A 2-D intercept problem using a neural extended Kalman filter and controller modification

The neural extended Kalman filter is a technique that learns unmodeled dynamics while performing state estimation in the feedback loop of a control system. This coupled system performs the standard estimation of the states of the plant while estimating a function to approximate the difference between the given state-coupling function model and the behavior of the true plant dynamics. At each sample step, this new model is added to the existing model to improve the state estimate. The new model can also be used to recalculate the control gains at various intervals either continuously or at some given criteria. This paper investigates the application of the neural extended Kalman filter to a two dimensional intercept problem. Using the intercept control approach of computing an impact point based on a priori knowledge of the trajectory of a target, the neural state estimation is applied when modeling errors or target maneuvers are incorporated into the system. The new trajectory model can be better approximated in flight allowing a closer intercept of the target.