Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems
Test, verification, and development activities of vehicles with ADAS (Advanced Driver Assistance Systems) and ADF (Automated Driving Functions) generate large amounts of measurement data. To efficiently evaluate and use this data, a generic understanding and classification of the relevant driving sc...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-12-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/8/12/2590 |
id |
doaj-fe462fac16f4434889c0d1eb77ccae3c |
---|---|
record_format |
Article |
spelling |
doaj-fe462fac16f4434889c0d1eb77ccae3c2020-11-25T00:44:15ZengMDPI AGApplied Sciences2076-34172018-12-01812259010.3390/app8122590app8122590Deep Learning Applied to Scenario Classification for Lane-Keep-Assist SystemsHalil Beglerovic0Thomas Schloemicher1Steffen Metzner2Martin Horn3AVL List GmbH, Hans-List-Platz 1, 8020 Graz, AustriaAVL List GmbH, Hans-List-Platz 1, 8020 Graz, AustriaAVL List GmbH, Hans-List-Platz 1, 8020 Graz, AustriaInstitute of Automation and Control, Graz University of Technology, Inffeldgasse 21b, 8010 Graz, AustriaTest, verification, and development activities of vehicles with ADAS (Advanced Driver Assistance Systems) and ADF (Automated Driving Functions) generate large amounts of measurement data. To efficiently evaluate and use this data, a generic understanding and classification of the relevant driving scenarios is necessary. Currently, such understanding is obtained by using heuristic algorithms or even by manual inspection of sensor signals. In this paper, we apply deep learning on sensor time series data to automatically extract relevant features for classification of driving scenarios relevant for a Lane-Keep-Assist System. We compare the performance of convolutional and recurrent neural networks and propose two classification models. The first one is an online model for scenario classification during driving. The second one is an offline model for post-processing, providing higher accuracy.https://www.mdpi.com/2076-3417/8/12/2590deep learningscenario classificationautomated driving functionsautomated driving functions (ADF)advanced driver assistance systems (ADAS) |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Halil Beglerovic Thomas Schloemicher Steffen Metzner Martin Horn |
spellingShingle |
Halil Beglerovic Thomas Schloemicher Steffen Metzner Martin Horn Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems Applied Sciences deep learning scenario classification automated driving functions automated driving functions (ADF) advanced driver assistance systems (ADAS) |
author_facet |
Halil Beglerovic Thomas Schloemicher Steffen Metzner Martin Horn |
author_sort |
Halil Beglerovic |
title |
Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems |
title_short |
Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems |
title_full |
Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems |
title_fullStr |
Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems |
title_full_unstemmed |
Deep Learning Applied to Scenario Classification for Lane-Keep-Assist Systems |
title_sort |
deep learning applied to scenario classification for lane-keep-assist systems |
publisher |
MDPI AG |
series |
Applied Sciences |
issn |
2076-3417 |
publishDate |
2018-12-01 |
description |
Test, verification, and development activities of vehicles with ADAS (Advanced Driver Assistance Systems) and ADF (Automated Driving Functions) generate large amounts of measurement data. To efficiently evaluate and use this data, a generic understanding and classification of the relevant driving scenarios is necessary. Currently, such understanding is obtained by using heuristic algorithms or even by manual inspection of sensor signals. In this paper, we apply deep learning on sensor time series data to automatically extract relevant features for classification of driving scenarios relevant for a Lane-Keep-Assist System. We compare the performance of convolutional and recurrent neural networks and propose two classification models. The first one is an online model for scenario classification during driving. The second one is an offline model for post-processing, providing higher accuracy. |
topic |
deep learning scenario classification automated driving functions automated driving functions (ADF) advanced driver assistance systems (ADAS) |
url |
https://www.mdpi.com/2076-3417/8/12/2590 |
work_keys_str_mv |
AT halilbeglerovic deeplearningappliedtoscenarioclassificationforlanekeepassistsystems AT thomasschloemicher deeplearningappliedtoscenarioclassificationforlanekeepassistsystems AT steffenmetzner deeplearningappliedtoscenarioclassificationforlanekeepassistsystems AT martinhorn deeplearningappliedtoscenarioclassificationforlanekeepassistsystems |
_version_ |
1725275427410280448 |