摘要
车辆轨迹预测(VTP)是交通技术领域中的重要研究对象。传统VTP方法需要大量特征工程,且难以实时适应复杂变化的环境。深度学习(DL)通过多层神经网络实现高效数据表达,克服了传统方法的局限性。对基于DL的VTP方法进行了综述,探讨了其在VTP中的应用及性能表现。首先,回顾了传统VTP方法和基于DL的VTP方法,介绍了VTP主要考虑的问题和问题的表述;其次,分析并比较了各类VTP方案,包括输入数据、输出结果和预测方法;再次,介绍了常用的评估指标,比较了这些VTP方案的实验结果,分析了VTP的应用,并展示了DL在VTP中表现出的优异性能;最后,展望了VTP未来在数据集、建模和计算效率方面的研究方向,指出车辆交互协同建模、模型的泛化以及多模态融合将是未来的挑战和研究方向。
Vehicle trajectory prediction(VTP)was a significant research subject in the transportation technology field.Traditional VTP methods require extensive feature engineering and struggle to adapt to complex and dynamic environments in real-time.Deep learning(DL)overcomes the limitations of traditional methods by achieving efficient data representation through multi-layer neural networks.Therefore,in this study a comprehensive review of DL-based VTP methods was carried out to explore their applications and performance in VTP.Firstly,the traditional VTP and DL-based VTP methods were explored,and the main consideration problems and problem formulations in VTP were introduced.Secondly,various VTP schemes,including input data,output results and prediction methods were analyzed and compared.Subsequently,commonly used evaluation metrics was introduced,and the experimental results of these VTP approaches were compared,the applications of VTP were analyzed,and the superior performance of DL in VTP were demonstrated.Finally,future research directions of VTP are discussed in terms of datasets,modeling approaches,and computational efficiency.It identifies that vehicle interaction collaborative modeling,model generalization,and multimodal fusion would constitute the primary challenges and research frontiers in the field.
作者
刘凯
汪佳琴
李汉涛
LIU Kai;WANG Jiaqin;LI Hantao(School of Electronics and Information Engineering,Beihang University,Beijing 100191,China)
出处
《郑州大学学报(工学版)》
北大核心
2025年第5期77-89,共13页
Journal of Zhengzhou University(Engineering Science)
基金
国家自然科学基金资助项目(U2233216,U2033215)。
关键词
车辆轨迹预测
深度学习
序列网络
图神经网络
生成模型
网格方法
vehicle trajectory prediction
deep learning
sequential network
graph neural network
generative model
grid method
