Angle-Only Tracking in Three Dimensions

The angle-only filtering (AOF) problem in 3D arises in many real-world passive tracking problems. These problems include passive ranging using an infrared search and track (IRST) sensor, passive sonar, passive radar in the presence of jamming, and satellite-to-satellite passive tracking. We consider estimating the state of a target using noisy bearing and elevation measurements. We first describe the dynamic models of the target and senor and different coordinate systems. The filter initialization, which forms an important part of the filtering algorithm, is explained next. Several nonlinear filters including the extended Kalman filter (EKF), unscented Kalman filter (EKF), cubature Kalman filter (CKF), and particle filter (PF) are used for performance evaluation. We also compare the performance of these nonlinear filters with the posterior Cramér-Rao lower bound (PCRLB). Finally, we discuss other topics in the AOF problem not covered in this presentation.