Robust Fault Tolerant Control Method of Unmanned Aerial Vehicle Attitude Based on Neural Network Observer

Abstract

Based on neural network (Neural Network: Neural Network) technology and command filter inversion method, this application proposes a robust fault-tolerant control system design architecture based on command filter inversion. Firstly, the mathematical model of the NSV attitude control system is given, and on this basis, the uncertainty caused by modeling errors and external disturbances, and the state equation of the NSV attitude control system under control surface failures are considered. Its main design involves two units: one is the design of the auxiliary system, and the other is the design of the controller based on the auxiliary system. The auxiliary system introduces the neural network to ensure the robustness of the auxiliary system, and strictly proves the stability of the closed-loop system through the Lyapunov theorem. And the simulation is carried out on the attitude control system of the corresponding aircraft. The results show that the method proposed in this application can make the uncertain flight control system with external interference have ideal fault-tolerant tracking performance under the damage of the control surface.

Description

technical field [0001] The invention relates to the field of aircraft attitude fault-tolerant control, in particular to an unmanned aerial vehicle attitude robust fault-tolerant control method based on a neural network observer. Background technique [0002] Near space (Near Space) refers to the space area 20km~100km away from the sea level, and current human exploration activities rarely involve this height range. It has the atmospheric stratosphere region (height 20km~55km), the atmospheric mesosphere region (height 55km~85km) and a small part of the warming layer region (height above 85km), of which the region below 60km is the non-ionosphere, and the region above 60km is the The ionosphere, the atmospheric composition in most of its space is a homogeneous atmosphere (the area below about 90km in height). It is precisely because of its unique spatial location that the near space has a unique flight environment and nature, so it has great strategic significance and strate...

AI Technical Summary

Problems solved by technology

However, nonlinear predictive control is still immature in system stability analysis.

Inversion control is a controller design method based on Lyapunov stability theory. It has been widely concerned by researchers since it was proposed in the 1990s. However, it also has three main defects. (1) Differential expansion problem ; (2) requires a strict feedback form, (3) controls the constraint problem

In the flight control system, the defects (1) and (3) are the main obstacles to its practical application, especially the defect (3), if it is not considered in practice, it will cause the accumulation of errors and make the parameter estimation incorrect , causing system instability or even divergence

Images