摘要
为了减弱事故对上游路段的影响,预防潜在二次事故的发生,提出了适用于智能网联自动驾驶车辆(Connected and Automated Vehicle,CAV)与传统人工车(Human Vehicle,HV)混合存在场景下的CAV排列整合算法(CAV Arrangement Integration Algorithm,CAV-AIA)。由于在相同的CAV数量下,不同CAV间的排列关系(自动车队强度)对道路通行能力有一定影响,因此CAV-AIA以“就近”增大自动车队强度为目标,对事故路段上游的CAV进行管控。具体地,在横向车道方面,以提升自动车队强度为目标,考虑发生事故的车道位置,给出不同车道CAV合流到相邻车道自动车队中的次序;在纵向位置方面,以最小化自动车队中CAV间的距离为目标,通过考虑车辆安全、路段速度限制和车辆速度改变率等因素,对自动车队间的CAV进行纵向位置更新;而对于未在自动车队中的车辆及每个自动车队中的引导车(第1辆车),为了减少其对HV运行的干扰,应用增强的智能驾驶人车辆跟驰模型(Enhanced Intelligent Driver Model,EIDM)及最小化换道引起的总制动(Minimizing Overall Braking Induced by Lane Changes,MOBIL)车辆换道模型对其车辆位置信息进行更新。数值仿真结果表明:在高、中、低交通需求及不同CAV比例下,实施CAV-AIA后,车辆通过事故区域上游管控路段的平均行程时间均有所降低,其中当交通需求为6500 veh·h^(-1),CAV比例为0.8时,通行时间提升率可达41.8%;而当交通需求为2500 veh·h^(-1)时,由于车流密度较小即使不施加额外控制策略,车辆仍能以较高的速度通行,因此不同CAV比例场景下车辆的通行时间提升率较低。最后,以HV平均换道次数作为指标进一步从微观层面上研究了CAV-AIA对HV移动的干扰。结果表明:CAV-AIA能够有效减少HV的换道次数,特别是在高交通需求下,如当CAV比例为0.8时,实施CAV-AIA后,HV的平均换道次数可由0.99降为0.53。
This study proposed a connected and automated(CAV)arrangement integration algorithm(CAV-AIA)to reduce the impact of traffic accidents on the upstream of freeway sections and prevent the occurrence of potential secondary accidents.It is suitable for both connected and automated vehicles(CAVs)and traditional human vehicles(HVs).With the same number of CAVs,different CAV arrangement relationships(CAV platoon intensity)had a specific impact on the freeway capacity.CAV-AIA aimed to increase the platoon intensity“in the vicinity”to control the CAVs upstream of the accident section.Specifically,with respect to lateral lanes,considering the location of the accident lane,the order of CAVs on different lanes to merge into the CAV platoons of adjacent lanes was given to enhance the platoon intensity.In terms of updating the longitudinal position,the objective was to minimize the distance between CAVs in the same CAV platoon,considering related factors,such as safety conditions,freeway speed limits,and vehicle speed variation rates to update the position of CAVs in the CAV platoons.For vehicles that were not in the CAV platoons and the leader vehicle(the first vehicle)in each CAV platoon,the enhanced intelligent driver model(EIDM)and minimizing overall braking induced by lane changes(MOBIL)model were applied to update the vehicle information and reduce the interference to HVs.Numerical simulation results indicate that under high,medium,and low traffic demand,among different CAV MPR,CAV-AIA reduces the average travel time for vehicles to pass through the upstream control area of the accident area.When the traffic demand is 6500 veh·h^(-1)and the CAV MPR is 0.8,the improvement rate can increase by 41.8%.When the traffic demand is 2500 veh·h^(-1),the average travel time improvement rate under different MPR is lower,which can be attributed to the low traffic density.Even without an additional control strategy,vehicles can travel at a higher speed.Lastly,this study further explored the interference of CAV-AIA on HVs with the indicator of HV average lane-changing times from a microscopic perspective.The results show that CAV-AIA can effectively reduce the lane changing times of HVs,particularly under high traffic demand;for instance,when the CAV MPR is 0.8,the average lane changing times can reduce from 0.99 to 0.53 following the implementation of CAV-AIA.
作者
曹丹妮
吴建军
屈云超
刘浩
CAO Dan-ni;WU Jian-jun;QU Yun-chao;LIU Hao(State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China;Beijing Transportation Information Center,Beijing 100161,China)
出处
《中国公路学报》
EI
CAS
CSCD
北大核心
2022年第3期78-88,共11页
China Journal of Highway and Transport
基金
国家自然科学基金项目(71621001,71890972,71890970)
中央高校基本科研业务费专项资金项目(2021PT206).
关键词
交通工程
管控策略
微观交通流仿真
混合交通流
traffic engineering
control measure
microscopic traffic simulation
mixed flow
