MobileCity: An Efficient Framework for Large-Scale Urban Behavior Simulation

2025-04-18

AI Summary

Existing urban traffic simulation methods oversimplify travel choice modeling and struggle to support high-fidelity, real-time simulation of agent populations at the ten-thousand scale. This paper constructs a virtual city environment integrating diverse functional buildings and multimodal transportation (e.g., walking, bus, private car). Leveraging large-scale behavioral survey data and statistical modeling, it pioneers fine-grained representation of traffic decision-making mechanisms and heterogeneous population preferences within a generative agent framework. We propose a lightweight rule-based engine coupled with multi-granularity analysis—spanning microscopic trajectories and macroscopic density—to overcome computational bottlenecks, enabling real-time coordinated simulation of over 4,000 agents. Experimental validation confirms behavioral fidelity, demonstrated through both micro-level individual plausibility and macro-level distribution consistency. The framework further supports downstream tasks including crowd density forecasting and vehicle mode preference mining for demographic trend analysis.

Technology Category

Multiagent Systems: Agent-Based Simulation and Emergent BehaviorSearch and Optimization: Sampling/Simulation-based SearchPlanning, Routing, and Scheduling: Model-Based Reasoning

Application Category

Web Mining and Content Analysis: Web data generation and simulationUser Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systemsEconomics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Generative agents offer promising capabilities for simulating realistic urban behaviors. However, existing methods oversimplify transportation choices in modern cities, and require prohibitive computational resources for large-scale population simulation. To address these limitations, we first present a virtual city that features multiple functional buildings and transportation modes. Then, we conduct extensive surveys to model behavioral choices and mobility preferences among population groups. Building on these insights, we introduce a simulation framework that captures the complexity of urban mobility while remaining scalable, enabling the simulation of over 4,000 agents. To assess the realism of the generated behaviors, we perform a series of micro and macro-level analyses. Beyond mere performance comparison, we explore insightful experiments, such as predicting crowd density from movement patterns and identifying trends in vehicle preferences across agent demographics.

Problem

• Simulating realistic urban behaviors with generative agents
• Addressing oversimplified transportation choices in modern cities
• Reducing computational costs for large-scale population simulation

Innovation

• Virtual city with multiple functional buildings and transport
• Survey-based modeling of behavioral and mobility choices
• Scalable framework simulating 4000+ agents efficiently