Projects | UrbanXAI

Urban Mobility Sensing, Intelligence, and Services

Advancing data- and AI-driven methods for sensing, modeling, predicting, and improving urban mobility systems and services. This research integrates multimodal sensing and mobility data with spatial and network methods to extract traffic information, model movement and system dynamics, forecast mobility conditions, and support routing, navigation, and operational decisions. It develops intelligent mobility and location-based services that connect continuous sensing, real-time analytics, optimization, and feedback, translating mobility intelligence into better transportation planning, responsive operations, and effective service delivery.

Big-Data-Enabled Urban Analytics, Monitoring, and Decision-Making

Developing data-driven methods and analytical frameworks that transform heterogeneous urban data into knowledge, ongoing monitoring, and informed decision-making for cities and communities. This research integrates spatiotemporal and multivariate analytics, spatial statistics, AI, and domain knowledge to detect patterns and changes, characterize interactions and disparities, assess evolving urban conditions across scales, and connect analytical insights with decision-making, action, and feedback. Applications span population and demographic dynamics, urban development and planning, accessibility and inequality, sustainability and resilience, and public policy.

Scalable Spatial Optimization and GeoAI Systems

Developing scalable optimization methods, algorithms, and computational frameworks for spatial problems involving complex structures, objectives, and constraints. This research direction focuses on graph optimization, network partitioning, spatial allocation, regionalization, service-area design, and other large-scale spatial optimization problems. It translates algorithmic advances into reusable computational tools and GeoAI systems for real-world applications, with particular emphasis on solution quality, computational efficiency, reliability, and interpretable, human-in-the-loop decision support.

GeoAI Foundations & Multimodal Spatial Reasoning

Underlying these research directions is a shared foundation in methods that enable machines to represent, integrate, learn from, and reason about spatial information. This work advances spatial representation learning, geospatial data and knowledge engineering, multimodal information integration, and spatial reasoning across maps, imagery, trajectories, networks, text, sensor data, and urban knowledge. These capabilities provide a common methodological foundation for our research in urban mobility, spatial optimization, and big-data-enabled urban understanding, monitoring, and decision-making, while also advancing fundamental methods in GeoAI.