Research And Simulation Of Autopilot Based On Particle Swarm Optimization And Fuzzy Control

Ship autopilots, which used to keep the ships navigating in the set directions, are indispensable devices in the ship control systems. The demand of autopilot control precision is increasing as the increasing of navigation safety and economical demand.Ship maneuvering system is the research basis of ship autopilot control, and also the precondition of ship maneuver control. Based on the analysis of the ship maneuver model and the disturbance model, reduction mathematical model of the ship autopilot control system is constructed, also disturbance models of wind, wave and flow are constructed.Using ship autopilot mathematical model as the research object, PID controller and fuzzy PID controller are designed. Adjustment method of the PID controller is given, and the performance of the controller is analyzed in simulation. Fuzzy PID controller is designed based on the PID controller. Navigation error and error change are used as the fuzzy model inputs, and autopilot angle is used as the controller output. Controller performance is improved by adjusting the input and output regions, subjection functions and fuzzy control rules. Simulation results of"Yulong ship"indicate that the efficiency of the controller.Fundamental theory and mathematical description of particle swarm optimization algorithm is introduced. Four classic test functions are introduced and 3-Dimension figures of the functions are given. Particle swarm optimization algorithm is tested by the four functions, and simulation results show that the PSO algorithm has the global searching ability.Based on the designed fuzzy PID controller, parameter optimization of the controller using the particle swarm optimization algorithm is developed. Flow chart of the PSO based parameter optimization is given. Performance of the optimized fuzzy PID controller is improved, and overshoot is decreased, and this indicates that the optimized fuzzy PID controller is over performance the conventional PID controller and experience based fuzzy PID controller. Simulation results show that the particle swarm optimization based fuzzy PID controller can improve the controller performance, and the precise control of autopilot is achieved.