Deep-Learning Based Multi-Object Detection and Tracking using Range-Angle Map in Automotive Radar Systems

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

In this paper, a machine learning-based object detection and tracking approach in radar system is proposed via using the range-angle map as the input. Specifically, by using the You Only Look Once (YOLO) for object detection and Deep Simple Online and Realtime Tracking (D-SORT) for tracking, the proposed approach can improve the detection and tracking performance, reducing the parameters needed to be manually selected, and providing more relevant information, such as the shape, size, and category of the object. We conduct the realistic simulations to evaluate the proposed approach. Results show that our proposed approach can outperform the conventional radar processing approach in terms of detection and tracking performance. Furthermore, results indicate that the object categorization of the proposed approach is accurate.

Original languageEnglish
Title of host publication2022 IEEE 95th Vehicular Technology Conference - Spring, VTC 2022-Spring - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665482431
DOIs
StatePublished - 2022
Event95th IEEE Vehicular Technology Conference - Spring, VTC 2022-Spring - Helsinki, Finland
Duration: 19 Jun 202222 Jun 2022

Publication series

NameIEEE Vehicular Technology Conference
Volume2022-June
ISSN (Print)1550-2252

Conference

Conference95th IEEE Vehicular Technology Conference - Spring, VTC 2022-Spring
Country/TerritoryFinland
CityHelsinki
Period19/06/2222/06/22

Keywords

  • Multiple object Tracking
  • automotive radars
  • deep learning.
  • object detection
  • range-angle map

Fingerprint

Dive into the research topics of 'Deep-Learning Based Multi-Object Detection and Tracking using Range-Angle Map in Automotive Radar Systems'. Together they form a unique fingerprint.

Cite this