RuralAD is a comprehensive multi-modal, multi-season dataset recorded over an entire year in rural South Westphalia, Germany. Captured along a repeatedly driven 24 km long route, RuralAD encompasses diverse seasonal and weather conditions in 36 traversals. RuralAD integrates data from two lidar sensors, four automotive corner radar sensors, and six cameras for a diverse surround view. A RTK-DGNSS/INS system provides centimeter-accurate localization, while decoded vehicle CAN bus data supplies ego-vehicle status information. RuralAD includes a Lanelet2 HD map, a 3D point cloud map, and georeferenced aerial orthophotos. Instance-level object annotations are available for the first six months of recordings.  

The dataset and further documentation will soon be made available for academic use on this page. 

Citation 

F. Albers, S. Schütte and T. Bertram, „RuralAD: A Comprehensive Multi-Modal Multi-Season Dataset for Rural Automated Driving“, International Conference on Intelligent Transportation Systems (ITSC), 2026 (accepted)