基于改进PSO算法的四旋翼飞行器飞控系统PID参数整定

PID tuning for flight control system of quad-rotor aircraft based on improved particle swarm optimization

针对四旋翼飞行器飞控系统中存在PID控制器参数难以整定的问题,提出一种改进的粒子群算法,应用于PID参数的整定优化中.为了让粒子群在算法早期拥有较强的全局搜索能力,在算法后期拥有较强的局部开发能力和较快的收敛速度,该改进算法采用了一种可使惯性权重非线性下降的调整策略;同时,算法融合了遗传算子,进一步加快了收敛速度,避免算法陷入局部最优.将该算法应用于PID控制器的参数优化,以实数编码的形式直接生成与PID参数组对应的粒子群,并把控制系统的误差性能指标作为评价粒子群的适应度函数.通过与标准粒子群算法与手动调参的阶跃响应对比分析,发现改进算法其阶跃响应曲线超调量更小,调节时间更短,响应速度更快,动态性能更优.提出的改进算法能对四旋翼飞行器飞控系统中的PID参数进行较好的优化,实现更好的控制效果,使得飞行器在飞行过程中更加平稳.

Aimed at the difficulty of PID tuning for flight control system of quad-rotor aircraft,an improved particle swarm optimization algorithm is proposed to optimize the PID parameters.In order to make the particle swarm have strong global search ability in the early stage,as well as have strong local development capacity and fast convergence rate in the later stage,this improved algorithm adopts an adjustment strategy that makes the inertia weight nonlinear decline;at the same time,the genetic operator is fused to this algorithm,which makes the algorithm can speed up convergence rate and avoid algorithm from local optimum.Applying the algorithm to the parameter optimization of PID controller,the particle swarm corresponding to the PID parameter group is generated directly by real number encoding.At the same time,an error performance index of the control system is taken as adaptation function for evaluating particle swarm.By comparing with standard particle swarm optimization algorithm and manual adjustment parameters,it is found that the improved algorithm proposed in this paper has smaller overshoot,shorter transient time,faster response speed and better dynamic performance in step response curve.The improved algorithm proposed in this paper can optimize PID parameters of flight control system of quad-rotor aircraft much better and achieve better control effect,which makes the aircraft more stable during flight process.