Apparatus and method for predicting traffic information
By calculating vehicle spacing and density and combining it with deep learning algorithms, a density-speed model is constructed, which solves the error problem caused by variables such as weather in existing traffic information prediction methods, and achieves more accurate speed prediction and future traffic information prediction.
Patent Information
- Application Number
- CN202110829538.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-10-26
- Filing Date
- 2021-07-22
- Publication Date
- 2026-02-06
- Estimated Expiration
- 2041-07-22
AI Technical Summary
Existing traffic information prediction methods rely on past speed patterns and cannot effectively cope with changes in variables such as weather and seasons, resulting in a high probability of prediction errors, especially at the micro level, such as speed prediction of link units.
By calculating data on vehicle spacing, front-to-back spacing, and vehicle density, and using data acquired by sensors and cameras installed on vehicles, a density-speed pattern is constructed. Combined with deep learning algorithms, a representative speed is derived and existing speed predictions are corrected.
It improves the accuracy of traffic information prediction, enabling accurate speed prediction under different density conditions, reducing errors, and achieving continuity and future prediction of traffic conditions.