Prioritized Closed-Loop Inverse Kinematic Algorithms for Redundant Robotic Systems with Actuator Velocity Saturations

Standard kinematics prioritized task based motion control solutions do not take into account the physical limitations in terms of maximum actuator speed of robots. In this paper, a prioritized task based kinematics control solution is presented that, under given conditions on the kind of concurrent tasks to be pursued, guarantees task error stability and convergence. Moreover the joint velocities are guaranteed to be bounded by a desired threshold. As for other a null-space projection techniques known in the literature, joint speed commands are computed in such a way that lower priority tasks do not interfere with higher priority ones in the assumption that joint speeds can be arbitrarily large: in addition, if joint speeds are to be bounded by a desired value, joint velocity commands are limited by dynamically chosen values depending on the task priority. As a result, joint velocities are always bounded such that, if necessary, higher priority tasks are executed first.