A novel neuro-estimator and it’s application to parameter estimation in a remotely piloted vehicle

Abstract The paper presents a novel neuro-computing approach to the problem of state estimation by means of a hybrid combination of a Hopfield neural network and a feedforward multilayer neural net capable to solve certain optimization problems. This neuro-estimator is very appropriate for the real-time implementation of nonlinear state estimators, especially when the modeling of uncertainty is considered in the problem. The proposed estimator is applied to estimate the aerodynamic parameters of a remotely piloted vehicle. Simulation results show the effectiveness of the proposed method.