A machine learning framework for acoustic reflector mapping

Usama Saqib, Letizia Marchegiani, Jesper Rindom Jensen
{"title":"A machine learning framework for acoustic reflector mapping","authors":"Usama Saqib, Letizia Marchegiani, Jesper Rindom Jensen","doi":"arxiv-2409.12094","DOIUrl":null,"url":null,"abstract":"Sonar-based indoor mapping systems have been widely employed in robotics for\nseveral decades. While such systems are still the mainstream in underwater and\npipe inspection settings, the vulnerability to noise reduced, over time, their\ngeneral widespread usage in favour of other modalities(\\textit{e.g.}, cameras,\nlidars), whose technologies were encountering, instead, extraordinary\nadvancements. Nevertheless, mapping physical environments using acoustic\nsignals and echolocation can bring significant benefits to robot navigation in\nadverse scenarios, thanks to their complementary characteristics compared to\nother sensors. Cameras and lidars, indeed, struggle in harsh weather\nconditions, when dealing with lack of illumination, or with non-reflective\nwalls. Yet, for acoustic sensors to be able to generate accurate maps, noise\nhas to be properly and effectively handled. Traditional signal processing\ntechniques are not always a solution in those cases. In this paper, we propose\na framework where machine learning is exploited to aid more traditional signal\nprocessing methods to cope with background noise, by removing outliers and\nartefacts from the generated maps using acoustic sensors. Our goal is to\ndemonstrate that the performance of traditional echolocation mapping techniques\ncan be greatly enhanced, even in particularly noisy conditions, facilitating\nthe employment of acoustic sensors in state-of-the-art multi-modal robot\nnavigation systems. Our simulated evaluation demonstrates that the system can\nreliably operate at an SNR of $-10$dB. Moreover, we also show that the proposed\nmethod is capable of operating in different reverberate environments. In this\npaper, we also use the proposed method to map the outline of a simulated room\nusing a robotic platform.","PeriodicalId":501031,"journal":{"name":"arXiv - CS - Robotics","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Robotics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.12094","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Sonar-based indoor mapping systems have been widely employed in robotics for several decades. While such systems are still the mainstream in underwater and pipe inspection settings, the vulnerability to noise reduced, over time, their general widespread usage in favour of other modalities(\textit{e.g.}, cameras, lidars), whose technologies were encountering, instead, extraordinary advancements. Nevertheless, mapping physical environments using acoustic signals and echolocation can bring significant benefits to robot navigation in adverse scenarios, thanks to their complementary characteristics compared to other sensors. Cameras and lidars, indeed, struggle in harsh weather conditions, when dealing with lack of illumination, or with non-reflective walls. Yet, for acoustic sensors to be able to generate accurate maps, noise has to be properly and effectively handled. Traditional signal processing techniques are not always a solution in those cases. In this paper, we propose a framework where machine learning is exploited to aid more traditional signal processing methods to cope with background noise, by removing outliers and artefacts from the generated maps using acoustic sensors. Our goal is to demonstrate that the performance of traditional echolocation mapping techniques can be greatly enhanced, even in particularly noisy conditions, facilitating the employment of acoustic sensors in state-of-the-art multi-modal robot navigation systems. Our simulated evaluation demonstrates that the system can reliably operate at an SNR of $-10$dB. Moreover, we also show that the proposed method is capable of operating in different reverberate environments. In this paper, we also use the proposed method to map the outline of a simulated room using a robotic platform.
声学反射体绘图的机器学习框架
几十年来,基于声纳的室内测绘系统一直被广泛应用于机器人领域。虽然这类系统在水下和管道检测环境中仍是主流,但随着时间的推移,其易受噪声影响的特性降低了其广泛使用的普遍性,转而使用其他模式(如文本、相机、激光雷达),而这些模式的技术正在取得非凡的进步。不过,由于声学信号和回声定位与其他传感器相比具有互补性,因此利用它们绘制物理环境地图可以为机器人在逆境中导航带来显著优势。相机和激光雷达在恶劣的天气条件下、在缺乏光照或面对非反射墙壁时,确实很难发挥作用。然而,声学传感器要想生成准确的地图,就必须妥善有效地处理噪声。在这种情况下,传统的信号处理技术并不总能解决问题。在本文中,我们提出了一个框架,利用机器学习来帮助更传统的信号处理方法处理背景噪声,从声学传感器生成的地图中去除异常值和伪影。我们的目标是证明传统回声定位绘图技术的性能即使在噪声特别大的情况下也能得到极大提升,从而促进声学传感器在最先进的多模式机器人导航系统中的应用。我们的模拟评估表明,该系统可以在信噪比为 10 美元分贝的条件下可靠运行。此外,我们还证明了所提出的方法能够在不同的混响环境中工作。在本文中,我们还使用所提出的方法,利用机器人平台绘制了一个模拟房间的轮廓图。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信