POSITION CONTROL OF MECHANICAL MANIPULATORS BY USING NONLINEAR ADAPTIVE CONTROL ALGORITHM (FLEXIBLE, BEAMS).

摘要

The manipulator is a complicated nonlinear coupling system whose dynamics is influenced by the coriolis effect and the gravity effect of the configuration change. Conventional robot control methods which are designed based on linear systems are only applicable to slow moving condition, because at higher speeds the nonlinear effects will deteriorate the performance of the control system. This is one of the industrial robots limitations, furthermore the payload capacity of industrial robots are very small compare to its arms weight. With growing demand on manipulator performance there is an increasing need for improved dynamic modelling and control technique of manipulators.; In the present work two nonlinear adaptive control algorithms are developed for manipulator systems to solve these two problems. The first scheme is designed for the rigid arm robot systems (deterministic), and the second one is designed for the flexible beam manipulator systems (stochastic). These algorithms directly use nonlinear dynamic model in controller design to account for the nonlinear effects of the system. The least-square time varying parameter identification scheme has been used to identify the changes in configuration and payload of the system.; The dynamic model retains all the nonlinear forces and dynamic coupling of the interlink. The assumed modes method is used to describe the flexible motion of the link, and a finite number of vibration modes are chosen independently for each link in deriving the dynamic equations. However, the controller is not designed based on finite number of modes, so this nonlinear control algorithm is not limited the distributed system with finite number of modes.; Under bounded input and output considerations, the global stability of the system can be proven. It is also demonstrated that the system output tracking error converges to zero and the solutions of controller parameters are not unique. A number of numerical examples have been worked out, and the results shows that this algorithm is robust and easy to apply in real-time control systems.