Conference Agenda

Urban Digital Twins (UDTs) represent a powerful tool for understanding and managing the complex dynamics of urban environments. However, current UDT implementations often face challenges related to data interoperability, system integration, and scalability. Urban digital twins must integrate data from various sources, such as sensors, 3D models, and IoT devices, which often have different formats and semantics. This hetero-geneity complicates data interpretation and integration (Gu et al., 2024, Rafamatanantsoa et al., 2024). Moreover, The com-plexity of urban environments, with their diverse stakeholders and systems, makes it challenging to achieve interoperability. Stakeholders often struggle to understand and manage the com-plexity involved in integrating various data sources (Raes et al., 2021) Addressing these gaps, the Urban Digital Twin Interoperabil-ity Pilot (UDTIP)1, organized by the Open Geospatial Consor-tium (OGC), explores the development of a functional and in-teroperable UDT ecosystem through the integration of diverse urban data and standards-based workflows. The project fo-cuses on two main applications: urban traffic noise modeling and Geo-AI analysis. Noise modeling leverages 3D city mod-els in CityGML format, traffic profiles, and sensor readings to simulate and visualize urban noise levels, providing insights for planning and mitigation strategies. For Geo-AI analysis, camera imagery, INS metadata, and labeled training data are processed to enable object detection and road surface classi-fication tasks within urban environments. Central to the pro-ject is the use of OGC APIs to ensure seamless data exchange between modules. By aligning persistent elements, such as 3D city models, with dynamic information from IoT sensors and AI-driven analysis, the project demonstrates a viable pathway towards scalable and modular urban digital twins. Furthermore, stakeholder engagement with organizations such as the Land and Housing Agency of Korea and the United Nations ensures that the project outcomes address real-world needs and priorities.

Optimal transport seeks the most efficient way to transform one probability distribution into another, typically under constraints that preserve total mass. However, in many practical applications, such as point cloud and image co-registration, the source and target data distributions may have unequal mass. Herein, this is overcome by relaxing the aformentioned constraints. This proposed ``partial'' optimal transport framework adaptively selects and matches subsets of both source and target distributions, thereby enabling robust outlier rejection and noise reduction. The method relaxes the strict constraints typically used in linear programming formulations of optimal transport, thus allowing flexibility on both sides of the matching problem. The resulting transport plan is then refined through branch-and-cut and mass bounding procedures that enforce binary mass assignments and further prune undesired points. For scalability, traditional linear programming solvers are replaced by an efficient gradient-based algorithm. This approach is validated on synthetic two-dimensional examples and real-world three-dimensional point cloud data.

In smart cities, diverse modes of transport coexist, making the integration of multimodal networks essential. However, existing approaches typically simplify modal transitions into single points, which does not fully reflect real-world. To overcome these limitations, we propose a data model that represents mode transitions in three-dimensional space. The proposed model defines the transition between different transport modes as a 3D transfer space to represent internal routes in detail, and integrates the networks between individual transport networks through this space to form a unified multimodal transportation network. This approach enables precise modeling of transfer paths across micro-scale spaces and allows for the flexible integration of individual transport networks without requiring modifications. We expect that the proposed model will effectively address the demands of multilayered transportation environments and serve as a foundation for developing human-oriented multimodal navigation systems.