PSO-based robust PID control for flexible manipulator systems

Abstract

In this paper, a new control law is designed for flexible manipulator systems. Kharitonov theorem is used as a design tool to derive the robust control in robotic systems. Stability is guaranteed not only at one operating condition but also for a wide range of system uncertainty. Both the non-linear behaviour of the gearbox stiffness and the end load variations are tackled as interval uncertainties in the model. Particle swarm optimisation (PSO) is used to tune controller parameters, such that, the greatest real parts of closed loop eigenvalues among Kharitonov-extreme polynomials is minimised to assure the highest possible relative stability. The flexible manipulator is a realistic industrial benchmark robot manipulator. The proposed controller is tested as well for rejecting disturbances injected at the motor and the tool. The robustness performance of the controller is evaluated in terms of reference tool position tracking in the presence of the mentioned disturbances and uncertainties. The results of the simulation are presented and compared to the results of traditional controllers and very promising outcomes are obtained.