Researchand Implementation For Missile Autopilot

Missile motion has characteristics of model inaccuracy, nonlinearity and time-varying.The traditional control method can not meet the demand for missile flight control. Using themethod of intelligent control into missile autopilot is an effective way to solve the controlproblem of complex objects.This paper designs the controller of the missile pitch autopilot via selecting a featurepoint on a trajectory based on frozen-coefficient method. Firstly, it combines the fuzzy controlwith PID control into fuzzy PID controller (FC-PID). The purpose is to realize the self-tuningon PID parameters. Then it combines the genetic algorithm (GA) with simulated annealing(SA) into simulated annealing genetic algorithm (GASA). SA becomes part of GA in order toavoid local optimal solution. Then it uses the GASA to optimize membership functionparameters for fuzzy PID controller (FC-PID) to design a simulated annealing genetic fuzzyPID controller (GASAFC-PID).Matlab simulation results of PID controller, FC-PID controller, GAFC-PID controllerand GASAFC-PID controller show that the ability of searching of GASA is better than GAand GASAFC-PID controller has good dynamic performance, robustness and goodperformance of every point on the trajectory. The result of semi-physical simulation showsthat the controller can meet the requirement of real-time by using fuzzy lookup table and has agood engineering application prospect.