The 2nd Workshop on Foundation Models for Autonomous Driving (FMAD) at IEEE ITSC aims to identify challenges and opportunities of foundation models for autonomous driving and how their potential will disrupt future systems. It brings together leading researchers, engineers, and practitioners to explore the transformative potential of foundation models in autonomous driving systems. This workshop will examine cutting-edge applications of foundation models in perception, prediction, planning, and decision-making, while addressing critical challenges in safety-critical automotive environments.
The integration of foundation models into Autonomous Driving (AD) systems has the potential to revolutionize the field. Built on architectures with significant capacity, foundation models are capable of utilizing vast data collections through self-learning approaches. This enables them to achieve remarkable performance across diverse sets of tasks in different domains. Modern models such as DALL-E, CLIP, and SAM stand to be cornerstones for a wide range of solutions in the AD domain and have initiated new research directions.
The application of AD systems provides the opportunity to collect large amounts of data as they are equipped with high quality, multi-modal sensor suits. While the manual annotation of this data proves to be costly and time-intensive, foundation models not only allow tapping into this data source but also rapidly adapting to new tasks. Among other challenges, integrating different sensor modalities, considering the spatial and temporal nature of the data, and determining how to use it for planning and prediction remain open research questions.
Additionally, issues like limited availability of computational resources and the importance of safety considerations are inherent to AD platforms and need to be addressed. Overall, research interest is increasing in the field of foundation models for AD, which we want to explore with the following topics:
November 18, 2025
ITSC 2025, Gold Coast, Australia
Hybrid
We invite contributions on a broad range of topics related to foundation models and their application in autonomous driving systems.
Foundation models for camera-based perception, LiDAR processing, sensor fusion, and multi-modal understanding.
Motion forecasting, trajectory prediction, behavior modeling, and planning with foundation models.
Vision-language models, multi-modal approaches, and end-to-end systems for autonomous navigation.
Safety validation, robustness testing, adversarial scenarios, and verification methodologies.
Large-scale datasets, self-supervised learning, transfer learning, and data-efficient methods.
Real-time inference, edge computing, model compression, and practical deployment challenges.
| Time | Event |
|---|---|
| 08:45 | Greetings and Introduction — Hendrik Königshof & Frank Bieder (FZI & KIT) |
| 09:00 | Manuel Schwonberg, CARIAD Generalization Across Domains: Foundation Models for Autonomous Driving |
| 09:30 | Alejandro Galindo, Zoox Foundation Models @ Zoox |
| 10:00 | Break |
| 10:30 | José M. Álvarez, NVIDIA Foundation Models @ NVIDIA |
| 11:00 | Yağız Nalçakan, Yonsei University How robust are the foundation models for autonomous driving under adverse conditions? Challenges and Opportunities. |
| 11:30 | Youssef Shoeb, Conti/TU Berlin Finding the Unknowns: A Data Engine for Out-of-Distribution Perception |
| 12:00 | Yinzhe Shen, Karlsruhe Institute of Technology Motion and Semantic Learning in End-to-End Autonomous Driving |
| 12:25 | Discussion & Farewell — Hendrik Königshof & Frank Bieder (FZI & KIT) |
| 12:30 | Lunch |
The workshop is organized by leading researchers and practitioners in autonomous driving and AI.
For questions, please reach out to the organizing committee.
Email: bieder@fzi.de
Updates: Follow this website for announcements
