GATSim: Urban Mobility Simulation with Generative Agents

2025-06-29

AI Summary

Traditional rule-based traffic simulation struggles to capture the complexity and adaptivity of human travel decision-making. This paper proposes GATSim, a novel generative agent simulation framework that pioneers the integration of large language models (LLMs) with cognitive architectures to instantiate individual agents endowed with social attributes, psychological mechanisms, memory and reflective reasoning, tool-use capabilities, and lifelong learning. By embedding these generative agents within a multi-agent traffic simulation environment, GATSim achieves dynamic coupling between micro-level behavioral evolution and macro-level traffic flow patterns. Experimental results demonstrate that the simulated travel behaviors achieve fidelity comparable to human-annotated ground truth, while naturally emerging traffic patterns align closely with real-world observations. The framework advances traffic simulation toward cognitively grounded, adaptive, and scalable agent-based modeling. The source code is publicly available.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Traditional agent-based urban mobility simulations rely on rigid rule-based systems that fail to capture the complexity, adaptability, and behavioral diversity characteristic of human travel decision-making. Recent advances in large language models and AI agent technology offer opportunities to create agents with reasoning capabilities, persistent memory, and adaptive learning mechanisms. We propose GATSim (Generative-Agent Transport Simulation), a novel framework that leverages these advances to create generative agents with rich behavioral characteristics for urban mobility simulation. Unlike conventional approaches, GATSim agents possess diverse socioeconomic attributes, individual lifestyles, and evolving preferences that shape their mobility decisions through psychologically-informed memory systems, tool usage capabilities, and lifelong learning mechanisms. The main contributions of this study include: (1) a comprehensive architecture combining an urban mobility foundation model with agent cognitive systems and transport simulation environment, (2) a fully functional prototype implementation, and (3) systematic validation demonstrating that generative agents produce believable travel behaviors. Through designed reflection processes, generative agents in this study can transform specific travel experiences into generalized insights, enabling realistic behavioral adaptation over time with specialized mechanisms for activity planning and real-time reactive behaviors tailored to urban mobility contexts. Experiments show that generative agents perform competitively with human annotators in mobility scenarios while naturally producing macroscopic traffic evolution patterns. The code for the prototype system is shared at https://github.com/qiliuchn/gatsim.

Problem

• Simulate human-like urban mobility with adaptive generative agents
• Overcome rigid rule-based systems in traditional mobility simulations
• Integrate AI reasoning and memory for realistic travel behaviors

Innovation

• Generative agents with reasoning and memory
• Urban mobility foundation model integration
• Lifelong learning for behavioral adaptation