6G-enabled Situation-Aware ML-assisted UTM (6G-SAMU)

In Unmanned Aerial System (UAS) Traffic Management (UTM) environment, small UASs generally fly at lower altitudes than large manned aircraft and their transmission power is very limited. They also fly shorter distances and more densely populated in the airspace than their manned counterparts. Therefore, a finer control and more frequent updates are necessary for their traffic control. To cope with the special requirements in UTM, we here explore and compare the potential of three machine learning models to help safe UAS operation, built on the 3GPP ecosystem model, through awareness about weather conditions. The proposed algorithm operates in two steps, wherein the first step it predicts the wind speed that it exploits during the second step to allow or stop a flight depending on the characteristics of the drone. The prediction process also covers two modes: i) premature 24 -hour prediction for an early decision, and ii) One-hour prediction for a refined final decision. Depending on the weather conditions (stable or frequently changing), the proposed algorithm can use either the premature or the hourly mode.