基于PET算法的匝道合流区交通冲突识别模型

为了识别入口匝道汇入车辆与主线直行车辆间的交通冲突，开展了匝道合流区车辆交通冲突识别研究.本文结合车辆运动信息，考虑车辆尺寸对交通冲突的影响，构建了基于后侵入时间(Post Encroachment Time, PET)算法的匝道合流区冲突识别模型；给出车辆交通冲突严重程度的确定方法，采用仿真分析验证了所建模型的有效性.结合实测交通数据，确定了PET阈值范围.结果表明，采用后侵入时间算法的匝道合流区交通冲突识别准确率为91.71%，说明该模型能有效识别匝道合流区的潜在冲突.研究成果可为车路协同环境下匝道合流车辆提供安全预警，进而减少车辆碰撞事故的发生，提升整个交织区域的道路交通安全水平.

Traffic Conflict Identification Model Based on Post Encroachment Time Algorithm in Ramp Merging Area

In order to identify the traffic conflicts between the entrance ramp and the main line, the study of vehicle traffic conflict identification in ramp merging area is carried out. We built a traffic conflict identification model based on PET (Post Encroachment Time) algorithm in ramp merging area, considering vehicle movement information and the influence of vehicle size on traffic conflict and gave the method for determining the severity of traffic conflict. The validity of the model is verified by simulation analysis, and the PET threshold range is confirmed by the measured data. The results show that the accuracy rate using PET algorithm in ramp merging area is 91.71%, which can effectively identify the potential conflicts in the ramp zone. The results can provide a safe early warning for confluence vehicle under collaborative environment, reduce accidence of the vehicle collision and raise the level of road traffic safety in the mixed zone.