Design of a nonlinear TACAN bearing tracking servo using the extended Kalman filter

An analytical and simulation study was used to develop a near-real-time optimum digital computer TACAN bearing tracking algorithm. An extended Kalman filter estimator was utilized to determine a tracking loop configuration and its expected performance. Examination of the filter covariance matrix and gains revealed that a few key elements of the covariance matrix were the significant contributors to the filter gains. This observation suggested that a set of time-switched constant gains multiplied by the partial derivatives of the observation with respect to the state would provide adequate performance with significant real-time processing reduction. Fortran computer simulations verified the concept. The algorithm was implemented on a microcomputer, and the system was successfully flight-tested