On-line, point-to-point near minimum-time feedback control of robot manipulators

Solving minimum-time point-to-point motion control problems, for robot manipulators with actuator bounds, has been an important research area in the robotics literature. Execution-time of specific tasks and productivity are directly related for industrial robots, and similarly size and energy requirements are critical for space robots. Finding minimum-time strategies, however, even when the dynamic equations and parameters are precisely known, is understood to be difficult and computationally intensive, so that minimum-time trajectory planning is typically performed off-line. In this paper, a simple real-time feedback control scheme is proposed, which is effective in achieving point-to-point tasks in near minimum-time, using a so-called 'two-step' Lyapunov approach. The performance of the proposed method is simulated and compared with true optimal solutions.