Design of Multi-Sensor Array Payload for UAV Bridge Concrete Surface Crack Detection

Bridges with concrete structures are prone to phenomena such as cracking, which poses a significant threat to railway operational safety. Currently, unmanned aerial vehicles (UAV) has been employed for bridge health monitoring. However, the existing UAV payloads are not specifically designed for bridge inspection and encounter several issues during operation. In light of this, this paper proposes a low-power, omni-directional vision payload specifically designed for bridge inspection. The payload is equipped with four cameras, a laser rangefinder, a millimeter-wave radar, an RTK (Real-Time Kinematic), inertial navigation system, and a high-performance computing platform. It enables real-time acquisition, detection, and annotation of crack information on the sides and underside of bridges, while offering an extended flight time. Moreover, experimental results have demonstrated that the payload can discern targets with a width of 0.19 mm at a distance of 20 m from the bridge.