Fault Detection and Recovery for Automotive Perception Sensors Based on AI and Regarding Driver Comfort

@inproceedings{Scheffmann2024,
title = {Fault Detection and Recovery for Automotive Perception Sensors Based on AI and Regarding Driver Comfort},
author = {Marco Scheffmann and Alia Salah and Prof. Dr. Hans-Christian Reuss},
editor = {Andras Kemeny and Jean-Rémy Chardonnet and Florent Colombet and Stéphane Espié},
isbn = {978-2-9573777-5-6},
year  = {2024},
date = {2024-09-18},
urldate = {2024-09-18},
booktitle = {Proceedings of the Driving Simulation Conference 2024 Europe VR},
pages = {195-202},
address = {Strasbourg, France},
organization = {Driving Simulation Association},
abstract = {The accuracy of perception sensors on board the vehicle is one of the major aspects contributing towards autonomous driving. In order to ensure fault-free operation and support the functional safety, the reliability of these sensors need to be guaranteed. In this contribution, a unique fault detection and compensation technique for lidar sensors is proposed. Specifically targeted in this work, is the distance-keeping function, which depends mainly on the distance provided by the lidar sensor measurement. This technique of fault detection and compensation takes place online and while driving. The concept of fault detection is mainly based on deep learning algorithms, which are trained using timeseries of certain features indicating the reliability of the sensor. Fault compensation utilizes machine learning clustering to isolate the relevant points for distance measurements. The implementation of artificial intelligence algorithms takes advantage of the modern techniques like digital twins and over-the air-updates, which enable the execution and outsourcing of these algorithms to high-performance server architectures. For the implementation and testing of this technique, a co-simulation framework is designed as an intermediate testing environment before the real implementation on the driving simulator. The later allows further investigating the fault detection and compensation technique and the effect on the driver comfort. The preliminary results of the realized algorithm for distance-keeping functions within the co-simulation environment and the driving simulator are promising and demonstrate a great potential for this specific implementation.},
keywords = {},
}

Download .bib file

TY  - CONF
TI  - Fault Detection and Recovery for Automotive Perception Sensors Based on AI and Regarding Driver Comfort
AU  - Scheffmann, Marco
AU  - Salah, Alia
AU  - Reuss, Prof. Dr. Hans-Christian
C1  - Strasbourg, France
C3  - Proceedings of the Driving Simulation Conference 2024 Europe VR
DA  - 2024/09/18
PY  - 2024
SP  - 195
EP  - 202
LA  - en-US
PB  - Driving Simulation Association
SN  - 978-2-9573777-5-6
L2  - https://proceedings.driving-simulation.org/proceeding/dsc-2024/fault-detection-and-recovery-for-automotive-perception-sensors-based-on-ai-and-regarding-driver-comfort
ER  - 

Download .ris file