Abstract
Unmanned aerial vehicles (UAVs) play a significant role in the aviation industry nowadays. Their portability and lower cost compared to traditional meteorological towers mean that their use is gaining momentum in many meteorological applications. In particular, UAV-based wind measurements are exploited in atmospheric energy balance research, precision agriculture, climate change studies, among others. This work aims to review the state-of-the-art of UAV-based wind measurement techniques by comparing the different working principles and highlighting their main challenges. The analyzed methodologies are divided into two categories: direct wind measurements (using anemometers mounted on UAVs) and indirect wind measurements (using velocity and force balances). Key aspects, such as the use of computational fluid dynamics (CFD) simulations, the most common sensor onboarding strategies, and the set-up of experimental tests in wind tunnels or in the field to validate the wind measurement accuracy, are addressed. Furthermore, novel developments based on machine learning and data filtration techniques for data quality enhancement are detailed. Based on a quantitative analysis of the recent relevant literature on this topic, we can conclude that multirotor UAVs are preferred to fixed-wing UAVs for scientific purposes, with the main challenge being the effect of propeller perturbation in the case of direct method wind measurements. Finally, it is shown that in most of the studies analyzed, sonic anemometers are chosen among all other types of sensors. Alternatively, the simplest version of the indirect method, namely the tilt model, is a common choice.