CoDesignAI: An AI-Enabled Multi-Agent, Multi-User System for Collaborative Urban Design at the Conceptual Stage

Abstract

Public participation has become increasingly important in collaborative urban design; yet, existing processes often face challenges in achieving efficient and scalable citizen engagement. To address this gap, this study explores how large language models (LLMs) can support cooperation among community members in participatory design. We introduce CoDesignAI, a collaborative urban design tool that combines multiple users, representing residents or stakeholders, with multiple AI agents, representing domain experts who provide facilitation and professional knowledge during the conceptual stage of urban design. This paper presents the system architecture and main components of the tool, illustrating how users interact with AI agents within a collaborative and iterative design workflow. Specifically, the system integrates generative AI with spatial mapping services to support street-level visualization of design proposals. AI agents assist users by summarizing discussion content, extracting shared design intentions, and generating prompts for presenting design interventions. The system also enables users to revise and refine their ideas over multiple rounds while documenting the design process. By combining conversational AI, multi-user interaction, and image-based design grounded in real-world urban contexts, this study argues that AI-enabled design systems can help shift urban design from an expert-centered practice to a more open and participatory process. The paper contributes a new web-based platform for AI-assisted collaborative design and offers an early exploration of how AI agents may expand the capacity for public participation in urban design.

Figures (8)

• Figure 1: The figure illustrates the concept of a multi-agent, multi-user system and explains why such a framework is needed to coordinate multiple participants. First, selected community representatives enter the design platform as users, where an AI facilitator tracks the conversation and summarizes shared insights, thereby improving the flexibility and efficiency of the discussion process. Second, simulated AI expert agents (e.g., AI planner, AI designer) are integrated into the system to answer users’ questions, thereby enhancing professional support and collaboration. To reduce the potential bias introduced by involving only a limited number of representatives, CoDesignAI enables that the process can be repeated n times, with m users selected in each round, until a broader portion of the community or even the entire community is covered.
• Figure 2: System architecture of CoDesignAI. The platform supports round-based collaborative urban design through interactions between users, backend coordination modules, AI services, and persistent storage. User discussions are coordinated through a room-based round mechanism, processed by AI facilitator and AI expert agents, and transformed into structured design prompts and image-based visualizations derived from Google Street View scenes.
• Figure 3: Users who share the same “Room ID” are assigned to the same collaborative workspace and wait until all users are ready to start. During this stage, they may also choose to add AI expert agents. Once all users are ready, they enter the collaborative workspace and begin the design process.
• Figure 4: Any user in the system may initiate the conversation. However, to encourage balanced participation, the AI facilitator provides a summary and response only after all users have contributed their input.
• Figure 5: The AI facilitator generates prompts based on users’ inputs, which are then presented as key design intentions. Users can review these suggestions by editing, removing, or adding new items. After confirming the AI-generated design intentions, they can click “Generate AI Image” to browse the generated results.