论文部分内容阅读
为了揭示驾驶动作、车辆行驶状态与驾驶员疲劳孕育过程之间的联系,为驾驶员疲劳早期预警提供理论依据,构建了一套涵盖驾驶员疲劳孕育过程的驾驶员模型。采用预瞄—跟随理论推导出预瞄神经网络权值的解析表达,得出神经网络权值仅与车辆本身的动力学参数有关而与驾驶路径无关的结论。利用人体疲劳特征符合Sigmoid函数的高低增益区的数学特性推导出了驾驶员神经反应时间、动作反应时间与疲劳度之间的定量关系。以双移线、垂直弯道两种路况为例,对疲劳孕育过程中驾驶员操作动作及车辆行驶状态变化进行仿真,结果显示与已有的试验研究结论相符,从而证明提出模型的正确性。
In order to reveal the connection between driving motion, vehicle driving status and driver fatigue during pregnancy, a theoretical basis for driver fatigue early warning is provided, and a driver model covering driver fatigue incubation process is constructed. The pre-sight-following theory is used to derive the analytic expression of the weight of the pre-cursor neural network. The conclusion is obtained that the weight of the neural network is only related to the dynamic parameters of the vehicle and not to the driving route. The quantitative relationship between driver reaction time and fatigue degree was deduced based on the mathematical characteristics of the human body fatigue characteristics in accordance with the high and low gain regions of Sigmoid function. Taking the two kinds of road with double shift line and vertical curve as an example, the simulation of driver’s operation and vehicle’s driving status during the process of fatigue inoculation is carried out. The result is in good agreement with the existing experimental results, which proves the correctness of the proposed model.