A posteriori aircraft control signal recovery via sparsity-inducing norm minimization

The inclusion of large-scale data-monitoring and data-storage systems has presented researchers with new methods and opportunities for understanding air traffic systems. Traditionally, the study of air traffic systems has required laborious real-time monitoring of human pilots and controllers for analysis that yields only qualitative descriptions of the subjects´ behavior. Computer systems can rapidly automate many analysis processes on historical data sets covering weeks, months, and even years. This paper takes a step in the direction of data-driven analysis by introducing an off-line algorithm for detecting aircraft trajectory changes in heading, speed, and altitude. The algorithm recovers control signal inputs (e.g. heading changes) via sparsity-inducing norm minimization. The algorithm is able to overcome noise in position measurements to construct reliable estimates of the number of maneuvers and inter-maneuver aircraft states (e.g. ground-speed, heading, and their derivatives).