Global versus Local Optimization in Redundancy Resolution of Robotic Manipulators

Abstract

In redundancy resolution of robotic manipulators with more than the required degrees of freedom at the kinematic level, the minimization of the Euclidean norm of the joint velocities at the dynamic level, and the minimization of the kinetic energy of the manipulator at any moment of the task execu tion have long been suggested in the literature. In this paper the global optimization of the above characteristics are worked out by using integral-type criteria. The results are in surprisingly simple and elegant forms and give new insights to the redundancy resolution of robotic manipulators. The relations between local and global optimization are discussed.