Data underlying the publication: A unified probabilistic approach to traffic conflict detection

This dataset includes the resulting data of the research: A unified probabilistic approach to traffic conflict detection. It contains a bundled collection of processed experimental data that excludes raw sensor recordings and initial preprocessing outputs. The study is on traffic conflict detection using a unified probabilistic approach. The highD dataset is used for lane-changing dynamics, and the 100Car NDS is used for crash and near-crash events to test the proposed approach. The raw data went through pre-processing steps that add heading directions, extract lane-change trajectories, and align crash data, enabling both conflict probability estimation and intensity evaluation. The scripts that produced these data are open-sourced at https://github.com/Yiru-Jiao/UnifiedConflictDetection