主题词：无信号交叉口 网联车辆 碰撞预警 驾驶人特性非合作博弈中图分类号：U471.15 文献标志码：A DOI： 10.19620/j.cnki.1000-3703.20240752

A Collision Warning Strategy for Connected Vehicles at Unsignalized IntersectionsConsideringDriverCharacteristics

WangRunmin1²，FengHao²，FanHaijin²，HeJiajun² （1.Collaborative InnovationCenterforWestern Traffc Safetyand IntellgentControlbyProvinceand Ministry，Chang'an University，Xi'an 71O064; 2.SchoolofInformation Engineering，Chang'anUniversity，Xi'an 710064）

【Abstract】Addressing the limitations of current intersection collsion warning systems，including non-line-of-sight isuesandlimitedconsiderationof drivers’characteristics，thispaperproposesacooperativecolisionwarning strategy for connectedvehiclesatintersections，incorporatingrivertraits.irstlyrivinghaviorsatinersetionsaeategordito straightandturning，andaturning speed modeltailoredtodrivercharacteristics is builtusingthe InDdataset.Secondly， vehicleturingtrajectorypredictionisenhancedwithaconstantyawratemodelandExtendedKalmanFilter，whilecolision risksaredynamicallassessedusingadual-circlevehiclegeometrymodelbasedonTimeExposedtoRisk.Thirdly，atwo-level warningstrategygroundedinnon-cooperativegametheoryisdevised，consideringdriverheterogeneityanddynamic interactions inunsignalizedconflicts.Finaly，thestrategyisvalidatedthroughsimulationsandreal-vehicletests.Results indicate the strategy successfully detected all colisions with a 100% warning rate，reduced collisions by up to 100% among diverse drivers，and decreased accidents by 95.06% and kinetic energy by 52.71% even with aggressive drivers.

Keywords：Unsignalized intersections，Connected vehicles，Collision warning，Drivers' characteristics，Non-cooperativegame

【引用格式】王润民，冯皓，凡海金，等.考虑驾驶人特性的无信号交叉口网联车辆碰撞预警策略[J].汽车技术，2025（7）：1-12. WANGRM，FENGH，FANHJ，etal.AColisionWarningStrategyforConnecedVehiclesatUnsignaliedIntersections ConsideringDriverCharacteristics[J].Automobile Technology，2O25（7）：1-12.

1前言

交叉口是交通网络的关键节点，其中，因缺乏时空隔离管控，无信号交叉口成为交通事故的高发区域。即使在有信号交叉口，信号灯失效或驾驶人分心等情况易使驾驶人面临无信号交叉口的通行场景。随着车联网（VehicletoEverything，V2X）技术的发展，协作式交叉口碰撞预警（Collaborative Intersection Collision Warning，CICW）通过网联车辆间的实时信息交互，预测风险并实现协同避碰[，有效减少了交通事故。因此，提升驾驶人的碰撞风险感知能力和应对能力，对提升无信号交叉□的交通安全具有重要意义。

目前，无信号交叉口的网联车辆碰撞预警系统的研究集中于车车碰撞风险检测和碰撞预警策略的制定。碰撞风险检测是预测车辆行为与潜在碰撞的关键，Yi等2提出一种改进的动态时空注意力网络模型预测车辆轨迹，研究主要考虑了车辆的直线运动，却忽略了其转弯行为，从而限制了模型在复杂交通场景中的预测准确性。Li等通过加入车辆转向速度，对车辆偏航率和侧向加速度的预测结果进行修正，但未考虑驾驶人真实驾驶行为的转弯车辆速度特性，无法准确反映车辆在交叉口的行为。Qu等利用碰撞时间（Time ToCollision，TTC）将车辆建模为矩形，虽然增强了检测的精确度，但计算复杂度较高。

现有的碰撞预警策略基于实时风险，仅向固定主车发出警告，忽略了交通系统的整体动态，从而影响通行效率。……