Second-order adaptive spherical target state estimation

A second-order adaptive spherical filter formulation is derived and the resulting nonlinear estimator is applied to a tactical air-to-air missile intercept problem. In particular, a second-order Gaussian filter (SOGF) is used to estimate the target acceleration in the line-of-sight (LOS) coordinate frame. The target acceleration is modeled as a first order Gauss-Markov process. An adaptive filter structure is developed to enable the inverse target acceleration time constant to be estimated in real-time. Additionally, a kinematic constraint is used to improve the estimated target acceleration by using a fictitious measurement to improve the target acceleration estimates. A comparison of the extended Kalman filter, adaptive extended Kalman filter, the SOGF, and the adaptive second-order Gaussian adaptive filter is performed using a simulated air-to-air engagement scenario with a maneuvering target. These results are presented in terms of Monte-Carlo simulations.