基于神经网络的应急疏散状态下车辆跟驰模型  

Car Following Model Under Emergency Evacuation Situation Based on the BP Artificial Neural Network

作　　者：徐志[1] 杨孝宽[1] 赵晓华[1] 李玲洁[1] 

机构地区：[1]北京工业大学建筑工程学院,北京100124

出　　处：《北京工业大学学报》2011年第6期882-887,共6页Journal of Beijing University of Technology

基　　金：国家自然科学基金项目资助(50578003);国家科技支撑计划资助项目(2006BAG01A01)

摘　　要：利用情景模拟方法为被试营造了应急疏散的驾驶环境,以主观量表、心电仪等实验心理学手段从主观和客观2方面对所营造的应急疏散驾驶环境的有效性进行了验证.对驾驶模拟舱采集的数据使用灰色综合关联度进行关联分析,以后车加速度为参考序列,以被试车车速、前车车速、两车速度差、车头间距、前车加速度为比较序列计算其关联系数,最终选择前车加速度、两车速度差、前车车速指标作为神经网络的输入向量.设计BP神经网络的结构,网络输出为被试车加速度.利用实测数据对BP神经网络进行训练、仿真实验,结果表明模型仿真效果良好.Scenario simulation is employed to create a driving environment in the emergency evacuation and the questionnaire investigation and the Miriam ECG are used to verify the validity of the driving environment in the emergency evacuation from subjective and objective aspects.Gray-correlation Analysis is undertaken to determine which factors have more impact on the acceleration of the following car.Acceleration of the following car is set as the mother factor series and the speed of the following car,speed of the front car,speed difference between vehicles,headway between vehicles,acceleration of the front car are set as the sub-factor series.Finally acceleration of the front car,speed difference between vehicles,speed of the following car are determined as three most important factors.BP neural network is designed and the three factors mentioned above are selected as input variables and the acceleration of the following car is selected as output variable.Simulation of the BP neural network with data gathered from driving simulator reveals that BP neural network has a high precision in the prediction of the car-following model.

关 键 词：跟驰模型 应急疏散 灰关联度 BP神经网络 

分 类 号：U491[交通运输工程—交通运输规划与管理]