Lane Change Intention Recognition Dataset

If you plan to use the dataset in your work, please cite the related open-source paper at https://ieeexplore.ieee.org/document/11271346. This dataset supports research in lane change (LC) intention recognition and time-to-LC prediction in highway scenarios. Generated using the open-source CARLA driving simulator, it includes over 3,400 annotated LC maneuvers performed by 50 virtual drivers on a 60 km-long highway track with variable curvature and segment lengths. Each maneuver is synchronized with ego-vehicle dynamics, road geometry, and surrounding traffic context. The dataset is provided in both .csv and .h5 format. This resource is intended to facilitate reproducible research and foster development in automated driving.