Mobile structural health monitoring using quadruped robots

Abstract

To mitigate infrastructure deterioration, structural health monitoring (SHM) has been employed for more than half a century, gaining momentum with recent advancements in information, communication, and sensing technologies. In particular, wireless sensor networks have gradually been incorporated into SHM, offering new opportunities towards enhanced flexibility and scalability, as compared to cable-based SHM systems. However, wireless sensor nodes are installed at fixed locations and, causing high installation costs, need to be employed at high density to reliably monitor large infrastructure. This feasibility study proposes quadruped robots for wireless SHM of civil infrastructure, leveraging advantages regarding cost-efficiency and maneuverability. Aiming at cost-efficiency, the quadruped robots are implemented using off-the-shelf components. The robots are equipped with sensors to collect acceleration data relevant to SHM of civil infrastructure, with cameras for navigation, and with embedded algorithms, facilitating autonomous data processing, analysis, synchronization, and communication. The accuracy of the quadruped robots is validated in laboratory tests on a shear-frame structure by comparing the SHM data collected and analyzed by the quadruped robots with SHM data collected by a high-precision cable-based SHM system. Furthermore, the maneuverability and efficiency of the quadruped robots is demonstrated through field tests conducted on a road bridge by comparing the sensor information collected by the robots with the respective sensor information collected by a comprehensive benchmark SHM system. The results confirm that the quadruped robots, as compared to stationary wireless sensor nodes, require a smaller number of nodes to achieve the same sensor information and, as compared to wheeled robots, offer better maneuverability, as critical parts of civil infrastructure may be hard to reach. In summary, this feasibility study represents a first step towards robotic fleets employed for autonomous SHM. calculates the movements to be performed, and the movements are sequentially forwarded through the classes ProcessManager , CommunicationManager and Arduino to the robot unit, which performs the movements. The PathDetectionLocator class processes image data of the sensing unit, leveraging computer vision, to obtain position data based on the longitudinal centerline of the robot, which represents its reference position.