A geometric structure based self-motion control algorithm for redundant manipulators

Abstract

This paper presents a real-time self-motion planning algorithm for redundant manipulators. In most situations, the self-motion pattern of a redundant manipulator is restricted by the geometric structure of it. So it is reasonable and feasible to plan the self-motion of the manipulator on the analysis of the geometric structure. In our work, according to a kind of redundant manipulators, which have one extra degree of freedom (DOF) and a special self-motion pattern, a linear constraint based minimal least square algorithm is developed to obtain a vector based self-motion planning. In this algorithm, the motion planning of the end-effector is achieved by a minimal least square-based scheme. The principle that the motion of the end-effector should not be influenced by the self-motion of the manipulator is treated as a set of constraint conditions. A unified expression for self-motion planning is obtained by such a mathematic derivation. This algorithm is suitable for some applications like real-time obstacle avoiding or dexterous manipulating. Numerical simulation is also presented in the paper to verify the algorithm.