WSN and UAV-Assisted Aircraft Structural Health Monitoring During Maintenance
Prashanth Kotur, R Surendran
Abstract
The research study is focused on tracking the structural health during maintenance by incorporating Wireless Sensor Network (WSN) and Unmanned Aerial Vehicle (UAV). WSN embedded in aircraft are strain gauges, accelerometers, and temperature sensors to measure the real-time changes of parameters like stress, strain, and vibration. These are used to forward data from onboard sensors through a low-power long-range communication standard called Long Range (LoRa) to a base station. UAV is equipped with high-definition cameras, thermal infrared cameras, and Light Detection and Ranging (LiDAR) would be modified to inspect remote areas of airplanes. The UAV would be a video and thermal data capture traveling relay, hence relaying the sensor data back to the sensor nodes for processing. Data fusion techniques subsequently combine data from the WSN and UAV for the best aircraft maintenance. The fused data is taken into account by an Attention Mechanism-based Long Short Term Memory (AM-LSTM) model for predictive maintenance. The AM-LSTM model monitors the time-series sensor data to identify the patterns of structure and the need for future maintenance. These models enhance detection as they focus on prominent data features that can show early damage. Hence, proactive damage detection minimizes aircraft downtime and saves expensive repairs. The research proves the possibility of enhancement in the efficiency, safety, and reliability of aircraft maintenance using collaborative application of WSN, UAV, and deep learning, resulting in timely repair and more trustworthy operations.