驾驶员制动意图辨识的方法研究

Research on Driver Braking Intent Identification

采用基于模糊推理的多参数驾驶员制动意图识别的方法 ,以制动踏板力、制动踏板开度以及制动踏板开度变化率作为输入参数,将驾驶意图分为平稳制动、一般制动、紧急制动;通过Matlab自带的模糊推理工具箱,采用图形界面可视化工具建立驾驶意图识别的模糊推理系统。最后借助Matlab的Fuzzy工具箱,搭建驾驶员制动意图辨识模型的离线验证Simulink程序,利用大量实车的试验数据验证辨识模型的准确性。仿真结果证明该模型辨识度高,模型准确可靠。本模型也可用于制动辅助系统的控制策略的开发,从而提高制动辅助系统的响应时间,减轻驾驶员的负担,提高汽车的行驶安全性。

Using driver＇s braking intention recognition method based on fuzzy reasoning, with the brake pedal force, brake pedal displacement and brake pedal displacement rate as input parameters, driving intention can be divided into smooth braking, general braking and emergency braking; And through fuzzy reasoning toolbox of Matlab, using graphical interface visualization tools of driver intention recognition, fuzzy inference system is established. Finally, the offline validation Simulink program of driver braking intention recognition model is established with the Matlab Fuzzy toolbox, and then accuracy of the identification model is verified by using a large number of real vehicle test data. The simulation results show that the Simulink model has a high recognition rate, which demonstrates reliably. The model can also be used to develop control strategy of brake assist system to improve response time of the braking system, which will reduce the burden of drivers and improve driving safety of cars.