ADAM: Adaptive Monitoring of Runtime Anomalies in Small Uncrewed Aerial Systems

Abstract

Small Uncrewed Aerial Systems (sUAS), commonly referred to as drones, have become ubiquitous in many domains. Examples range from drones taking part in search-and-rescue operations to drones being used for delivering medical supplies or packages. As sUAS commonly exhibit safety-critical behavior, ensuring their safe operation has become a top priority. Thus, continuous and rigorous monitoring of sUAS at runtime is essential. However, sUAS generate vast amounts of data, for example, multi-variate time series which need to be analyzed to detect potential emerging issues. This poses a significant challenge, due to resource constraints imposed on the onboard computation capabilities of sUAS. To alleviate this problem, we introduce ADAM, a novel adaptive monitoring anomaly detection framework for sUAS. ADAM selectively monitors a subset of data streams, which serve as indicators of anomalous behavior. In the event of a raised alert, ADAM adjusts its monitoring strategy, enabling additional detectors and taking further mitigation actions. We evaluated the effectiveness of ADAM through simulations in Gazebo, analysis of real flight logs taken from sUAS forums, and tests with real drones. Results confirm that ADAM can enhance safety and efficiency of sUAS operations, by dynamically managing anomaly detection, reducing CPU and memory usage by up to 65%.