L.C. Kwek, Alan W.C. Tan, and E.K. Wong
Generalized predictive control, iterative learning control, real-time feedback control, robotic manipulator, disturbance compensation
The paper presents an improved generalized predictive control (GPC) scheme integrated with a disturbance compensation scheme based on iterative learning control (ILC) and real-time feedback control (RFC). The proposed GPC–ILC–RFC method is simulated on a two-link rigid robotic manipulator system that performs pick-and-place task repeatedly. A discrete-time model of the robotic manipulator is used to predict the system output over a prediction horizon such that optimal control inputs that minimize the position and velocity trajectory errors can be determined. The main contribution of this paper is the inclusion of the current cycle feedback error information into the disturbance estimations. This error information provides the controller the capability of making more immediate corrections with respect to disturbances. Simulation results reveal that the proposed GPC–ILC–RFC scheme outperforms the existing repetitive GPC controller under comparison and provides satisfactory tracking despite the presence of periodic and non-periodic disturbances.
Important Links:
Go Back
