朱耀堃1,
吴欣虹1,
肖云鹏1,
吴海峰2
1.重庆邮电大学软件工程学院 重庆 400065
2.海口市气象局 海口 571199
基金项目:国家自然科学基金(61772098)，重庆市教委科技研究项目(KJQN201800641)，重庆邮电大学博士高端人才项目(BYJS2017004)，重庆市技术创新与应用发展专项面上项目(cstc2020jscx-msxmX0150)
详细信息
作者简介:李暾：男，1980年生，高级工程师，研究方向为机器学习和网络安全
朱耀堃：男，1994年生，硕士，研究方向为智慧交通、人工智能
吴欣虹：女，1995年生，硕士生，研究方向为社交网络、机器学习
肖云鹏：男，1979年生，教授，研究方向为大数据、移动互联网、信息安全
吴海峰：男，1978年生，助理工程师，研究方向为计算机网络
通讯作者:李暾 litun@cqupt.edu.cn
中图分类号:TP391计量
文章访问数:490
HTML全文浏览量:199
PDF下载量:68
被引次数:0
出版历程
收稿日期:2020-02-28
修回日期:2020-10-15
网络出版日期:2020-10-21
刊出日期:2021-05-18
Vehicle Trajectory Prediction Method Based on Intersection Context and Deep Belief Network
Tun LI1,,,Yaokun ZHU1,
Xinhong WU1,
Yunpeng XIAO1,
Haifeng WU2
1. School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
2. Haikou Meteorological Service, Haikou 571199, China
Funds:The National Natural Science Foundation of China (61772098), The Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN201800641), The Doctoral Top Talents Program of Chongqing University of Posts and Telecommunications (BYJS2017004), Chongqing Technology Innovation and Application Development Special General Project (cstc2020jscx-msxmX0150)
摘要
摘要:针对车辆轨迹预测中节点序列的时序特性和实际路网中的空间关联性，该文提出一种基于深度置信网络和SoftMax (DBN-SoftMax)轨迹预测方法。首先，考虑到轨迹在节点集合中的强稀疏性和一般特征学习方法对新特征的泛化能力不足，该文利用深度置信网络(DBN)较强的无监督特征学习能力，达到提取轨迹局部空间特性的目的；然后，针对轨迹的时序特性，该文采用逻辑回归的预测思路，用当前轨迹集在路网特征空间中的线性组合来预测轨迹；最后，结合自然语言处理领域中的词嵌入的思想，基于实际轨迹中节点存在的上下文关系，运用节点的向量集表征了节点间的交通时空关系。实验结果表明该模型不仅能够有效地提取轨迹特征，并且在拓扑结构复杂的路网中也能得到较好的预测结果。
关键词:智能交通/
轨迹预测/
卡口上下文分析/
特征提取/
深度置信网络
Abstract:For the temporal features of trajectory intersection sequence and spatial correlation of the actual road network, a trajectory prediction method based on the Deep Belief Networks and SoftMax (DBN-SoftMax) is proposed. At first, considering the sparsity of trajectory in an intersection set and the insufficiency of generalization ability in general feature learning methods for new features, the strong unsupervised feature learning ability of Deep Belief Network (DBN) is used to extract the local spatial features of trajectory. Secondly, considering the temporal features of the trajectory, the logistic regression method and the linear combination of the current trajectory set in the road network features space are used to predict the trajectory. Finally, Based on the idea of word embedding in the field of natural language processing and the contextual relationship of intersections in the actual trajectory, the vector set of intersections is used to represent the spatiotemporal relationship of traffic between intersections. The experimental results show that the model can not only extract the trajectory features effectively, but also obtain better prediction performance in a road network with complex topology.
Key words:Intelligent traffic/
Trajectory prediction/
Trajectory intersection context analysis/
Feature extract/
Deep Belief Network(DBN)
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=b438e087-c065-40e6-8404-dee3184a841f
