Anomaly Detection in Multi-Agent Trajectories for Automated Driving

Julian Wiederer, Arij Bouazizi, Marco Troina, Ulrich Kressel, Vasileios Belagiannis
Proceedings of the 5th Conference on Robot Learning, PMLR 164:1223-1233, 2022.

Abstract

Human drivers can recognise fast abnormal driving situations to avoid accidents. Similar to humans, automated vehicles are supposed to perform anomaly detection. In this work, we propose the spatio-temporal graph auto-encoder for learning normal driving behaviours. Our innovation is the ability to jointly learn multiple trajectories of a dynamic number of agents. To perform anomaly detection, we first estimate a density function of the learned trajectory feature representation and then detect anomalies in low-density regions. Due to the lack of multi-agent trajectory datasets for anomaly detection in automated driving, we introduce our dataset using a driving simulator for normal and abnormal manoeuvres. Our evaluations show that our approach learns the relation between different agents and delivers promising results compared to the related works. The code, simulation and the dataset are publicly available.

Cite this Paper

BibTeX
@InProceedings{pmlr-v164-wiederer22a, title = {Anomaly Detection in Multi-Agent Trajectories for Automated Driving}, author = {Wiederer, Julian and Bouazizi, Arij and Troina, Marco and Kressel, Ulrich and Belagiannis, Vasileios}, booktitle = {Proceedings of the 5th Conference on Robot Learning}, pages = {1223--1233}, year = {2022}, editor = {Faust, Aleksandra and Hsu, David and Neumann, Gerhard}, volume = {164}, series = {Proceedings of Machine Learning Research}, month = {08--11 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v164/wiederer22a/wiederer22a.pdf}, url = {https://proceedings.mlr.press/v164/wiederer22a.html}, abstract = {Human drivers can recognise fast abnormal driving situations to avoid accidents. Similar to humans, automated vehicles are supposed to perform anomaly detection. In this work, we propose the spatio-temporal graph auto-encoder for learning normal driving behaviours. Our innovation is the ability to jointly learn multiple trajectories of a dynamic number of agents. To perform anomaly detection, we first estimate a density function of the learned trajectory feature representation and then detect anomalies in low-density regions. Due to the lack of multi-agent trajectory datasets for anomaly detection in automated driving, we introduce our dataset using a driving simulator for normal and abnormal manoeuvres. Our evaluations show that our approach learns the relation between different agents and delivers promising results compared to the related works. The code, simulation and the dataset are publicly available.} }
Endnote
%0 Conference Paper %T Anomaly Detection in Multi-Agent Trajectories for Automated Driving %A Julian Wiederer %A Arij Bouazizi %A Marco Troina %A Ulrich Kressel %A Vasileios Belagiannis %B Proceedings of the 5th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2022 %E Aleksandra Faust %E David Hsu %E Gerhard Neumann %F pmlr-v164-wiederer22a %I PMLR %P 1223--1233 %U https://proceedings.mlr.press/v164/wiederer22a.html %V 164 %X Human drivers can recognise fast abnormal driving situations to avoid accidents. Similar to humans, automated vehicles are supposed to perform anomaly detection. In this work, we propose the spatio-temporal graph auto-encoder for learning normal driving behaviours. Our innovation is the ability to jointly learn multiple trajectories of a dynamic number of agents. To perform anomaly detection, we first estimate a density function of the learned trajectory feature representation and then detect anomalies in low-density regions. Due to the lack of multi-agent trajectory datasets for anomaly detection in automated driving, we introduce our dataset using a driving simulator for normal and abnormal manoeuvres. Our evaluations show that our approach learns the relation between different agents and delivers promising results compared to the related works. The code, simulation and the dataset are publicly available.
APA
Wiederer, J., Bouazizi, A., Troina, M., Kressel, U. & Belagiannis, V.. (2022). Anomaly Detection in Multi-Agent Trajectories for Automated Driving. Proceedings of the 5th Conference on Robot Learning, in Proceedings of Machine Learning Research 164:1223-1233 Available from https://proceedings.mlr.press/v164/wiederer22a.html.

Related Material