PID结构广义预测控制算法的仿真及分析

Simulation and Analysis of PID Generalized Predictive Control

为了使智能车转弯平滑,提出了一种具有PID结构的广义预测控制算法.利用PID控制与广义预测控制规律的相似性,通过改变一般预测控制中的目标函数,确立了广义预测控制算法的递推关系.并对其进行了仿真,结果表明:舵机通过路径检测传感器采集到当前路径信息做出转弯响应,可以预测出下一时刻的转向控制信息.在干扰信号的作用下,对于不同的路径信息,目标函数中增量u的输出曲线稳定,时间明显缩短,在7s后出现周期性变化,使得舵机的预测输出轨迹接近给定输入轨迹,实现了对PID参数自动整定和优化.

For the smooth turning of the intelligent vehicle,a PID generalized predictive control algorithm is proposed.With the similarity of PID control and generalized predictive control,the objective function of the general predictive control is changed,the recursive relation of the generalized predictive control is established.The simulation illustrates：The current path information is collected by the path detection sensor,the next time steering control information is predicted,then steering mechanism responds to the turning.For different path information with interference signals,the curve of increment output u of objective function is smooth and changes cyclically in 7 seconds.The respose time is reduced obviously.The output forecast track of steering closes with input given tracks to realize the automatic PID parameters setting and optimization.