CroTad: A Contrastive Reinforcement Learning Framework for Online Trajectory Anomaly Detection

TL;DR

CroTad introduces a novel contrastive reinforcement learning framework for online, threshold-free trajectory anomaly detection that effectively handles noisy, irregular data and identifies anomalies at both sub-trajectory and point levels.

Contribution

It presents a new contrastive reinforcement learning approach that improves real-time, fine-grained anomaly detection in trajectories without relying on thresholds.

Findings

Outperforms existing methods in real-world datasets

Robust to noisy and irregularly sampled data

Capable of detecting anomalies at sub-trajectory and point levels

Abstract

Detecting trajectory anomalies is a vital task in modern Intelligent Transportation Systems (ITS), enabling the identification of unsafe, inefficient, or irregular travel behaviours. While deep learning has emerged as the dominant approach, several key challenges remain unresolved. First, sub-trajectory anomaly detection, capable of pinpointing the precise segments where anomalies occur, remains underexplored compared to whole-trajectory analysis. Second, many existing methods depend on carefully tuned thresholds, limiting their adaptability in real-world applications. Moreover, the irregular sampling of trajectory data and the presence of noise in training sets further degrade model performance, making it difficult to learn reliable representations of normal routes. To address these challenges, we propose a contrastive reinforcement learning framework for online trajectory anomaly detection, CroTad. Our method is threshold-free and robust to noisy, irregularly sampled data. By incorporating contrastive learning, CroTad learns to extract diverse normal travel patterns for different itineraries and effectively distinguish anomalous behaviours at both sub-trajectory and point levels. The detection module leverages deep reinforcement learning to perform online, real-time anomaly scoring, enabling timely and fine-grained identification of abnormal segments. Extensive experiments on two real-world datasets demonstrate the effectiveness and robustness of our framework across various evaluation scenarios.