Aircraft global finite time neural network control method based on switching mechanism

Abstract

The present invention relates to an aircraft global finite time neural network control method based on a switching mechanism, belonging to the field of aircraft control. The problem of aircraft globalfinite time neural network control is solved. The method comprises the steps of: decoupling an aircraft vertical model to a height subsystem and a speed subsystem, employing backstepping control forthe height subsystem, and employing PID control for the speed subsystem. For the height subsystem, a switching mechanism is introduced to achieve switching between an effective approaching intra-areaneural network control and approaching outer-area robust control, and based on the tracking errors and the modeling errors, performing updating of the neural network weight so as to improve the learning performance of the neural network, and giving a robust design scheme on this basis, achieving finite time convergence of the tracking errors of the system. The aircraft global finite time neural network control method ensures that the aircraft neural network control is still worked in the effective approaching area to achieve the global stability of the closed-loop system and ensure the performance demand of the practical engineering application.

Description

technical field [0001] The invention relates to an aircraft control method, in particular to an aircraft global finite time neural network control method based on a switching mechanism, and belongs to the field of aircraft control. Background technique [0002] Facing the new requirements for aircraft technology in the military and civilian fields, the flight envelope of modern aircraft continues to expand. The innovative configuration design of aircraft and the complex flight environment lead to complex nonlinear and strong uncertainty in aircraft dynamics. Neural networks can approximate unknown dynamics and model uncertainties, and are widely used in aircraft control. However, most current methods assume that neural networks can always perform effective approximation in the entire region for controller design, which makes the closed-loop The system can only guarantee semi-global stability, which is difficult to guarantee in practical applications. "Global neural dynamic ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that current aircraft neural network control methods rarely consider whether the neural network approximation is always effective during the control process, the present invention designs a global finite-time neural network control method for aircraft based on a switching mechanism, which uses the switching mechanism to achieve effective approximation The switch between the neural network control in the domain and the robust control outside the effective approximation domain ensures that the neural network works in the effective approximation domain and realizes the global stability of the closed-loop system. Based on the learning performance of the network, a robust design scheme is given, which can realize the finite time convergence of the system tracking error

Images