Application of a colored noise Kalman filter to a radio-guided ascent mission
D. JOHNSON · 1969
Abstract : A set of general sequential filter equations is derived for nonlinear system dynamics and a nonlinear observation model, but is obtained with the assumption of a linear estimator. These equations, which are based upon previously developed formulas, include the colored measurement noise statistics and the statistics of nonestimated model parameter errors. Several simulations made with this filter are compared with the simulation results of an estimation procedure constructed with the utilization of the standard white noise assumptions. The difference between the white noise filter results and the colored noise filter results is found to be minimal. Instability occurred when the statistics of the effective exhaust velocity in the acceleration model were not properly accounted for.