Unknown hostile environment-oriented autonomous WSN deployment using a mobile robot

Publisher:
Elsevier BV
Publication Type:
Journal Article
Citation:
Journal of Network and Computer Applications, 2021, 182, pp. 103053-103053
Issue Date:
2021-05-15
Filename Description Size
Unknown hostile environment.pdfPublished version6.27 MB
Adobe PDF
Full metadata record
In this study, we consider the Internet of Things (IoT) with an autonomous deployment framework and seek optimal localizable k-coverage (OLKC) strategies to preserve the connectivity and robustness in IoT networks to assist robots during disaster recovery activities. Therefore, we define localizable k-coverage as the covered region within which a mobile robot can localize itself aided by k neighboring beacon nodes (BNs) in a wireless sensor network (WSN). To this end, we first propose the optimal localizable k-coverage WSN deployment problem (OLKWDP) and present a novel framework that preserves WSN connectivity and robustness for mobile robots. To localize a mobile robot with at least k BNs and overcome the network hole problem that can occur in unknown hostile environments, we propose a hole recovery method for the OLKC achieved by a mobile robot that knows the concurrent mapping, deployment and localization of the WSN. We then present a mapping-to-image transformation method to reveal the interactions between the WSN deployment and the network holes for the OLKC while constructing the online mapping. To solve the OLKWDP, we also develop two optimality conditions to achieve maximum coverage by the proposed OLKC in the unknown hostile environment using the minimum number of sensors. Moreover, we analyze the factors that influence the probability of success of the OLKC and the factors that influence the performance of a mobile robot when determining the WSN deployment. The simulation results illustrate that our framework outperforms the trilateration and spanning tree (TST) method in unknown hostile environment exploration and can achieve the OLKC in a WSN. In 27 simulated situations, our framework achieved average rates of nearly 100% 1-coverage, 91.34% 2-coverage and 89.00% 3-coverage.
Please use this identifier to cite or link to this item: