Neural network-based control with disturbance estimation using Kalman filter for missile longitudinal channel autopilot

Nasrollahi, Saeed; Amuei, Mostafa

doi:10.22034/joae.2025.456173.1227

Neural network-based control with disturbance estimation using Kalman filter for missile longitudinal channel autopilot

Document Type : Original Article

Authors

Saeed Nasrollahi ¹

Mostafa Amuei ²

¹ Faculty of Electrical and Computer, Malek-Ashtar University of Technology, Iran.

² Faculty of Electrical & Computer Engineering , Malek ashtar University of Technology, Iran

10.22034/joae.2025.456173.1227

Abstract

In this paper, the design and simulation of the longitudinal channel missile autopilot based on the neural network and disturbance estimation using Kalman filter is presented. It is assumed that the dynamics of this disturbance is equal to zero and its value is constant over time. After the disturbance compensation, a neural network based controller is used for the longitudinal channel autopilot of the missile. One of the advantages of the proposed method is the use of the Kalman filter to estimate the disturbance, which is an optimal estimator. Also, the neural network is used as an autopilot and there is no need to design a controller.

Keywords

Autopilot

Kalman Filter

Neural network

Disturbance estimation

Equivalent disturbance

Subjects