UVTM: Universal Vehicle Trajectory Modeling with ST Feature Domain Generation

2024-02-11

AI Summary

Incomplete GPS trajectory features—such as sparsity and missing data—hinder effective multi-task modeling and necessitate maintaining multiple task-specific models. Method: We propose the first general-purpose vehicle trajectory foundation model for multi-task learning. Our approach introduces a novel spatiotemporal feature tri-domain independent masking and generation mechanism, enabling a sparse-to-complete trajectory reconstruction pretraining paradigm. It integrates spatiotemporal-decoupled representation learning, domain-aware feature generation, and a Transformer-based architecture. Contribution/Results: Evaluated on three real-world trajectory datasets, the single unified model achieves or surpasses state-of-the-art (SOTA) performance across four downstream tasks—including travel time estimation, trajectory recovery, and trajectory prediction—under both zero-shot and lightweight fine-tuning settings. This significantly reduces deployment complexity and maintenance overhead compared to conventional task-specific modeling.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Multi-modal Vision
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Vehicle movement is frequently captured in the form of GPS trajectories, i.e., sequences of timestamped GPS locations. Such data is widely used for various tasks such as travel-time estimation, trajectory recovery, and trajectory prediction. A universal vehicle trajectory model could be applied to different tasks, removing the need to maintain multiple specialized models, thereby reducing computational and storage costs. However, creating such a model is challenging when the integrity of trajectory features is compromised, i.e., in scenarios where only partial features are available or the trajectories are sparse. To address these challenges, we propose the Universal Vehicle Trajectory Model (UVTM), which can effectively adapt to different tasks without excessive retraining. UVTM incorporates two specialized designs. First, it divides trajectory features into three distinct domains. Each domain can be masked and generated independently to accommodate tasks with only partially available features. Second, UVTM is pre-trained by reconstructing dense, feature-complete trajectories from sparse, feature-incomplete counterparts, enabling strong performance even when the integrity of trajectory features is compromised. Experiments involving four representative trajectory-related tasks on three real-world vehicle trajectory datasets provide insight into the performance of UVTM and offer evidence that it is capable of meeting its objectives.

Problem

• Modeling vehicle trajectories with partial or sparse features
• Adapting a single model to multiple trajectory-related tasks
• Reducing computational costs by avoiding specialized models

Innovation

• Divides trajectory features into three domains
• Masks and generates each domain independently
• Pre-trains by reconstructing dense from sparse trajectories