EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application
This paper presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/1/89 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841548936373862400 |
|---|---|
| author | Leanne Miller Pedro J. Navarro Francisca Rosique |
| author_facet | Leanne Miller Pedro J. Navarro Francisca Rosique |
| author_sort | Leanne Miller |
| collection | DOAJ |
| description | This paper presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the Efficient Net convolutional module. To train the architecture, RGB and depth images were used together with inertial data as inputs to predict the driving speed and steering wheel angle. To pretrain the architecture, a synthetic multimodal dataset for autonomous driving applications was created. The dataset includes driving data from 100 diverse weather and traffic scenarios, gathered from multiple sensors including cameras and an IMU as well as from vehicle control variables. The results show that including edge detection layers in the architecture improves performance for transfer learning when using synthetic and real-world data. In addition, pretraining with synthetic data reduces training time and enhances model performance when using real-world data. |
| format | Article |
| id | doaj-art-98dcf9f4b0b747afa1c40d4440275490 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-98dcf9f4b0b747afa1c40d44402754902025-01-10T13:20:50ZengMDPI AGSensors1424-82202024-12-012518910.3390/s25010089EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving ApplicationLeanne Miller0Pedro J. Navarro1Francisca Rosique2División de Sistemas e Ingeniería Electrónica (DSIE), Campus Muralla del Mar, s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, SpainDivisión de Sistemas e Ingeniería Electrónica (DSIE), Campus Muralla del Mar, s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, SpainDivisión de Sistemas e Ingeniería Electrónica (DSIE), Campus Muralla del Mar, s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, SpainThis paper presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the Efficient Net convolutional module. To train the architecture, RGB and depth images were used together with inertial data as inputs to predict the driving speed and steering wheel angle. To pretrain the architecture, a synthetic multimodal dataset for autonomous driving applications was created. The dataset includes driving data from 100 diverse weather and traffic scenarios, gathered from multiple sensors including cameras and an IMU as well as from vehicle control variables. The results show that including edge detection layers in the architecture improves performance for transfer learning when using synthetic and real-world data. In addition, pretraining with synthetic data reduces training time and enhances model performance when using real-world data.https://www.mdpi.com/1424-8220/25/1/89end-to-end architecturesmultimodal synthetic datasetautonomous driving |
| spellingShingle | Leanne Miller Pedro J. Navarro Francisca Rosique EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application Sensors end-to-end architectures multimodal synthetic dataset autonomous driving |
| title | EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application |
| title_full | EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application |
| title_fullStr | EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application |
| title_full_unstemmed | EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application |
| title_short | EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application |
| title_sort | edgenet an end to end deep neural network pretrained with synthetic data for a real world autonomous driving application |
| topic | end-to-end architectures multimodal synthetic dataset autonomous driving |
| url | https://www.mdpi.com/1424-8220/25/1/89 |
| work_keys_str_mv | AT leannemiller edgenetanendtoenddeepneuralnetworkpretrainedwithsyntheticdataforarealworldautonomousdrivingapplication AT pedrojnavarro edgenetanendtoenddeepneuralnetworkpretrainedwithsyntheticdataforarealworldautonomousdrivingapplication AT franciscarosique edgenetanendtoenddeepneuralnetworkpretrainedwithsyntheticdataforarealworldautonomousdrivingapplication |