Cargando…

Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving

Deep neural network algorithms have achieved impressive performance in object detection. Real-time evaluation of perception uncertainty from deep neural network algorithms is indispensable for safe driving in autonomous vehicles. More research is required to determine how to assess the effectiveness...

Descripción completa

Detalles Bibliográficos
Autores principales: Yang, Mingliang, Jiang, Kun, Wen, Junze, Peng, Liang, Yang, Yanding, Wang, Hong, Yang, Mengmeng, Jiao, Xinyu, Yang, Diange
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10007375/
https://www.ncbi.nlm.nih.gov/pubmed/36905068
http://dx.doi.org/10.3390/s23052867
_version_ 1784905505583398912
author Yang, Mingliang
Jiang, Kun
Wen, Junze
Peng, Liang
Yang, Yanding
Wang, Hong
Yang, Mengmeng
Jiao, Xinyu
Yang, Diange
author_facet Yang, Mingliang
Jiang, Kun
Wen, Junze
Peng, Liang
Yang, Yanding
Wang, Hong
Yang, Mengmeng
Jiao, Xinyu
Yang, Diange
author_sort Yang, Mingliang
collection PubMed
description Deep neural network algorithms have achieved impressive performance in object detection. Real-time evaluation of perception uncertainty from deep neural network algorithms is indispensable for safe driving in autonomous vehicles. More research is required to determine how to assess the effectiveness and uncertainty of perception findings in real-time.This paper proposes a novel real-time evaluation method combining multi-source perception fusion and deep ensemble. The effectiveness of single-frame perception results is evaluated in real-time. Then, the spatial uncertainty of the detected objects and influencing factors are analyzed. Finally, the accuracy of spatial uncertainty is validated with the ground truth in the KITTI dataset. The research results show that the evaluation of perception effectiveness can reach 92% accuracy, and a positive correlation with the ground truth is found for both the uncertainty and the error. The spatial uncertainty is related to the distance and occlusion degree of detected objects.
format Online
Article
Text
id pubmed-10007375
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-100073752023-03-12 Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving Yang, Mingliang Jiang, Kun Wen, Junze Peng, Liang Yang, Yanding Wang, Hong Yang, Mengmeng Jiao, Xinyu Yang, Diange Sensors (Basel) Article Deep neural network algorithms have achieved impressive performance in object detection. Real-time evaluation of perception uncertainty from deep neural network algorithms is indispensable for safe driving in autonomous vehicles. More research is required to determine how to assess the effectiveness and uncertainty of perception findings in real-time.This paper proposes a novel real-time evaluation method combining multi-source perception fusion and deep ensemble. The effectiveness of single-frame perception results is evaluated in real-time. Then, the spatial uncertainty of the detected objects and influencing factors are analyzed. Finally, the accuracy of spatial uncertainty is validated with the ground truth in the KITTI dataset. The research results show that the evaluation of perception effectiveness can reach 92% accuracy, and a positive correlation with the ground truth is found for both the uncertainty and the error. The spatial uncertainty is related to the distance and occlusion degree of detected objects. MDPI 2023-03-06 /pmc/articles/PMC10007375/ /pubmed/36905068 http://dx.doi.org/10.3390/s23052867 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Yang, Mingliang
Jiang, Kun
Wen, Junze
Peng, Liang
Yang, Yanding
Wang, Hong
Yang, Mengmeng
Jiao, Xinyu
Yang, Diange
Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving
title Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving
title_full Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving
title_fullStr Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving
title_full_unstemmed Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving
title_short Real-Time Evaluation of Perception Uncertainty and Validity Verification of Autonomous Driving
title_sort real-time evaluation of perception uncertainty and validity verification of autonomous driving
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10007375/
https://www.ncbi.nlm.nih.gov/pubmed/36905068
http://dx.doi.org/10.3390/s23052867
work_keys_str_mv AT yangmingliang realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT jiangkun realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT wenjunze realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT pengliang realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT yangyanding realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT wanghong realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT yangmengmeng realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT jiaoxinyu realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving
AT yangdiange realtimeevaluationofperceptionuncertaintyandvalidityverificationofautonomousdriving