基于RBF神经网络的高速公路防追尾模型研究

Study on Rear-end Avoidance Model of Highway Based on RBF Neural Network

为提高高速公路车辆跟车安全,在常用安全距离算法基础上,结合汽车制动过程分析,提出安全度判别准则的安全控制方法,建立RBF神经网络防追尾模型,讨论RBF网络设计过程中的参数设置,并运用m atlab编程进行训练和仿真,仿真结果表明RBF模型是有效的;同时,模型为高速公路汽车追尾建模提供一种新思路,为汽车预警设计提供理论基础。

In order to improve the car-following security on the highway,a kind of safety control method based on security criterion was proposed combining with the common safety distance algorithm and the analysis of vehicle braking process.The rear-end avoidance model of RBF neural network was set up,then the parameter setting in designing process was discussed and the whole model was trained and simulated by matlab programming.The simulation results indicated the effectiveness of the RBF model.Meanwhile,it provided a new idea for the modeling of vehicle rear-end on highway and afforded theoretical basis for pre-warning design of automobile.