An Approach to the Control of Kinematically Redundant Robot Manipulators

The study of redundant manipulators presents a number of challenges. In particular, it is of interest to investigate how best to take advantage of the increased number of joint space trajectories solving the inverse kinematics problem which are made available by the redundant design. Here the inverse kinematics problem is investigated for those robotic manipulators possessing redundancy. An algorithm is described based on the generalized inverse approach, allowing generation of trajectories performing some desired subtask in addition to prespecified end effector motion. The subtask, or subtasks, are represented by a condition on a scalar cost function. Several possible applications are illustrated by simulation results.