Maschinelles Lernen für Fahrzeugsicherheitssysteme

Methoden der statistischen Signalverarbeitung haben durch die Einführung vorausschauender Sensoren vermehrt Einzug in Fahrzeuge erhalten. Die Verfahren spielen eine zentrale Rolle im Tracking von Objekten und in Schätzalgorithmen von Funktionen der integralen Fahrzeugsicherheit. Um die Vermeidung von Unfällen in komplexen Verkehrssituationen, mit vielen Verkehrsteilnehmern zu gewährleisten, werden Methoden des maschinellen Lernens erforscht. Im Fokus stehen dabei Verfahren, die es ermöglichen, maschinelles Lernen in Kombination mit modellbasierten Methoden in sicherheitskritischen Anwendungen zu nutzen.

Laufende Projekte

IDA

Intelligente Datenaufnahme im Bereich Automotive

ITraKS

Interaktionsbasierte Trajektorienprädiktion zur zuverlässigen Kollisionsvermeidung

KIaF

KI-Algorithmen für automatisiertes Fahren

Abgeschlossene Projekte

VbD

Absicherungsfähigkeit und Interpretation von maschinellen Lernverfahren für automatisiertes Fahren durch Entwurf

OLAF

Online Maschinelle Lernverfahren für Anwendungen der Fahrzeugsicherheit

MLPOG

Prädizierte Belegungskarten und statistische Lernverfahren zur Planung und Validierung von Fahrzeugsicherheitssystemen

SUP

Sichere Unfallprognose

HySLEUS

Hybride Statistische Lernmethoden für die Embedded-Umsetzung von Sicherheitsfunktionen im Fahrzeug

Veröffentlichungen

P. Riegl, K. C. Sekaran and M. Botsch, "Generation of Realistic Traffic Scenarios for Virtual and Real Test Drives Based on a Hybrid Machine Learning Framework," 2024 IEEE International Conference on Vehicular Electronics and Safety (ICVES), Ahmedabad, India, 2024. [ IEEE ]
M. Neumeier, S. Dorn, M. Botsch and W. Utschick, "Reliable Trajectory Prediction and Uncertainty Quantification with Conditioned Diffusion Models," 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Seattle, WA, USA, 2024. [ IEEE ]
K. C. Sekaran, L. Balasubramanian, M. Botsch and W. Utschick, "Open-Set Object Detection for the Identification and Localization of Dissimilar Novel Classes by means of Infrastructure Sensors," 2024 IEEE Intelligent Vehicles Symposium (IV), Jeju Island, Korea, Republic of, 2024. [ IEEE ]
A. Fertig, L. Balasubramanian and M. Botsch, "Clustering and Anomaly Detection in Embedding Spaces for the Validation of Automotive Sensors," 2024 IEEE Intelligent Vehicles Symposium (IV), Jeju Island, Korea, Republic of, 2024. [ IEEE ]

M. Neumeier, A. Tollkühn, S. Dorn, M. Botsch and W. Utschick, "Optimization and Interpretability of Graph Attention Networks for Small Sparse Graph Structures in Automotive Applications ", 35th IEEE Intelligent Vehicles Symposium, Anchorage, United States, 2023. [ IEEE | arXiv ]
M. Neumeier, A. Tollkühn, S. Dorn, M. Botsch and W. Utschick, " Gradient Derivation for Learnable Parameters in Graph Attention Networks", 2023. [ arXiv ]
M. Neumeier, S. Dorn, M. Botsch and W. Utschick, " Prediction and Interpretation of Vehicle Trajectories in the Graph Spectral Domain", 26thIEEE International Conference on Intelligent Transportation Systems, Bilbao, Spain, 2023. [ IEEE | arXiv ]
L. Balasubramanian, J. Wurst, R. Egolf, M. Botsch, W. Utschick and K. Deng, "SceneDiffusion: Conditioned Latent Diffusion Models for Traffic Scene Prediction," 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, 2023. [ IEEE ]
P. Nadarajan, M. Botsch and S. Sardina, "Continuous Probabilistic Motion Prediction Based on Latent Space Interpolation," 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, 2023. [ IEEE ]
K. C. Sekaran, L. Balasubramanian, M. Botsch and W. Utschick, "Metric Learning Based Class Specific Experts for Open-Set Recognition of Traffic Participants in Urban Areas Using Infrastructure Sensors," 2023 IEEE Intelligent Vehicles Symposium (IV), Anchorage, AK, USA, 2023. [ IEEE ]
L. Balasubramanian, J. Wurst, M. Botsch and K. Deng, "Open-World Learning for Traffic Scenarios Categorisation," in IEEE Transactions on Intelligent Vehicles, vol. 8, no. 5, pp. 3506-3521, May 2023. [ IEEE ]
M. Botsch, W. Huber, L. Balasubramanian, A. F. Fernández, M. Geisler, C. Gudera, M. R. Morales Gomez, P. Riegl, E. Sánchez Morales, M. Weinzierl, and K. C. Sekaran, "Data Collection and Safety Use Cases in Smart Infrastructures", In Adjunct Proceedings of the 15th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI '23 Adjunct), Association for Computing Machinery, New York, NY, USA, 2023. [ ACM ]

T. Elter, T. Dirndorfer, M. Botsch, W. Utschick, "Interaction-aware Prediction of Occupancy Regions Based on a POMDP Framework", IEEE International Conference on Intelligent Transportation Systems, Macau, Macau, 2022. [ IEEE ]
M. Neumeier, A. Tollkühn, M. Botsch and W. Utschick, " A Multidimensional Graph Fourier Transformation Neural Network for Vehicle Trajectory Prediction ", IEEE International Conference on Intelligent Transportation Systems, Macau, Macau, 2022. [ IEEE ]
L. Balasubramanian, J. Wurst, R. Egolf, M. Botsch, W. Utschick, K. Deng, "ExAgt: Expert-guided Augmentation for Representation Learning of Traffic Scenarios", IEEE International Conference on Intelligent Transportation Systems, Macau, Macau, 2022. [ IEEE | arXiv ]
J. Wurst, L. Balasubramanian, M. Botsch, W. Utschick, "Expert-LaSTS: Expert-Knowledge Guided Latent Space for Traffic Scenarios", IEEE Intelligent Vehicles Symposium, Aachen, Germany, 2022. [ IEEE | arXiv ]

M. Neumeier, A. Tollkühn, T. Berberich, and M. Botsch, "Variational autoencoder-based vehicle trajectory prediction with an interpretable latent space", 24th IEEE International Conference on Intelligent Transportation Systems, Indianapolis, United States, 2021. [arxiv pre-print]
L. Balasubramanian, F. Kruber, M. Botsch and K. Deng, "Open-set Recognition based on the Combination of Deep Learning and Ensemble Method for Detecting Unknown Traffic Scenarios", IEEE Intelligent Vehicles Symposium, Nagoya, Japan, 2021. [arxiv pre-print]
L. Balasubramanian, J. Wurst, M. Botsch and K. Deng, "Traffic Scenario Clustering by Iterative Optimisation of Self-Supervised Networks Using a Random Forest Activation Pattern Similarity", IEEE Intelligent Vehicles Symposium, Nagoya, Japan, 2021. [arxiv pre-print]
O. Gallitz, O. d. Candido, M. Botsch, and W. Utschick, “Interpretable Early Prediction of Lane Changes Using a Constrained Neural Network Architecture”, IEEE International Conference on Intelligent Transportation Systems, USA, September 2021. [IEEE]
J. Wurst, L. Balasubramanian, M. Botsch, W. Utschick, "Novelty Detection and Analysis of Traffic Scenario Infrastructures in the Latent Space of a Vision Transformer-Based Triplet Autoencoder ", IEEE Intelligent Vehicles Symposium, Nagoya, 2021. [arxiv pre-print]

O. Gallitz, O. d. Candido, R. Melz, M. Botsch, and W. Utschick, “Interpretable Machine Learning Structure for an Early Prediction of Lane Changes”, Artificial Neural Networks and Machine Learning – ICANN 2020, Slovakia, September 2020. [Springer link]
J. Wurst, A. Flores Fernández, M. Botsch and W. Utschick, "An Entropy Based Outlier Score and its Application to Novelty Detection for Road Infrastructure Images", IEEE Intelligent Vehicles Symposium, Las Vegas, 2020.

A. Chaulwar, M. Botsch, and W. Utschick, “Efficient Hybrid Machine Learning Algorithm for Trajectory Planning in Critical Traffic-Scenarios”, International Conference on Intelligent Transportation Engineering, Singapore, September 2019.
O. Gallitz, O. d. Candido, M. Botsch, and W. Utschick, “Interpretable Feature Generation using Deep Neural Networks and its Application to Lane Change Detection”, IEEE International Conference on Intelligent Transportation Systems, New Zealand, October 2019.

M. Müller, X. Long, M. Botsch, D. Böhmländer, and W. Utschick „Real-Time Crash Severity Estimation with Machine Learning and 2D Mass-Spring-Damper Model”, IEEE International Conference on Intelligent Transportation Systems, 2018.
M. Müller, M. Botsch, D. Böhmländer, W. Utschick, „A Simulation Framework for Vehicle Safety Testing / Ein Simulationsframework für die Absicherung von Fahrzeugsicherheitsfunktionen“, Fachbuch/Tagung "Aktive Sicherheit und Automatisiertes Fahren", expert Verlag, pp. 135-155, ISBN: 978-3-8169-3405-9, 2017.

M. Müller, M. Botsch, D. Böhmländer, W. Utschick, "Machine Learning Based Prediction of Crash Severity Distributions for Mitigation Strategies", Journal of Advances in Information Technology, Vol. 9, No. 1, pp. 15-24, February 2018. doi: 10.12720/jait.9.1.15-24.
P. Nadarajan, M. Botsch, and S. Sardina, "Machine Learning Architectures for the Estimation of Predicted Occupancy Grids in Road Traffic", Journal of Advances in Information Technology, Vol. 9, No. 1, pp. 1-9, February 2018. doi: 10.12720/jait.9.1.1-9.
A. Chaulwar, M. Botsch, and W. Utschick, “Generation of Reference Trajectories for Safe Trajectory Planning”, International Conference on Artificial Neural Networks, 2018.

G. Notomista, M. Botsch, "A Machine Learning Approach for the Segmentation of Driving Maneuvers and its Application in Autonomous Parking", Journal of Artificial Intelligence and Soft Computing Research, Volume 7, Issue 4, pp. 243-255, 2017.
A. Chaulwar, M. Botsch, W. Utschick, “A Machine Learning based Biased-Sampling Approach for Planning Safe Trajectories in Complex, Dynamic Traffic-Scenarios”, IEEE Intelligent Vehicles Symposium, 2017.
P. Nadarajan, M. Botsch, S. Sardina, “Predicted-Occupancy Grids for Vehicle Safety Applications based on Auotencoders and the Random Forest Algorithm”, International Joint Conference on Neural Networks, 2017.

A. Chaulwar, M. Botsch, W. Utschick, “A Hybrid Machine Learning Approach for Planning Safe Trajectories in Complex Traffic-Scenarios”, IEEE International Conference on Machine Learning and Applications, 2016.
M. Müller, P. Nadarajan, M. Botsch, W. Utschick, D. Böhmländer, S. Katzenbogen, “A Statistical Learning Approach for Estimating the Reliability of Crash Severity Predictions”, IEEE Intelligent Transportation Systems Society Conference, 2016.
G. Notomista, A. Kammenhuber, P. Nadarajan, M. Botsch, M. Selvaggio, “Relative Motion Estimation Based on Sensor Eigenfusion Using a Stereoscopic Vision System and Adaptive Statistical Filtering”, International Symposium on Robotics, 2016.
P. Nadarajan; M. Botsch, “Probability Estimation for Predicted-Occupancy Grids in Vehicle Safety Applications Based on Machine Learning”, IEEE Intelligent Vehicles Symposium, 2016.
Chaulwar, M. Botsch, T. Krueger und T. Miehling, “Planning of safe trajectories in dynamic multi-object traffic-scenarios”, Journal of Traffic and Logistics Engineering, Vol.4, no.2, 2016.

G. Notomista und M. Botsch, "Maneuver Segmentation for Autonomous Parking Based on Ensemble Learning", International Joint Conference on Neural Networks, 2015.