Wavelet Neural Network Based on SSUKF and its Applications in Aerodynamic Force Modeling for Flight Vehicle

Abstract

To overcome the shortcomings of traditional Wavelet Neural Network (WNN), a WNN algorithm based on modified Unscented Kalman Filter (UKF) is proposed. The algorithm uses a UKF based on Spherical Simplex sigma-point (SSUKF) to estimate the WNN parameters, which can improve the learning capability of WNN. The aerodynamic force modeling experiment for flight vehicle indicate that, compared with BP, EKF and UKF, SSUKF for the WNN training has a better ability with features of convergence, precision and calculation, and is also a good method for aerodynamic force modeling for flight vehicle.