In the last decade, there has been a growing interest in Unmanned Aerial Vehicles (UAVs) or drones, leading to significant technological advancements in avionics. Thanks to artificial intelligence and machine learning, UAVs are becoming more autonomous. However, a major challenge is the limited observability, testability, and predictability of their behavior, which can lead to fatal crashes, even involving humans. Current testing approaches involve expensive and unscalable field testing, while simulation-based testing shows promise in being cheaper and safer. However, simulations may not fully mirror real-world performance, leading to a concern called the Reality Gap. The AERIALIST project aims to enable simulation-based testing and test automation for UAVs by addressing the Reality Gap. It investigates runtime monitoring for early detection of misbehaviors, reduces simulation-based testing costs by predicting outcomes without running simulations, and extracts lower-level tests from simulated ones. The project also aims to use logged data from misbehaviors and failing tests to enhance UAV performance in new scenarios.