The invention belongs to the technical field of intelligent traffic, and particularly relates to a track restoration system based on an urban traffic network. The system comprises a traffic trajectory data acquisition module, a GPS trajectory feature extraction module and a GPS trajectory restoration learning module. Wherein the traffic trajectory data acquisition module acquires latitude and longitude data of a vehicle running on an urban road in a running process, and the GPS trajectory feature extraction module inputs acquired low-sampling-rate trajectory points into a neural network to extract spatial-temporal features. And the GPS trajectory restoration learning module inputs the output of the previous stage into a new neural network model to predict latitude and longitude information of missing sampling points in the input low-sampling-rate trajectory. According to the method, the utilization of low-sampling-rate trajectory data can be greatly improved; and meanwhile, in combination with a topological structure of an urban road network, the restored GPS track is more in line with a real track, and the utilization efficiency of track data is effectively improved.
本发明属于智能交通技术领域,具体为一种基于城市交通路网的轨迹还原系统。本发明系统包括交通轨迹数据采集模块、GPS轨迹特征抽取模块、GPS轨迹还原学习模块;其中交通轨迹数据采集模块采集城市道路上行驶的车辆在行驶过程中的经纬度数据,GPS轨迹特征抽取模块对采集到的低采样率的轨迹点输入神经网络中提取时空特征,GPS轨迹还原学习模块将上一阶段的输出输入到新的神经网络模型中预测输入的低采样率轨迹中缺失的采样点的经纬度信息。本发明能够大幅度提高低采样率轨迹数据的利用;同时结合城市路网的拓扑结构,使得还原出来的GPS轨迹更加符合真实的轨迹,有效提升对轨迹数据的利用效率。
Trajectory restoration system based on urban traffic road network
一种基于城市交通路网的轨迹还原系统
2023-08-18
Patent
Electronic Resource
Chinese
Trajectory reconstruction technology-based traffic restoration method
European Patent Office | 2022
|Road network traffic data restoration method based on DTW-RGCN
European Patent Office | 2020
|