On Uncertainty And Evaluating Domain Of Attraction With Saturation Of Robotic Control

The control problems of robotic manipulator have received great attention in theoretical and engineering for many years. It is well known that the robotic manipulator is a very complicated MIMO nonlinear system with time-varying, strong-coupling and nonlinear dynamic characteristics, so the control for such a system is quite difficult. Motion control in joint space is a basic problem in robot control where the goal is to make the manipulator's joint track a desired trajectory. When the robot model is exactly known, the technique of feedback linearization in nonlinear system can solve the problem very well.However, the system can be influenced by uncertainties such as payload change and the change in parameters of dynamic model of robot when it goes about its task. Meanwhile the system may be also subject to the saturating nonlinear actuators. All these factors probably result in deterioration of control system performance, even instability. Therefore, good self-adaptability and the ability to deal with saturation in actuator are required in robot control system.So far, a great number of control systems are available in robot control. However, most controllers are designed based on arbitrary large torque of the actuator. In fact actuator has its physical limits and when the input exceeds the limit the actuator suffers from saturation so that in the practical application, good performances of the controllers can not always be guaranteed and even sometimes it results in control failure. Therefore, the controller should be designed on the basis of considering the actuator saturation, and finding such a controller that supports a domain of attraction as large as possible have great theoretical and practical importance.Based on multiple model adaptive control method, this paper presents a new algorithm for calculating the control-weight aiming at the problem that great change in parameters always leads large transient error. Moreover, simulation has been done on a two-link robot and the results show that new algorithm increases tracking accuracy comparing with the conventional approach. At last, this paper considers a robotic system subject to actuator saturation, and proposes a method to evaluate domain of attraction of the system. For the control method combing composite nonlinear feedback and computed-torque control, an iterative Ricatti equation algorithm is given to design such a linear gain that leads to a domain of attraction as large as possible. And simulation results show the effectiveness of the presented approach to decrease the conservative nature greatly.