TY  - RPRT
AU  - Aach, Marcel
AU  - Adolph, Laurenz
AU  - Baumann, Stefan
AU  - Benassou, Sabrina
AU  - Bernhard, David
AU  - Bernhard, Sebastian
AU  - Bouzidi, Mohamed-Khalil
AU  - Braun, Markus
AU  - Brox, Thomas
AU  - Burdorf, Sven
AU  - Dauner, Daniel
AU  - Graaff, Thies de
AU  - Doğan, Samed
AU  - Derajic, Bojan
AU  - Flohr, Fabian
AU  - Galesso, Silvio
AU  - Ganzer, Malte
AU  - Gottschalk, Hanno
AU  - Heid, Dominik
AU  - Heue, Falk
AU  - Hu, Tao
AU  - Hubschneider, Christian
AU  - Kästingschäfer, Marius
AU  - Keser, Mert
AU  - Khongsab, Peerayut
AU  - Kowol, Kamil
AU  - Krause, Felix
AU  - Kromm, Edward
AU  - Lucente, Giovanni
AU  - Lukin, Artem
AU  - Mariani, Annajoyce
AU  - Mittal, Sudhanshu
AU  - Mousakhan, Arian
AU  - Mualla, Firas
AU  - Molin, Adam
AU  - Neuhöfer, Jonas
AU  - Niemeijer, Joshua
AU  - Bernal, Christian Ojeda
AU  - Ommer, Björn
AU  - Ourania Tze, Christina
AU  - Peyinghaus, Sven
AU  - Piecha, Pascal
AU  - Prestel, Ulrich
AU  - Ramazzina, Andrea
AU  - Reichardt, Jörg
AU  - Rist, Christoph
AU  - Ritter, Werner
AU  - Rochau, Dennis
AU  - Savarino, Fabrizio
AU  - Schenkel, Philipp
AU  - Schlauch, Christian
AU  - Schmidt, Julian
AU  - Taş, Ömer Şahin
AU  - Thiel, Laurenz
AU  - Vivekanandan, Abhishek
AU  - Walz, Stefanie
AU  - Wang, Jiangtao
AU  - Wang, Zhaoze
AU  - Wiederer, Julian
AU  - Winter, Katharina
AU  - Wunderlich, Carolin
AU  - Yadav, Harsh
AU  - Yao, Yue
AU  - Yang, Yutong
TI  - Automotive Foundation Models: Fundamentals, Training Concepts and Applications
IS  - Whitepaper 1
M1  - FZJ-2026-00918
M1  - Whitepaper 1
SP  - 177 p.
PY  - 2025
AB  - Generative AI has gained wide attention through large-scale models such as DALL-E 3, StableDiffusion, and GPT-4, creating new opportunities for innovation while raising important ethicalquestions. At the core of these advancements are foundation models: massive neural networkspretrained on extensive datasets and designed to adapt across diverse tasks and data modalities.Their ability to unify camera, LiDAR, radar, and other sensor data has significant potential forenhancing perception, prediction, and planning in autonomous driving. <br>This report discusses the fundamental concepts underlying foundation models, including theirarchitectural structures, training regimens, and interpretability considerations. It examines cur-rent literature for foundation models and their application and use in the domain of autonomousdriving. Techniques for generative sensor data synthesis are outlined, demonstrating how arti-ficial datasets can replicate complex driving environments and reduce the expense of collectingreal-world samples. Methods for extending these generative approaches to sequential or videodata are also highlighted, enabling realistic motion forecasting and scenario simulation. Fi-nally, the report explores how abstract representations, such as semantic maps or symbolic data,can increase explainability and computational efficiency when applied to autonomous driving.This whitepaper aims to provide both foundational understanding and practical guidance ofthe current state-of-the-art for leveraging generative AI and foundation models in the field ofautonomous driving.
LB  - PUB:(DE-HGF)29
UR  - https://juser.fz-juelich.de/record/1052307
ER  -