A Kalman filter algorithm for terminal-area navigation with sensors of moderate accuracy

This paper considers translational state estimation in terminal-area operations, using a set of position, air data, and acceleration sensors which are commonly available or considered for IFR and area navigation operations. Kalman filtering is applied to obtain maximum estimation accuracy from the sensors. Feasibility of the Kalman filter in real-time computations requires a variety of approximation s and devices aimed at minimizing the required computation time with negligible sacrifice of accuracy. The principal investigative tool is a simulation of the system. This is used to settle design issues regarding the selection of states, filter partitioning, and multirate execution of the computations; and to determine accuracy behavior for 14 states throughout the terminal area and its relation to sensor accuracy.