摘要
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。
A fatigue driving model was established to investigate the relationship of driver fatigue with driving action and driving dynamics parameters.The target of this model provides a theoretical basis for early warning study of fatigue driving.The model includes parameters of perception,decision-making and driving operations.The preview follow theory is used to obtain the weight analytical solutions of preview optimized artificial neural network and proved the weight relate to the vehicle dynamic parameter,but has nothing to do with driving path.The sigmoid function was used to deduce the quantitative relation of the reaction time of nervous and motion and the degree fatigue by taking account the characteristics of the human body fatigue.Driving fatigue process of a driver in double-lane and vertical cross lane was simulated by using the proposed model.The simulated results qualitatively coincide with the reported experimental observations,verifying the feasibility of the method.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2011年第7期1481-1485,共5页
Journal of System Simulation
基金
国家自然科学基金(51005176)
关键词
驾驶员
疲劳孕育
驾驶员—汽车—道路闭环系统
预瞄神经网络
driver
fatigue germinating
driver-vehicle-road closed-loop system
preview optimized artificial neural network
