Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards Urban Planing AI Agent in the Age of Agentic AI

Rui Liu, Tao Zhe, Zhong-Ren Peng, Necati Catbas, Xinyue Ye, Dongjie Wang, Yanjie Fu

TL;DR

The paper argues that urban planning requires agentic AI and multimodal AI to overcome limitations of current GenAI approaches. It proposes an Agentic AI Urban Planner framework with Task Planner, Specialized Agents, Fusion, and Multimodal Generation, integrated into the planning workflow for problem scoping, visioning, data analysis, alternative generation, and deliberation. It discusses leveraging Vision-Language Models, multimodal reasoning, and domain tools to address real human needs and flood resilience, and outlines learning considerations including human–AI collaboration and digital twins. The practical impact is enabling transparent, participatory planning and faster evaluation of alternatives in complex urban contexts.

Abstract

Generative AI, large language models, and agentic AI have emerged separately of urban planning. However, the convergence between AI and urban planning presents an interesting opportunity towards AI urban planners. Existing studies conceptualizes urban planning as a generative AI task, where AI synthesizes land-use configurations under geospatial, social, and human-centric constraints and reshape automated urban design. We further identify critical gaps of existing generative urban planning studies: 1) the generative structure has to be predefined with strong assumption: all of adversarial generator-discriminator, forward and inverse diffusion structures, hierarchical zone-POI generative structure are predefined by humans; 2) ignore the power of domain expert developed tools: domain urban planners have developed various tools in the urban planning process guided by urban theory, while existing pure neural networks based generation ignore the power of the tools developed by urban planner practitioners. To address these limitations, we outline a future research direction agentic urban AI planner, calling for a new synthesis of agentic AI and participatory urbanism.

Towards Urban Planing AI Agent in the Age of Agentic AI

TL;DR

The paper argues that urban planning requires agentic AI and multimodal AI to overcome limitations of current GenAI approaches. It proposes an Agentic AI Urban Planner framework with Task Planner, Specialized Agents, Fusion, and Multimodal Generation, integrated into the planning workflow for problem scoping, visioning, data analysis, alternative generation, and deliberation. It discusses leveraging Vision-Language Models, multimodal reasoning, and domain tools to address real human needs and flood resilience, and outlines learning considerations including human–AI collaboration and digital twins. The practical impact is enabling transparent, participatory planning and faster evaluation of alternatives in complex urban contexts.

Abstract

Generative AI, large language models, and agentic AI have emerged separately of urban planning. However, the convergence between AI and urban planning presents an interesting opportunity towards AI urban planners. Existing studies conceptualizes urban planning as a generative AI task, where AI synthesizes land-use configurations under geospatial, social, and human-centric constraints and reshape automated urban design. We further identify critical gaps of existing generative urban planning studies: 1) the generative structure has to be predefined with strong assumption: all of adversarial generator-discriminator, forward and inverse diffusion structures, hierarchical zone-POI generative structure are predefined by humans; 2) ignore the power of domain expert developed tools: domain urban planners have developed various tools in the urban planning process guided by urban theory, while existing pure neural networks based generation ignore the power of the tools developed by urban planner practitioners. To address these limitations, we outline a future research direction agentic urban AI planner, calling for a new synthesis of agentic AI and participatory urbanism.

Paper Structure

This paper contains 31 sections, 1 figure.

Table of Contents

  1. Introduction and Motivation
  2. Existing Studies: Urban Planning as Generative AI
  3. Recent Studies in GenAI for Urban Planning
  4. The Underlying Generative Urban Planning Framework
  5. Representation: Modeling Urban Contexts and Human Needs
  6. Generation: Producing Optimal Land-Use and Built Environment Configurations
  7. Embedding Generative Urban Planning to Real Systems: Flooding Resilience
  8. Limitations of Current AI Urban Planner Research
  9. The Vision: Integrating VLM, Multimodal Reasoning, and Agentic AI for Urban Planning AI Agents
  10. Agentic AI Abilities: Implications for Urban Planning
  11. Reasoning
  12. Task decomposition
  13. Domain-specific tool usage
  14. The Agentic AI Urban Planner Framework
  15. Task Planner Agent
  16. ...and 16 more sections

Figures (1)

  • Figure 1: Framework Overview of Agentic AI based Automated Urban Planner.