Development of an atmospheric boundary layer detection system based on a rotary-wing unmanned aerial vehicle.

The article focuses on the development and validation of an atmospheric boundary layer (ABL) meteorological detection system based on a rotary-wing unmanned aerial vehicle (UAV). Using computational fluid dynamics (CFD), the study optimized sensor placement on the UAV and designed a novel radiation shield to reduce temperature measurement errors caused by solar radiation. To further enhance accuracy, temperature error correction models were developed employing random forest and multilayer perceptron (MLP) neural network algorithms trained on CFD-simulated data, with the MLP model showing superior performance. Ground-based and UAV hovering experiments demonstrated that the corrected temperature measurements closely matched reference data, with average discrepancies in temperature, humidity, and pressure of approximately 0.6 °C, 1.6% relative humidity, and 0.77 hPa, respectively, supporting the feasibility of UAV-based ABL meteorological monitoring.