Cargando…
Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data
Accurate segmentation of drivable areas and road obstacles is critical for autonomous mobile robots to navigate safely in indoor and outdoor environments. With the fast advancement of deep learning, mobile robots may now perform autonomous navigation based on what they learned in the learning phase....
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9268933/ https://www.ncbi.nlm.nih.gov/pubmed/35808244 http://dx.doi.org/10.3390/s22134751 |
_version_ | 1784744108748701696 |
---|---|
author | Nguyen, Toan-Khoa Nguyen, Phuc Thanh-Thien Nguyen, Dai-Dong Kuo, Chung-Hsien |
author_facet | Nguyen, Toan-Khoa Nguyen, Phuc Thanh-Thien Nguyen, Dai-Dong Kuo, Chung-Hsien |
author_sort | Nguyen, Toan-Khoa |
collection | PubMed |
description | Accurate segmentation of drivable areas and road obstacles is critical for autonomous mobile robots to navigate safely in indoor and outdoor environments. With the fast advancement of deep learning, mobile robots may now perform autonomous navigation based on what they learned in the learning phase. On the other hand, existing techniques often have low performance when confronted with complex situations since unfamiliar objects are not included in the training dataset. Additionally, the use of a large amount of labeled data is generally essential for training deep neural networks to achieve good performance, which is time-consuming and labor-intensive. Thus, this paper presents a solution to these issues by proposing a self-supervised learning method for the drivable areas and road anomaly segmentation. First, we propose the Automatic Generating Segmentation Label (AGSL) framework, which is an efficient system automatically generating segmentation labels for drivable areas and road anomalies by finding dissimilarities between the input and resynthesized image and localizing obstacles in the disparity map. Then, we train RGB-D datasets with a semantic segmentation network using self-generated ground truth labels derived from our method (AGSL labels) to get the pre-trained model. The results showed that our AGSL achieved high performance in labeling evaluation, and the pre-trained model also obtains certain confidence in real-time segmentation application on mobile robots. |
format | Online Article Text |
id | pubmed-9268933 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-92689332022-07-09 Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data Nguyen, Toan-Khoa Nguyen, Phuc Thanh-Thien Nguyen, Dai-Dong Kuo, Chung-Hsien Sensors (Basel) Article Accurate segmentation of drivable areas and road obstacles is critical for autonomous mobile robots to navigate safely in indoor and outdoor environments. With the fast advancement of deep learning, mobile robots may now perform autonomous navigation based on what they learned in the learning phase. On the other hand, existing techniques often have low performance when confronted with complex situations since unfamiliar objects are not included in the training dataset. Additionally, the use of a large amount of labeled data is generally essential for training deep neural networks to achieve good performance, which is time-consuming and labor-intensive. Thus, this paper presents a solution to these issues by proposing a self-supervised learning method for the drivable areas and road anomaly segmentation. First, we propose the Automatic Generating Segmentation Label (AGSL) framework, which is an efficient system automatically generating segmentation labels for drivable areas and road anomalies by finding dissimilarities between the input and resynthesized image and localizing obstacles in the disparity map. Then, we train RGB-D datasets with a semantic segmentation network using self-generated ground truth labels derived from our method (AGSL labels) to get the pre-trained model. The results showed that our AGSL achieved high performance in labeling evaluation, and the pre-trained model also obtains certain confidence in real-time segmentation application on mobile robots. MDPI 2022-06-23 /pmc/articles/PMC9268933/ /pubmed/35808244 http://dx.doi.org/10.3390/s22134751 Text en © 2022 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 Nguyen, Toan-Khoa Nguyen, Phuc Thanh-Thien Nguyen, Dai-Dong Kuo, Chung-Hsien Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data |
title | Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data |
title_full | Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data |
title_fullStr | Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data |
title_full_unstemmed | Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data |
title_short | Effective Free-Driving Region Detection for Mobile Robots by Uncertainty Estimation Using RGB-D Data |
title_sort | effective free-driving region detection for mobile robots by uncertainty estimation using rgb-d data |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9268933/ https://www.ncbi.nlm.nih.gov/pubmed/35808244 http://dx.doi.org/10.3390/s22134751 |
work_keys_str_mv | AT nguyentoankhoa effectivefreedrivingregiondetectionformobilerobotsbyuncertaintyestimationusingrgbddata AT nguyenphucthanhthien effectivefreedrivingregiondetectionformobilerobotsbyuncertaintyestimationusingrgbddata AT nguyendaidong effectivefreedrivingregiondetectionformobilerobotsbyuncertaintyestimationusingrgbddata AT kuochunghsien effectivefreedrivingregiondetectionformobilerobotsbyuncertaintyestimationusingrgbddata |