Learning-empowered Intelligent Transportation Systems: Foundation Vehicles and Coordination Technique

Scope

This invited session aims to bring together researchers and engineers from academia and industry to discuss the advanced techniques in the evolving landscape of intelligent vehicles and transportation systems. It is the first year of our section, which covers all the areas of perception, decision-making, motion planning and control, dynamic interactions, networking and communication strategies of the connected automated vehicles. The emergence of advanced learning-based techniques has renewed vitality into the domain, ushering in novel opportunities and solutions, enabling intelligent connected vehicles to adeptly navigate increasingly complex scenarios. This year, the focus of this section is keenly set on exploring and deploying the foundation models as the bedrock of intelligent capabilities, leveraging artificial general intelligence to related automated vehicle missions, such as coordinated perception, trustworthy decision-making, robust multi-agent control, and continuous improvement of driving policies. Foundation models, trained from vast and diverse large-scale data, are seen as possessing human-like intelligence and common sense, demonstrating the potential to coordinate the multi-agent control and handle the rare long-tail cases, thereby significantly enhancing road safety. Furthermore, the session will also delve into emerging themes like continual/lifelong learning technologies, driving safety, system heterogeneity, resilience, privacy preservation, human-centric AI, and human-machine interaction. We are soliciting original contributions that are not published or currently under consideration by any other journals/conferences.

1. Data-driven and knowledge-driven autonomous driving
2. Fault-tolerance control and game-theory for connected intelligent vehicle systems
3. Optimization and control technique for connected intelligent vehicle systems
4. Heterogeneous coordinated technique for intelligent vehicle systems
5. Resilient and privacy-preserving control for intelligent vehicle systems
6. Vision foundation models (VFM) and Large language models (LLMs) for the perception, decision-making, planning, and control of connected automated vehicles
7. Continual/Lifelong learning, reinforcement learning, and imitation learning-based autonomous driving
8. Human-centric AI and Human-AI teaming for intelligent vehicle systems
9. Driving safety validation and safety of the intended functionality for connected automated vehicles
10. Digital twin and cyber security of connected automated vehicles