Optimal estimation of missile free-flight trajectory: Comparative results of linear and nonlinear Kalman filter approaches

This is a technical report on the application of optimal estimation theory for estimating the trajectory of a free-flight missile, as a means of evaluating missile performance. Several approaches were investigated (linear Kalman filter working in Cartesian Coordinates and nonlinear extended Kalman filter implemented in Cartesian Coordinates). A six-degree of freedom simulation was used to generate the free-flight trajectory of the missile. A simulated radar, positioned at the origin of the inertial ground reference frame, generated range, azimuth, and elevation observation data. An analysis was conducted in the following three areas: (1) initialization, (2) Q matrix weighting, and (3) observation of noise variances. The nonlinear filter approach (linear dynamics with nonlinear measurements) proved to exhibit superior performance over a strictly linear filter approach. Further investigations in specific areas, are recommended.