摘要
针对智能驾驶系统在驾驶风险预警中存在的动态交互特征捕捉不足、多模态轨迹预测精度有限以及碰撞风险量化物理指标过度单一等问题,研究了基于多模态轨迹预测与概率量化耦合的预见性碰撞风险评估模型。在轨迹预测部分,研究了分层图注意力网络,通过图注意力机制融合高精地图、车道线以及车辆历史轨迹特征,能够有效捕捉车辆行驶环境中的动态变化;针对传统模型中先预测再细化的两阶段解码结构,引入滑动窗口优化解码器,能够准确预测临近车辆的未来轨迹。在碰撞风险评估部分,研究了1种基于概率量化的碰撞风险评估方法,通过结合预测的未来轨迹与碰撞风险,估算自车与周边车辆发生碰撞的概率,实现对车辆危险行为的提前预警。实验结果表明:在Argoverse数据集上最小终点位移误差、最小平均位移误差和漏检率分别为0.785、1.157和0.126,与HiVT与LaneGCN相比,在终点预测方面误差分别减少了1%和15.1%。在城市交通能力仿真软件(simulation of urban mobility,SUMO)上验证预测风险与实际风险的偏差约为5%,从数据波动性上看,危险程度波动幅度为0.3,与碰撞时间(time to collision,TTC)方法和动态安全指数(dynamic safety index,DSI)方法相比,波动幅度分别减少33.3%和18.75%,在持续驾驶场景中展现出更优秀的风险评估水准;证明了基于障碍车辆轨迹预测的驾驶碰撞风险模型在预测未来潜在驾驶风险的准确性。
In order to address critical challenges in intelligent driving systems,including insufficient dynamic interaction modeling,limited accuracy in multimodal trajectory prediction,and over-reliance on single physical metrics for collision risk quantification.A proactive collision risk assessment framework is proposed by integrating probabilistic quantification with multimodal trajectory prediction.For trajectory prediction,a hierarchical graph attention network is developed to capture dynamic environmental features through adaptive fusion of high-definition maps,lane geometries,and vehicle motion history.A sliding window-optimized decoder is introduced within the conventional two-stage prediction architecture to refine trajectory outputs.For risk assessment,a probabilistic collision quantification method is designed to calculate collision likelihood between ego and surrounding vehicles based on predicted trajectories.Results on the Argoverse dataset demonstrate state-of-the-art performance with minimum final displacement error(=0.785),average displacement error(=1.157),and miss rate(=0.126),achieving 1%and 15.1%error reduction in endpoint prediction compared to HiVT and LaneGCN respectively.simulation of urban mobility,SUMO simulations reveal 5%deviation between predicted and actual risks,with risk fluctuation amplitude reduced by 33.3%and 18.75%against time to collision(TTC)and dynamic safety index(DSI)methods.The proposed model shows enhanced stability in continuous driving scenarios(risk fluctuation=0.3)and demonstrates superior accuracy in forecasting potential collision risks through systematic integration of trajectory prediction and probabilistic analysis.These findings validate the framework’s effectiveness in proactive safety warning for intelligent vehicles.
作者
杨厚新
陆丽萍
秦恒
杨奥
褚端峰
YANG Houxin;LU Liping;QIN Heng;YANG Ao;CHU Duanfeng(Communication and Information Center of Hubei Provincial Department of Transportation,Wuhan 430030,China;School of Computer Science and Artificial Intelligence,Wuhan University of Technology,Wuhan 430079,China;Intelligent Transportation Systems Research Center,Wuhan University of Technology,Wuhan 430063,China)
出处
《交通信息与安全》
北大核心
2025年第1期42-51,共10页
Journal of Transport Information and Safety
基金
国家自然科学基金面上项目(52172392)
湖北省交通运输厅科技项目(2023-121-3-2)资助。
关键词
交通安全
预见性碰撞风险评估
车辆轨迹预测
图神经网络
风险概率量化
traffic safety
predictive crash risk assessment
vehicle trajectory prediction
graph neural networks
Risk probability quantification
作者简介
第一作者:杨厚新(1968-),本科,高级工程师.研究方向:交通运输行业信息化研究.E-mail:331011781@qq.com;通信作者:陆丽萍(1977-),博士,副教授.研究方向:无人驾驶技术.E-mail:luliping@whut.edu.cn。
