Improving minimum-time task-space path-following algorithms for redundant manipulators

This paper deals with the minimum-time task-space path following problem for redundant manipulators. Although there exist a number of efficient algorithms for solving this problem for non-redundant manipulators, the question of how to exploit kinematic redundancy to reduce the execution time is still an open problem. The problem investigated here is what to choose between different solutions to the inverse kinematics problem for redundant manipulators. The aim is to obtain a joint-space path which configures the manipulator so as to improve acceleration/deceleration capabilities of the end-effector. Numerical results are presented for an "easy-to-understand" three degree-of-freedom planar manipulator executing in 2D paths.