A neurocomputing algorithm for linear state estimation

A linearized Hopfield neural network (HNN) is used as a computing tool to solve a continuous-time linear state estimation problem. The estimation problem is treated as a dynamic optimization problem, where the objective is to find the system state that optimizes a performance measure. It is shown that, by appropriate choice of the weights of the HNN, the optimal state can be obtained as the sealed output of an HNN.<
>