Controller Designs for Dynamic Positioning System using Four Thrusters

This work proposes a control system designs for dynamic positioning system (DPS) with four azimuth thruster using a linear and nonlinear control, that being compared, both with feedforward control. The linear controls are LQ optimal control and PD control while the nonlinear control is the Sliding Mode. The controller is design based on the vessel model and then the controller is simulated on Simulink and tested in a real experiment with the expected result that sliding mode control perform better than the linear one. The optimation control allocations use a lagrangian multiplier technique. Experimental control testing use a real DPS where all the subsystems are connected by mosquito MQTT and the system is supported by high accuracy of DGPS. Based on the experiment, sliding mode with wind feedforward control gives the best result with the RMS error are smaller than all the other controllers.