TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation

TrajGenAgent：用于人类移动轨迹生成的层次化大模型智能体

Abstract: Human mobility data is important for transportation, urban planning, and epidemic control, but large-scale trajectory collection is often costly and privacy-constrained, motivating realistic synthetic trajectory generation. 摘要： 人类移动数据对于交通运输、城市规划和流行病防控至关重要，但大规模轨迹数据的采集往往成本高昂且受限于隐私保护，这促使了对逼真合成轨迹生成技术的需求。

Existing LLM-based generators typically rely on either prompt engineering, which preserves zero-shot reasoning but lacks fine-grained spatiotemporal grounding, or trajectory-level fine-tuning, which improves statistical precision but incurs substantial computational cost and may weaken general reasoning. 现有的基于大模型（LLM）的生成器通常依赖于两种方法：一是提示工程（Prompt Engineering），它保留了零样本推理能力，但缺乏细粒度的时空基础；二是轨迹级微调（Trajectory-level Fine-tuning），它提高了统计精度，但带来了巨大的计算成本，并可能削弱模型的通用推理能力。

We propose TrajGenAgent, a semantic-aware hierarchical LLM-agent framework for human mobility trajectory generation without model fine-tuning. 我们提出了 TrajGenAgent，这是一个无需模型微调、具备语义感知能力的层次化大模型智能体框架，用于人类移动轨迹生成。

TrajGenAgent uses a two-stage orchestrator-worker design: an LLM first synthesizes an individual- and weekday-conditioned activity chain from historical evidence via in-context learning, and a deterministic workflow then grounds each activity into a complete visit using personalized POI retrieval, distance-aware location selection, kinematics-aware travel-time propagation, and LLM-based duration estimation. TrajGenAgent 采用了“编排者-执行者”（orchestrator-worker）的两阶段设计：首先，大模型通过上下文学习（In-context Learning），根据历史证据合成一条基于个人特征和工作日条件的活动链；随后，确定性工作流将每项活动落实为完整的访问记录，具体手段包括个性化兴趣点（POI）检索、距离感知的位置选择、运动学感知的旅行时间传播以及基于大模型的时长估算。

To evaluate realism beyond aggregate spatiotemporal statistics, we introduce an anomaly-detection-based evaluation framework using two complementary detectors to assess behavioral and semantic plausibility. 为了评估超越聚合时空统计数据的真实性，我们引入了一个基于异常检测的评估框架，利用两个互补的检测器来评估行为和语义的合理性。

Experiments on benchmark and large-scale simulation datasets show that TrajGenAgent improves spatiotemporal fidelity, semantic coherence, and individual-specific behavioral realism over representative neural and LLM-based baselines, while avoiding parameter updates. 在基准数据集和大规模模拟数据集上的实验表明，TrajGenAgent 在提升时空保真度、语义连贯性以及个体行为真实性方面均优于现有的代表性神经模型和基于大模型的基准方法，同时避免了参数更新。