{"title":"Reinforcement Learning-Based Autonomous UAV Navigation for CO Source Localization","authors":"Pritika Marik;Harshit Kumar Sahu;Chiranjib Ghosh;Amit Ruidas;Soumajit Pramanik;Avishek Adhikary","doi":"10.1109/LSENS.2025.3596388","DOIUrl":null,"url":null,"abstract":"Online monitoring of carbon monoxide (CO) levels in urban and industrial areas may reduce the rising death toll from CO poisoning. A UAV-based gas sensing provides a dynamic solution to this problem by quickly locating the source through proper tracking. However, unmanned aerial vehicle (UAV) has a limited flight time; thus, an optimized search ensuring fast tracking of the source is crucial. In this letter, we propose a particle clustering deep Q-learning-based framework for autonomous localization of gas source using a UAV. The UAV structure is customized to mount the gas sensor MQ-9 in such a way that the effect of propeller turbulence is minimized. Besides, a modified Gaussian plum model is designed for augmenting real data for more accurate training. A comparison with the previous model highlights the higher success rate and lower step size achieved by this work.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"9 9","pages":"1-4"},"PeriodicalIF":2.2000,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Sensors Letters","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/11117180/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0
Abstract
Online monitoring of carbon monoxide (CO) levels in urban and industrial areas may reduce the rising death toll from CO poisoning. A UAV-based gas sensing provides a dynamic solution to this problem by quickly locating the source through proper tracking. However, unmanned aerial vehicle (UAV) has a limited flight time; thus, an optimized search ensuring fast tracking of the source is crucial. In this letter, we propose a particle clustering deep Q-learning-based framework for autonomous localization of gas source using a UAV. The UAV structure is customized to mount the gas sensor MQ-9 in such a way that the effect of propeller turbulence is minimized. Besides, a modified Gaussian plum model is designed for augmenting real data for more accurate training. A comparison with the previous model highlights the higher success rate and lower step size achieved by this work.