Maik Rosenberger, Mirco Andy Eilhauer, Raik Illmann, Martin Richter, Andrei Golomoz, Gunther Notni
{"title":"Learning sensor data fusion, an practical approach","authors":"Maik Rosenberger, Mirco Andy Eilhauer, Raik Illmann, Martin Richter, Andrei Golomoz, Gunther Notni","doi":"10.1016/j.measen.2025.101881","DOIUrl":null,"url":null,"abstract":"<div><div>Industrial measurement tasks often cannot be solved using simple sensor systems alone. To this end, the development of numerous field buses and protocols has attempted to achieve strong networking of various sensors and actuators. Technologies in the context of Industry 4.0 and IoT support these trends in the long term. Just like process measurement variables such as temperature, pressure, force, etc., image processing systems have also found their way into modern systems for process control and monitoring. Understanding these complex IoT systems consisting of typical process measurement variables and image-based variables and using them sensibly for measurement and automation tasks must be taught to students as part of their metrology and IT training. For this purpose, special electronics have been developed that combine selected sensors from process measurement technology with an image sensor. This offers the opportunity to develop a clear and practice-oriented exercise for sensor data fusion topics both for teaching in image processing, such as rotational position correction through sensor fusion of a rotation rate sensor and camera system, and for teaching in process measurement technology, such as calculating the dew point from humidity and temperature. The chosen division of the system into microcomputer architecture for recording the sensors and transmission to an evaluation PC as well as the free selection of possible software tools for the calculation of the sensor information among each other allows different learning scenarios to be developed for the system. This publication also presents the architecture of the system, the connection to an evaluation system and an initial application for sensor data fusion for use in teaching.</div></div>","PeriodicalId":34311,"journal":{"name":"Measurement Sensors","volume":"38 ","pages":"Article 101881"},"PeriodicalIF":0.0000,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Measurement Sensors","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2665917425000753","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Engineering","Score":null,"Total":0}
引用次数: 0
Abstract
Industrial measurement tasks often cannot be solved using simple sensor systems alone. To this end, the development of numerous field buses and protocols has attempted to achieve strong networking of various sensors and actuators. Technologies in the context of Industry 4.0 and IoT support these trends in the long term. Just like process measurement variables such as temperature, pressure, force, etc., image processing systems have also found their way into modern systems for process control and monitoring. Understanding these complex IoT systems consisting of typical process measurement variables and image-based variables and using them sensibly for measurement and automation tasks must be taught to students as part of their metrology and IT training. For this purpose, special electronics have been developed that combine selected sensors from process measurement technology with an image sensor. This offers the opportunity to develop a clear and practice-oriented exercise for sensor data fusion topics both for teaching in image processing, such as rotational position correction through sensor fusion of a rotation rate sensor and camera system, and for teaching in process measurement technology, such as calculating the dew point from humidity and temperature. The chosen division of the system into microcomputer architecture for recording the sensors and transmission to an evaluation PC as well as the free selection of possible software tools for the calculation of the sensor information among each other allows different learning scenarios to be developed for the system. This publication also presents the architecture of the system, the connection to an evaluation system and an initial application for sensor data fusion for use in teaching.