Related Experiment Video
Updated: Oct 11, 2025
14:55
Evaluating the Effect of Roadside Parking on a Dual-Direction Urban Street
Published on: January 20, 2023
3.6K
A high-resolution trajectory data driven method for real-time evaluation of traffic safety.
Yuping Hu1, Ye Li1, Helai Huang1
1School of Traffic and Transportation Engineering, Central South University, Changsha, Hunan, 410075, PR China.
Accident; Analysis and Prevention
|December 5, 2021
Summary
This study introduces a novel method for real-time road safety evaluation by combining traffic states and conflicts using high-resolution trajectory data. The random forest model demonstrated high accuracy in predicting conflict risk, enhancing proactive traffic safety management.
Area of Science:
- Traffic Safety Engineering
- Transportation Data Science
Background:
- Real-time safety evaluation is crucial for proactive traffic management and enhancing road safety.
- Existing methods require improvement for high-precision, lane-level safety assessments.
Purpose of the Study:
- To propose a novel method for real-time road safety evaluation.
- To explore the relationship between traffic states and conflicts using high-resolution trajectory data.
- To develop a lane-level safety assessment tool.
Main Methods:
- Utilized the HighD dataset for lane-based trajectory data collection.
- Employed the time-to-collision (TTC) index for surrogate safety measure and conflict identification.
- Applied binary logistic regression and machine learning algorithms (SVM, Decision Tree, Random Forest, Gradient Boosting) for model training.
Main Results:
- Trained 24 models using four distinct classifier algorithms.
- Achieved the best performance with the Random Forest model, yielding an overall accuracy of 0.85.
- Demonstrated that the proposed method effectively estimates conflict risk.
Conclusions:
- The developed method provides a high-precision evaluation of real-time traffic safety.
- Findings support the development of proactive traffic safety management strategies.
- Integration of traffic states and conflicts offers valuable insights into road safety.
