From Fragmented Mobility Data to Actionable Knowledge: Multimodal Data Fusion and Advanced AI for Sustainable Urban Transport

Urban transportation systems generate and depend on diverse data, including travel surveys, traffic and transit operations, mobile-phone and GPS traces, connected-vehicle and IoT sensors, Earth observation, street-level imagery, 3D city models, environmental measurements, energy use, administrative records, and citizen feedback. Yet these sources are often fragmented across platforms, modes, spatial and temporal scales, and semantic formats. This makes it difficult to understand how people and goods move, how transport systems interact with the urban environment, and where changes can support more sustainable, accessible, and resilient mobility.This Research Topic explores how multimodal data fusion and advanced artificial intelligence can turn fragmented urban mobility data into integrated knowledge and practical decision support. We welcome methods that combine data across transport modes and connect mobility with relevant urban factors, such as land use, infrastructure, emissions, energy, and socioeconomic conditions. Particular interest lies in approaches that use complementary spatial, temporal, semantic, and contextual information to describe, predict, diagnose, or improve transportation systems and mobility outcomes.Contributions may apply multimodal learning, graph-based AI, generative AI, foundation models, or AI agents to tasks such as travel-demand and traffic modelling, public transport and shared-mobility planning, accessibility analysis, transport equity, fleet electrification, charging and parking strategies, emissions reduction, and disruption or climate-resilience planning. Urban Digital Twins, knowledge graphs, and federated data environments may support integration, scenario testing, and operational use. Work should show a clear link to urban transportation systems or mobility and explain how its findings can inform planning, management, policy, or service design.Fusing these data is not a neutral technical exercise: choices about which sources are combined, how gaps and biases are handled, and whose mobility needs are foregrounded shape which transport problems become visible and actionable. This Research Topic therefore also welcomes work that critically examines these choices, asking who determines what is fused and for what purpose, and how transport authorities, service providers, and communities can use fused data to make decisions that are demonstrably more effective, equitable, and accountable, rather than simply better informed.Key challenges include cross-modal and spatiotemporal data fusion, semantic interoperability, data quality and uncertainty, privacy, model transferability, explainability, robustness, and the translation of AI outputs into trustworthy decisions. We welcome interdisciplinary research that addresses these challenges and advances sustainable, inclusive, and human-centred urban mobility.Topics of interest include, but are not limited to:• Multimodal fusion of transport, mobility, geospatial, Earth-observation, IoT, environmental, energy, socioeconomic, image, text, and citizen-generated data.• Spatiotemporal and cross-modal AI for travel behaviour, traffic, public transport, active travel, shared mobility, freight, and integrated mobility systems.• Graph neural networks, representation learning, hybrid AI, and other methods for modelling transport networks and mobility interactions.• Generative AI, large language models, foundation models, retrieval-augmented generation, and AI agents for mobility knowledge discovery, scenario generation, and decision support.• Trustworthy multimodal AI for mobility, including explainability, uncertainty, privacy, robustness, scalability, and transferability.• Critical and reflexive perspectives on data selection, bias, and governance in mobility data fusion, including who decides what is fused and how decisions based on fused data can be made more effective and accountable.Submissions should make urban transportation or mobility central to the research question and demonstrate a clear contribution to sustainable cities and communities, reduced inequalities, innovation and infrastructure, or climate action. Topic editors: Dessislava Petrova-Antonova, Mila Koeva, Youness Dehbi Source: https://www.frontiersin.org/research-topics/85799/from-fragmented-mobility-data-to-actionable-knowledge-multimodal-data-fusion-and-advanced-ai-for-sustainable-urban-transport