A Unified Model for Human Mobility Generation in Natural Disasters

2025-11-02

AI Summary

Existing human mobility modeling approaches for natural disasters suffer from poor generalizability due to reliance on narrow-domain data—typically confined to single cities and single disaster types—rendering them inadequate for unforeseen disaster scenarios. To address this, we propose the first generalizable human mobility generation model capable of cross-city and cross-disaster-type adaptation. Our method innovatively integrates physics-informed prompting, physics-guided alignment, and a meta-learning framework, employing a shared-private parameter architecture to jointly capture universal disaster dynamics and city-specific heterogeneities. Extensive experiments across multiple real-world datasets encompassing earthquakes, floods, and typhoons in diverse cities demonstrate that our model achieves an average 13.2% improvement in prediction accuracy over state-of-the-art methods. Moreover, it significantly enhances zero-shot and few-shot adaptability to unseen disasters and previously unobserved cities.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Search and Optimization: Metareasoning and Metaheuristics
• Computer Vision: Low Level & Physics-based Vision

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Human mobility generation in disaster scenarios plays a vital role in resource allocation, emergency response, and rescue coordination. During disasters such as wildfires and hurricanes, human mobility patterns often deviate from their normal states, which makes the task more challenging. However, existing works usually rely on limited data from a single city or specific disaster, significantly restricting the model's generalization capability in new scenarios. In fact, disasters are highly sudden and unpredictable, and any city may encounter new types of disasters without prior experience. Therefore, we aim to develop a one-for-all model for mobility generation that can generalize to new disaster scenarios. However, building a universal framework faces two key challenges: 1) the diversity of disaster types and 2) the heterogeneity among different cities. In this work, we propose a unified model for human mobility generation in natural disasters (named UniDisMob). To enable cross-disaster generalization, we design physics-informed prompt and physics-guided alignment that leverage the underlying common patterns in mobility changes after different disasters to guide the generation process. To achieve cross-city generalization, we introduce a meta-learning framework that extracts universal patterns across multiple cities through shared parameters and captures city-specific features via private parameters. Extensive experiments across multiple cities and disaster scenarios demonstrate that our method significantly outperforms state-of-the-art baselines, achieving an average performance improvement exceeding 13%.

Problem

• Modeling disaster mobility patterns across diverse scenarios
• Overcoming generalization limitations in new disaster types
• Addressing cross-city heterogeneity in emergency response planning

Innovation

• Physics-informed prompts guide mobility generation
• Meta-learning framework extracts universal city patterns
• Shared and private parameters handle city heterogeneity