基于动态行车安全间距的交织区换道模型

Lane change model for weaving area based on dynamic safety spacing

为模拟城市快速路交织区复杂换道决策行为,提升交织区运行安全性,基于微观车辆轨迹数据,构建考虑动态行车安全间距的交织区元胞自动机换道模型。首先,利用高空视频提取高峰时段交织区车辆换道间距信息,提出换道间距与换道次数关系式;然后,根据交织区不同车道换道差异,分别构建主线车道、集散车道换道行为模型;最后,采用流量、密度、换道类型等参数对模型进行评价。结果表明,相比经典换道概率分布模型,模型可更有效模拟交织区瓶颈现象和运行风险,反映真实换道需求,描述实际运行状态。

To simulate the complex lane-changing decision behavior in the weaving area of urban expressways and improve the operational safety in the interweaving zone,this paper constructs a meta-cellular automaton lane-changing model in the weaving area considering dynamic travel safety spacing based on microscopic vehicle trajectory data.First,more than 1000 highresolution vehicle trajectories with a time accuracy of 0.1 s and a spatial accuracy of 0.1 m/pixel are obtained by using the highaltitude video of the interweaving area captured by drones,including speed,acceleration,vehicle distance,density,and vehicle lane-changing position coordinates and directions,etc.By analyzing the information on the lane-changing distance of vehicles in the weaving area during peak hours,the relationship between the lane-changing distance and the number of lanechanging times is obtained;Then,according to the lanechanging differences of different lanes in the weaving area,taking into account the lane-changing demand,dynamic driving safety distance,parking safety distance,and other factors,the lane-changing behavior models of the mainline lane and the distribution lane are constructed,respectively,and the simulation analysis is carried out.The results are also compared with the classic lane-changing conflict handling multi-strategy model;Finally,the model is evaluated using typical parameters such as flow rate,density,and lane-changing type.The model validation results show that:compared with the measured data in the interweaving area,the errors of the flow,density,and velocity of the model in this paper are all in the range of 0.7%1.6%.The average error of lane changing times is 10.11%.The spatiotemporal characteristics and types of lane changing are also consistent with the actual situation.Compared with the classic lane-changing probability distribution model,the model in this paper can better simulate the bottleneck phenomenon and operational risk in the weaving area.This shows that the model in this paper can effectively simulate the complex lane-changing decision-making behavior of vehicles in the weaving area,reflecting the real lane-changing demand and describing the actual operating state.