Unmanned aerial vehicles (UAVs) hold significant potential for various indoor applications, such as mapping, surveillance, navigation, and search and rescue operations. However, indoor positioning is a significant challenge for UAVs, owing to the lack of GPS signals and the complexity of indoor environments. Therefore, this study was aimed at developing a Wi-Fi-based three-dimensional (3D) indoor positioning scheme tailored to time-varying environments, involving human movement and uncertainties in the states of wireless devices. Specifically, we established an innovative 3D indoor positioning system to meet the localisation demands of UAVs in indoor environments. A 3D indoor positioning database was developed using a deep-learning classifier, enabling 3D indoor positioning through Wi-Fi technology. Additionally, through a pioneering integration of fingerprint recognition into wireless positioning technology, we enhanced the precision and reliability of indoor positioning through a detailed analysis and learning process of Wi-Fi signal features. Two test cases (Cases 1 and 2) were designed with positioning height intervals of 0.5 m and 0.8 m, respectively, corresponding to the height of the test scene for positioning simulation and testing. With an error margin of 4 m, the simulation accuracies for the ( X ,   Y ) dimension reached 94.08% (Case 1) and 94.95% (Case 2). When the error margin was 0 m, the highest simulation accuracies for the H dimension were 91.84% (Case 1) and 93.61% (Case 2). Moreover, 40 real-time positioning experiments were conducted in the ( X ,   Y ,   H ) dimension. In Case 1, the average positioning success rates were 50.8% (Margin-0), 72.9% (Margin-1), and 81.4% (Margin-2), and the corresponding values for Case 2 were 52.4%, 74.5%, and 82.8%, respectively. The results demonstrated that the proposed method can facilitate 3D indoor positioning based only on Wi-Fi technologies.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Three-Dimensional Indoor Positioning Scheme for Drone with Fingerprint-Based Deep-Learning Classifier


    Beteiligte:
    Shuzhi Liu (Autor:in) / Houjin Lu (Autor:in) / Seung-Hoon Hwang (Autor:in)


    Erscheinungsdatum :

    2024




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Unbekannt




    PORTABLE MULTI DRONE SOCCER STADIUM AND MULTI DRONE SOCCER SYSTEM BASED INDOOR POSITIONING

    WI JIN AH / CHOI MIN HYEOK / HWANG JEONG HYEON | Europäisches Patentamt | 2023

    Freier Zugriff

    RSSI-Based Fingerprint Positioning System for Indoor Wireless Network

    Yang, R. / Zhang, H. | British Library Conference Proceedings | 2014


    RSSI-Based Fingerprint Positioning System for Indoor Wireless Network

    Yang, Ruohan / Zhang, Hao | Springer Verlag | 2014


    DRONE-ASSISTED COMMISSIONING OF INDOOR POSITIONING SYSTEM FOR LIGHT FIXTURES

    EPP ERIC / WILSON CHRISTOPHER / KUENZLER GIENN | Europäisches Patentamt | 2020

    Freier Zugriff

    Warehouse Drone: Indoor Positioning and Product Counter with Virtual Fiducial Markers

    Murat Ekici / Ahmet Çağdaş Seçkin / Ahmet Özek et al. | DOAJ | 2022

    Freier Zugriff