An AEGIS-FISST integrated detection and tracking approach to Space Situational Awareness

Space Situational Awareness (SSA) is composed of three interdependent tasks: discovery of new objects, tracking of detected objects, and characterization of tracked objects. Currently these problems are treated separately and independently of each other, which may result in the non-optimal processing of data, with a corresponding loss of potential information. Given the scarcity of sensors available to perform SSA missions, it is crucial that these resources be used as efficiently as possible. Detection and classification both involve estimation over the space of discrete variables (e.g., existence/nonexistence, satellite mission type), whereas tracking involves estimation over a space of continuously-valued variables (e.g., position and velocity). The current paper uses Finite Set Statistics (FISST) to formulate a hybrid SSA estimation problem, which consists of simultaneously estimating the number of objects and their tracks, in the presence of false alarms, misdetections and sensor noise. The main contribution of the paper is that, in order to reduce the computational burden entailed in FISST, we employ a Gaussian mixture approximation, not to the first-moment (as in GM-PHD) of the full FISST update equations, but apply the approximation directly to the full FISST equations. The specific GM technique we employ is the Adaptive Entropy-based Gaussian-mixture Information Synthesis (AEGIS). The approach is demonstrated on a simplified SSA application example.