采用改进粒子群算法与人工神经网络相结合的车辆转向控制研究  

Vehicle steering control based on improved particle swarm optimization and artificial neural network

作　　者：姚俊[1] 张劲恒[1] YAO Jun;ZHANG Jinheng(Institute of Electronics and Information,Nanchang University of Technology,Nanchang 330044,China)

机构地区：[1]南昌理工学院电子与信息学院,南昌330044

出　　处：《中国工程机械学报》2018年第6期480-485,共6页Chinese Journal of Construction Machinery

基　　金：湖南省自然科学基金资助项目(2016JJ0025)

摘　　要：为了提高车辆转向控制系统输出精度,改善车辆行驶的稳定性,提出了改进人工神经网络PID控制器.创建车辆平面参考模型简图,建立车辆运动参数的数学关系式,推导出车辆横摆角速度的动力学方程式.在传统PID控制器基础上,结合人工神经网络模型,采用改进粒子群算法对人工神经网络PID控制器进行在线优化,动态调整PID控制器参数,实现车辆转向控制系统的最优输出,在不同工况路面进行车辆横摆角速度仿真实验.结果表明:采用改进人工神经网络PID控制器,不仅可以提高车辆转向控制系统的响应速度,而且输出的摆动角速度误差较小.车辆在复杂工况路面行驶,其转向系统采用改进人工神经网络PID控制器,有利于提高车辆行驶的稳定性.In order to improve the output accuracy of vehicle steering control system and improve the stability of vehicle driving,the improved artificial neural network PID controller is proposed.The vehicle plane reference model diagram is created,the mathematical formula of vehicle motion parameters is established,and the dynamic equation of vehicle yaw rate is derived.On the basis of the artificial neural network model based on the traditional PID controller,the improved particle swarm optimization is used to optimize the PID controller online,dynamically adjust the parameters of the PID controller,realize the optimal output of the vehicle steering control system,and simulate the yaw velocity simulation of the vehicle in different working conditions.The results show that the improved artificial neural network PID controller can not only improve the response speed of the vehicle steering control system,but also have a small error in the output swing angle speed.Vehicle running on complex road surface,its steering system uses improved artificial neural network PID controller,which is conducive to improving the stability of vehicle driving.

关 键 词：改进粒子群算法 车辆 转向控制系统 人工神经网络 PID控制器 

分 类 号：U461[机械工程—车辆工程]