Autonomous wireless sensor deployment with unmanned aerial vehicles for structural health monitoring applications. Issue 6 (3rd March 2022)

• Record Type: Journal Article
• Authors: Zhou, Hao; Lynch, Jerome; Zekkos, Dimitrios
• Is Part Of: Structural control and health monitoring. Volume 29:Issue 6(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-03-03
• Subjects: autonomy -- computer vision -- Kalman filter -- mobility -- UAV -- wireless sensor; Structural engineering -- Periodicals; Structural control (Engineering) -- Periodicals; Automatic data collection systems -- Periodicals; Detectors -- Periodicals; 624.17
• DOI: 10.1002/stc.2942
• Languages: English
• ISSNs: 1545-2255

Abstract

Summary: Unmanned aerial vehicles (UAVs) equipped with onboard sensors empower end‐users to collect data within a wide range of civil engineering applications such as structural condition assessment. While UAVs have been used primarily as mobile sensing platforms to obtain imagery or other data, their mobility can also be used to deploy sensor networks. In this study, the feasibility of using an autonomous UAV to deploy wireless sensors in structural monitoring applications is proposed. While GPS‐based waypoint navigation is available for automating UAV flight operations, this approach does not provide the accuracy necessary for the precision placement of sensor payloads on structures. Computer vision‐based pose estimation is proposed to improve the accuracy of UAV localization for sensor placement. Variably sized fiducial markers integrated into a single pattern are applied to the surface of the structure and adopted as navigation and precision landing targets that identify sensor placement locations. Visual and inertial measurements are fused by means of a discrete Kalman filter to further increase the robustness of the relative position estimation algorithm that is included in the proportional‐integral‐derivative (PID) control law used for UAV landing. Outdoor experiments under realistic operational conditions are conducted to validate the proposed vision‐aided control of the UAV for sensor placement; the UAV is able to land on a predefined landing point within 10 cm. A UAV moving a wireless accelerometer to locations on a beam is used to experimentally show the validity of automating UAV sensor placement for modal analysis using reconfigurable sensor network topologies.