New Publication on Agentic Urban Digital Twins and Human-AI Co-Learning

GEAR Lab is pleased to share a new publication in Urban Informatics by Xinyue Ye, Wenjing Gong, Yifan Yang, Lei Zou, Zhengzhong Tu, Xiao Huang, Zhenlong Li, Huan Ning, and Ling Wu.

The paper, titled “Towards Agentic Urban Digital Twins (AUDiTs): Advancing New Urban Science through Human-AI Co-Learning Agents,” introduces Agentic Urban Digital Twins as a research agenda for the next generation of urban digital twin systems. The work explores how large language models, multimodal AI agents, and digital twin environments can work together to support context-aware reasoning, participatory scenario design, ethical deliberation, and human-AI co-learning.

This vision paper moves beyond viewing digital twins only as tools for monitoring, prediction, or optimization. Instead, it frames urban digital twins as collaborative platforms where humans and AI systems can learn from each other, test scenarios, explain complex urban processes, and support more responsible decisions for cities and communities.

The paper also identifies key challenges for this emerging direction, including AI bias and fairness, data limitations, computational sustainability, institutional alignment, and the need for transparent and value-sensitive design. By connecting urban science, GeoAI, digital twins, and human-centered AI, this work offers a timely framework for building more adaptive, explainable, and socially responsible urban intelligence systems.

Congratulations to the authors on this important contribution to urban informatics, new urban science, and the future of human-AI collaboration in city planning and governance.

Citation

Ye, X., Gong, W., Yang, Y., Zou, L., Tu, Z., Huang, X., Li, Z., Ning, H., & Wu, L. (2026). Towards Agentic Urban Digital Twins (AUDiTs): Advancing New Urban Science through Human-AI Co-Learning Agents. Urban Informatics, 5(1), Article 9. https://doi.org/10.1007/s44212-025-00099-3