摘要
为了应对高速公路入口合流区域并线事故频繁发生,在分析车辆并线行为影响因素的基础上,利用BP神经网络方法建立了车辆在该区域的决策模型,用来预测驾驶人的并线行为决策,保障车辆和驾驶人的安全。借助详细的车辆轨迹数据对模型进行了学习与测试,模型的测试结果表明:建立的BP神经网络模型用于预测驾驶人并线行为具有较高的准确度,并线车辆相对于目标车道前方车辆的相对速度是驾驶人并线时需要考虑的最重要因素。同时模型还可以进一步应用于交通仿真的研究以及驾驶人辅助系统的开发。
In order to response to the frequent accidents in on-ramp merging area of expressway,based on analyzing the influence of merging behavior of vehicles,the BP neural network is used to establish a decisionmaking model of vehicle in the area. The model is used to predict the decision-making of driver's merging behavior to guarantee the safety of vehicles and drivers. With detailed vehicle trajectory data,the model learning and testing are conducted. The model test result shows that the proposed BP neural network model has a high accuracy for predicting driver's merging behavior,and the relative speed between the merging vehicle and the front vehicle in the target lane is the most important factor that the driver must be considered for merging. The proposed model can be further applied in traffic simulation and driver assistance systems.
出处
《公路交通科技》
CAS
CSCD
北大核心
2014年第9期120-123,153,共5页
Journal of Highway and Transportation Research and Development
基金
教育部博士点基金资助项目(20113227110014)
江苏省大学生创新训练项目(201410299005Z)
关键词
交通工程
换道模型
BP神经网络
高速公路入口合流区域
automobile engineering
lane change model
BP neural network
expressway on-ramp merging area