Steve Willits
Papers
1
Total Citations
25
H-Index
1
About
Steve Willits is a leading figure in off-road autonomous driving, whose work has fundamentally advanced self-supervised learning for unstructured environments. His primary research areas center on large-scale robotic datasets, multimodal sensor fusion, and robust perception for extreme terrain. Willits’s major contribution is the creation of the TartanDrive dataset series, which has become a cornerstone resource for the field. His seminal work, *TartanDrive 2.0: More Modalities and Better Infrastructure to Further Self-Supervised Learning Research in Off-Road Driving Tasks* (2024, 25 citations), builds directly on his earlier 2021 release—one of the largest off-road driving datasets ever compiled. This follow-up captures seven hours of high-speed driving data at up to 15 m/s, providing the research community with richer modalities and improved infrastructure for developing models that learn without manual labels. By tackling the extreme challenges of off-road perception—where terrain is unpredictable and labels are scarce—Willits has enabled new approaches to autonomy that are both data-efficient and robust. His work is essential reading for any researcher pushing the boundaries of self-supervised learning in robotics.
Research Focus
Key Achievements
Top Papers
- 1