Road environment recognition for automotive FMCW RADAR systems through convolutional neural network

In this study, we propose a method to recognize road environments with automotive frequency-modulated continuous wave (FMCW) radar systems. For automotive radar systems on the road, diverse road environments are observed. Each road environment generates unnecessary echoes known as clutter, and the m...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2020-01, Vol.8, p.1-1
Main Authors: Sim, Heonkyo, Do, The-Duong, Lee, Seongwook, Kim, Yong-Hwa, Kim, Seong-Cheol
Format: Article
Language:English
Subjects:
Algorithms
Artificial neural networks
Automotive engineering
Automotive frequency-modulated continuous wave (FMCW) radar
Automotive radar
Autonomous vehicles
Classification
Clutter
Continuous radiation
convolutional neural network
Echoes
Iron
Neural networks
Object detection
Radar
Radar equipment
Radar systems
Recognition
road environment recognition
Roads
Target detection
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this study, we propose a method to recognize road environments with automotive frequency-modulated continuous wave (FMCW) radar systems. For automotive radar systems on the road, diverse road environments are observed. Each road environment generates unnecessary echoes known as clutter, and the magnitude distribution of received radar signal varies depending on road structures. Therefore, it is necessary to classify the road environment and adopt a target detection algorithm suitable for each road environment characteristic. To recognize the road environment in advance, it is necessary to identify the section where the road environment changes. In this paper, we define a changed road area as a transition region, and we classify the road environment and transition regions to improve the road environment recognition performance. Road environments are recognized by applying convolutional neural networks to the frequency-domain received signals of 77 GHz FMCW automotive radar systems. Experimental results in real-road environments demonstrate that the proposed method achieves 100% recognition performance, which is better if compared with that of the conventional methods.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2020.3013263