A comparison of tracking algorithms for supermaneuverable targets

Modern aircraft are capable of maneuvers exceeding those possible by purely aerodynamic design. This capability, called supermaneuverability, includes rapid changes in acceleration and high-G turns that are not feasible from traditional aircraft. Furthermore, newer aircraft often have a low radar cross-section (RCS) profile and/or RCS which varies rapidly with look angle. This paper summarizes the results of a performance evaluation of several conventional and emerging tracking approaches for supermaneuverable targets. The algorithms have been evaluated with respect to target maneuverability, along the continuum of mild maneuvers to supermaneuvers and as a function of target RCS. From this analysis, we draw the following broad conclusions: For high-SNR stable RCS targets exhibiting modest maneuvering, conventional Kalman filter (KF) trackers work well and are computationally appealing. For modern targets that exhibit a low RCS profile, high scintillation, and/or high maneuverability, KF approaches fail and more sophisticated approaches are required. While particle filter (PF) trackers are more computationally demanding than the KF, they provide a tracking capability not achievable by KF methods. This paper provides several case studies that backup these conclusions.