Trajectory Data Management and Mining: A Survey from Deep Learning to the LLM Era

TL;DR

This survey analyzes trajectory data management and mining through the lens of deep learning and the LLM era. It develops a unified taxonomy linking trajectory forms, management and mining tasks, and application domains, and it documents evolving methods from DL-based pipelines to foundation models and LLMs. It introduces the Trajectory Computing Project as a living repository of datasets, tools, and papers, and synthesizes cross-cutting themes such as cross-city transfer, multimodal fusion, and privacy-aware learning. The findings show that large models and multimodal representations can unify disparate trajectory tasks, enabling scalable, knowledge-rich mobility intelligence with practical implications for smart cities, transportation, and public safety.

Abstract

Trajectory computing is a pivotal domain encompassing trajectory data management and mining, garnering widespread attention due to its crucial role in various practical applications such as location services, urban traffic, and public safety. Traditional methods, focusing on simplistic spatio-temporal features, face challenges of complex calculations, limited scalability, and inadequate adaptability to real-world complexities. In this paper, we present a comprehensive review of the development and recent advances in trajectory computing, from deep learning to the more recent large language models. We first define trajectory data and provide a brief overview of widely-used deep learning models. Systematically, we explore deep learning applications in trajectory management (pre-processing, storage, analysis, and visualization) and mining (trajectory-related forecasting, trajectory-related recommendation, trajectory classification, travel time estimation, anomaly detection, and mobility generation). Furthermore, we discuss emerging research directions and recent advancements in large models (represented by foundation models and large language models) for trajectory computing, which promise to reshape the next generation of trajectory computing. Additionally, we summarize application scenarios, public datasets, and toolkits. Finally, we outline current challenges in trajectory computing research and propose future directions. Relevant papers and open-source resources have been collated and are continuously updated at: https://github.com/yoshall/Awesome-Trajectory-Computing.

Figures (11)

• Figure 1: Trajectory computing overview.
• Figure 2: Illustration of a trajectory.
• Figure 3: Taxonomy of this survey with representative trajectory data management and mining works.
• Figure 4: Pre-Processing example.
• Figure 5: Different pipeline of trajectory similarity methods.

Theorems & Definitions (5)

• Definition 1: Spatio-Temporal Point
• Definition 2: Trajectory
• Definition 3: Matrix
• Definition 4: Graph
• Definition 5: Raster