A road closure detection method based on multi-feature fusion

Abstract

The invention discloses a road closure detection method based on multi-feature fusion, which comprises the following steps: in the offline stage, grid processing is performed on the area to be detected and a grid index is established for trajectory data and road network data. Subsequently, the traffic correlation strength between road sections is extracted; at the same time, based on the historical trajectory data, the grid traffic flow sequence, the strongly associated road turning flow sequence of each road section, and the vehicle U-turn frequency sequence in a local area are obtained. In the online detection stage, the combined model of CNN and LSTM is used to predict the current traffic flow of each grid, and to screen candidate closed grids. The combined model of GCN and LSTM is used to predict the turning amount of the strongly associated roads in each road section and the U-turn frequency in a local area, and then the closed road section is identified according to the sudden decrease of the steering flow or the sharp increase of the U-turn frequency. Finally, combined with the driving yaw detection and the bicycle track data, the closed type of the road is further judged. The invention improves detection efficiency to ensure real-time performance.

Description

technical field [0001] The invention belongs to the technical field of trajectory mining, and in particular relates to a road closure detection method based on multi-feature fusion. Background technique [0002] With the widespread use of GPS devices, residents' daily travel is increasingly dependent on map navigation. In order to ensure the high precision of the electronic map, it is urgent to sense the dynamic changes of the road network to update the map. In recent years, a large amount of research has been devoted to the detection of road topology changes such as missing road discovery in the road network, intersection location and range identification. In the road network, due to traffic accidents, traffic control, road construction and other factors, some road sections will be impassable. The untimely detection of road closure events not only brings inconvenience to residents, but also leads to huge economic losses. For example, on October 10, 2019, the No. 1 bridge...

AI Technical Summary

Problems solved by technology

The invention proposes a road closure detection method based on multi-feature fusion for the phenomenon that various factors in daily life lead to temporary or long-term impassability of the road section, and the electronic map is not updated in time, resulting in inaccurate navigation.

Images