Vehicle trajectory prediction method and device based on road ETC data, and system
By extracting features and assigning weights to ETC data, and combining this with a time-series prediction model to process trajectory vector sequences, the problem of underutilization of the spatial relationship of ETC gantry in existing technologies is solved, thus improving the accuracy of vehicle trajectory prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ANHUI TRAFFIC CONTROL INFORMATION IND CO LTD
- Filing Date
- 2026-03-30
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, vehicle trajectory prediction methods fail to effectively integrate the spatial relationships between highway ETC gantries, resulting in inaccurate trajectory prediction results.
By acquiring ETC data, initial feature vectors are extracted, and a weighted fusion model is used to weight and fuse the feature vectors. The trajectory vector sequence is then processed by a time-series prediction model to generate a predicted state sequence, thereby improving the accuracy of trajectory prediction.
It enhances the accuracy of vehicle trajectory prediction by quantifying information and analyzing the spatial relationship between ETC gantries using a unified dimensional analysis, thereby achieving dynamic expression and information association of trajectory prediction.
Smart Images
Figure CN122416716A_ABST