LAUSR

Conventional dynamic positioning systems are based on PID controllers and an extended Kalman filter or a nonlinear state observer. However, it is nontrivial to tune the control parameters, and the station-keeping performance varies with environmental or loading conditions since the dynamics of the vessel are essentially nonlinear. To overcome these difficulties, a fuzzy rule-based PID controller is evaluated, which takes the estimated positioning error and low-frequency velocity as inputs, and outputs the time-varying PD control coefficients through fuzzy inference, while the integral control parameters are kept constant. The performance of the proposed controller is evaluated numerically through a time domain simulation of a dynamically positioned semi-submersible platform operating in variable environmental disturbances. Simulation results are compared with the conventional fixed gain PID controller, and the comparison results show that the proposed fuzzy PID controller can automatically tune the PD control coefficients according to the positioning accuracy and significantly improve the performance of the dynamic positioning system.