This special issue addresses how multimodal large language models and agentic AI are transforming transportation systems. By combining language understanding with visual, spatial, and sensor data, these technologies enable more intelligent autonomous driving, traffic management, and mobility services. The focus is on advancing safety, efficiency, accessibility, and sustainability across transportation networks.
The issue tackles key challenges in deploying these systems in transportation, including grounding language in real traffic scenes, handling diverse sensor inputs, generalizing across different cities and conditions, and ensuring trustworthy decision-making in safety-critical scenarios. Submissions should connect foundational AI methods to rigorous evaluation and real-world transportation outcomes.
The special issue welcomes research on architectures, algorithms, datasets, benchmarks, and deployment of multimodal and LLM-based systems across autonomous driving, traffic prediction, infrastructure coordination, public transit, and freight logistics. Research addressing safety, fairness, privacy, human oversight, and sustainability in transportation AI is also welcomed.