Acceleration-level analysis on velocity-based motion control of kinematically redundant manipulators

Acceleration-level control of kinematically redundant manipulators has its peculiar problems such as the instability of the local torque optimization and the end-motion joint velocities. A large number of studies on the problems have been carried out and one of the most promising solutions is the acceleration-level control based on a velocity-level method which never causes the problems. In the conventional approach, the joint accelerations to accomplish a task is derived from the acceleration-level kinematic equation, whereas the velocity-based control uses the joint accelerations obtained by time differentiation of the solution to the velocity-level kinematic equation. In this paper, control methods based on the task-priority concept which accomplish primary and secondary tasks are investigated at the acceleration level. The velocity-level solution is analytically differentiated by time and compared with the simple acceleration-level solution, which clarifies the difference between the velocity-based control and the simple acceleration-level control