Deep learning for lateral vehicle control – an end-to-end trained multi-fusion steering model
- Deep Learning based behavior reflex methods found their way into modern vehicles. To model the human driving behavior it is not sufficient to rely solely on individual, noncontiguous camera frames without taking vehicle signals or road specific features into account. In this work four temporal fusion methods are evaluated based on three different Deep Learning models. The proposed spatio-temporal Mixed Fusion model extends the present end-to-end models and consist of multiple levels of fusions. The raw image data from a single front facing camera is mixed with recorded vehicle data and a map based predicted road bank angle gradient vector. The model accesses multiple time axes: temporal features of multiple image frames are extracted through a combination of Convolution and LSTM layers while it can also make assumptions about the future road condition with the use of upcoming Ground Truth road bank angle changes. Experiments are performed on a recorded data set of real world drivings. Results show, that this approach leads to anDeep Learning based behavior reflex methods found their way into modern vehicles. To model the human driving behavior it is not sufficient to rely solely on individual, noncontiguous camera frames without taking vehicle signals or road specific features into account. In this work four temporal fusion methods are evaluated based on three different Deep Learning models. The proposed spatio-temporal Mixed Fusion model extends the present end-to-end models and consist of multiple levels of fusions. The raw image data from a single front facing camera is mixed with recorded vehicle data and a map based predicted road bank angle gradient vector. The model accesses multiple time axes: temporal features of multiple image frames are extracted through a combination of Convolution and LSTM layers while it can also make assumptions about the future road condition with the use of upcoming Ground Truth road bank angle changes. Experiments are performed on a recorded data set of real world drivings. Results show, that this approach leads to an accurate imitation of the human driver with an inference capability of more than 60 FPS.…
Author: | Johann Haselberger, Jian Chen, Bernhard SchickORCiD |
---|---|
DOI: | https://doi.org/10.1007/978-3-658-26435-2_17 |
Identifier: | 978-3-658-26435-2 OPAC HS OPAC extern |
Parent Title (English): | 10th International Munich Chassis Symposium 2019 |
Publisher: | Springer |
Place of publication: | Wiesbaden |
Document Type: | conference proceeding (article) |
Conference: | International Munich Chassis Symposium, 10th, 2019, Munich |
Language: | English |
Year of first Publication: | 2019 |
Tag: | MLPaSSAD Automated Driving; KI; automated lateral control; deep learning; vehicle dynamics |
Number of pages: | 27 Seiten |
First Page: | 201 |
Last Page: | 227 |
Institutes: | IFM - Institut für Fahrerassistenz und vernetzte Mobilität |
Dewey Decimal Classification: | 6 Technik, Medizin, angewandte Wissenschaften / 62 Ingenieurwissenschaften |
Research focus: | FSP2: Mobilität |
Publication Lists: | Schick, Bernhard |
Haselberger, Johann | |
Research Project: | MLPaSSAD – New Multi-Layer Platforms for Security and Safety-Relevant Automated Driving Functions |
Publication reviewed: | begutachtet |
Release Date: | 2020/07/10 |