一种公交到站时间实时预测方法及系统
By integrating multi-source data through vehicle-to-everything (V2X) technology, a time-series modeling and error correction architecture is constructed, and model parameters are dynamically adjusted. This solves the data fusion and adaptability problems in bus arrival time prediction, and achieves accurate and fast bus arrival time prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- EAST CHINA JIAOTONG UNIVERSITY
- Filing Date
- 2026-01-07
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for predicting bus arrival times rely on a single data source and do not fully integrate road network traffic light status, real-time traffic flow density, passenger flow changes, and weather conditions. The model parameters are fixed and cannot be dynamically adapted to different traffic flow states. They also lack the ability to identify and respond to abnormal events, resulting in insufficient prediction accuracy and increased errors.
By acquiring multi-source data in real time through vehicle-to-everything (V2X) communication technology, performing preprocessing and feature fusion, constructing a two-level architecture for time-series modeling and error correction, dynamically adjusting model parameters in conjunction with real-time traffic flow density, establishing an abnormal event monitoring and hierarchical response mechanism, and realizing multi-terminal information interaction.
It achieves stable prediction performance in scenarios with smooth traffic, slow traffic, and congestion, with prediction errors controlled within ≤3 minutes during peak hours, ≤2 minutes during off-peak hours, and ≤5 minutes at night. It responds quickly to changes in road conditions and provides accurate predictions of bus arrival times.
Smart Images
Figure CN121725658B_ABST