传统导弹的PID控制律设计仅考虑了幅值裕度、相位裕度等指标,无法满足多变量鲁棒性指标要求。文中提出了一种基于约束粒子群优化的导弹控制器设计方法,该方法以H_∞鲁棒性指标作为目标函数,自动优化得到控制参数。同时,对现有的粒子群算法进行了改进,采用映射函数法将约束优化问题转化为非约束优化问题。以样例导弹为对象设计了H_∞PID控制器并与罚函数法进行了比较,结果表明文中算法各项性能优于经典罚函数法。

Traditional design of missile's PID control law only considers robust performances such as amplitude margin, phase margin and so on, which is unable to guarantee multivariate robustness index.In this paper, an easy-to-use method for controller design was presented based on constrained particle swarm optimization (PSO) algorithm.This method takes H_∞ robustness index as objective function and automatically obtains various parameters.The existing PSO algorithm has been improved by a mapping function, so a constrained optimization problem can be converted into an unconstrained one.Taking the sample missile as the plant, the H_∞ PID controller was designed.Compared with penalty function algorithm, simulation results show that this method has better performance.