On the dynamic characteristics and control of balanced robotic manipulators

Abstract

Highly nonlinear and coupled characteristics of the manipulator dynamics have been one of the principal reasons in reducing path tracking accuracy and complicating controller design. This thesis presents the salient features of the balanced robot dynamics characterized by partially configuration-independent inertial properties and the absence of gravity loadings. The analysis was performed for a 3-DOF ideal robot model which has no offset distances between links. The analysis includes the effects of payload variations on the tracking accuracy and input torques of the balanced robot and was made via sensitivity theory. These simulation results were compared to those of the unbalanced robot. In addition, the effects of balancing masses on the dynamics of the balanced robot were evaluated for various manipulator speeds and work trajectories through simulation study. Based upon the above analyses, an efficient control algorithm is suggested, which is suitable for the balanced robot. To evaluate performance of the control algorithm simulation studies were made over a wide range of manipulator speeds and payloads conditions. The results show that the tracking accuracy obtained via this simple algorithm can be favorably compared with those obtained by other sophisticated controllers. Since the analytical results show that the balanced manipulator posseses many favorable effects on the dynamic characteristics and the controller simplification, the balancing mechanism was applied to the conventional 6-DOF robot which includes link offset distances and wrist motions. The dynamic characteristics of the balanced robot obtained for the case of the ideal robot were verified both analytically and experimentally using the PUMA-760 robot for various payload and speed conditions. The major result is that the balancing mechanism can enlarge the payload capacity of the conventional robot, increasing payload capacity more than two times than that of the unbalanced robot.