English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Predicting the future trajectories of surrounding vehicles based on theirhistory trajectories is a critical task in autonomous driving. However, whensmall crafted perturbations are introduced to those history trajectories, theresulting anomalous (or adversarial) trajectories can significantly mislead thefuture trajectory prediction module of the ego vehicle, which may result inunsafe planning and even fatal accidents. Therefore, it is of great importanceto detect such anomalous trajectories of the surrounding vehicles for systemsafety, but few works have addressed this issue. In this work, we propose twonovel methods for learning effective and efficient representations for onlineanomaly detection of vehicle trajectories. Different from general time-seriesanomaly detection, anomalous vehicle trajectory detection deals with muchricher contexts on the road and fewer observable patterns on the anomaloustrajectories themselves. To address these challenges, our methods exploitcontrastive learning techniques and trajectory semantics to capture thepatterns underlying the driving scenarios for effective anomaly detection undersupervised and unsupervised settings, respectively. We conduct extensiveexperiments to demonstrate that our supervised method based on contrastivelearning and unsupervised method based on reconstruction with semantic latentspace can significantly improve the performance of anomalous trajectorydetection in their corresponding settings over various baseline methods. Wealso demonstrate our methods' generalization ability to detect unseen patternsof anomalies.

Learning Representation for Anomaly Detection of Vehicle Trajectories