Nonlinear Estimators for Transfer Alignment/Navigation

The US Air Force has identified a need to quantify the benefits associated with applying nonlinear estimation techniques to air-launched weapon systems equipped with Inertial Measurement Units (IMU). Of particular interest is a new technique recently developed by the Air Force, referred to as the State Dependent Riccati Equation (SDRE) technique. The SDRE technique has the potential to achieve IMU transfer alignment and navigation accuracy comparable to (or better than) conventional nonlinear techniques but with a significantly reduced computational burden. The focus of this proposed research is to: (1) formulate nonlinear transfer alignment and navigation equations for an air-launched weapon grade IMU, (2) develop candidate nonlinear estimation algorithms, with particular emphasis on the SDRE technique, and (3) conduct simulation-based tradeoff analyses to characterize the benefits of the candidate algorithms. Detailed mathematical models will be formulated to characterize the nonlinear mechanization equation and linearized error equation for INS-to-IMU transfer alignment and IMU-based navigation candidate nonlinear estimation algorithms will be identified, with particular emphasis on the SDRE and extended Kalman filter algorithms. The models and algorithms will be incorporated into our existing monte carlo simulation and results will be generated over representative alignment and navigation trajectories. The simulation results for each candidate estimation algorithm will be contrasted, and the most promising technique will be selected for laboratory and van tests to be conducted under the subsequent Phase II effort.