematics methods for hyper-redundant manipulators in recent papers. One is based on the breadth-first search and the other is based on the divide-and-conquer search. Both methods do not involve a high dimensional Jacobian matrix and have their own favorable qualities. In this paper, we perform a detailed numerical comparison of these two methods. We focus on comparing the computation time, end-effector error and data storage requirements. This comparison will give users of these kinds of robots guidance as to when which method should be used.