Method for identifying maneuvering state of hypersonic aircraft

Abstract

The invention relates to the technical field of aircraft maneuvering state recognition, in particular to a hypersonic aircraft maneuvering state recognition method. Carrying out maneuvering modeling on the hypersonic flight vehicle; the method comprises the following steps: dividing maneuvering categories of a hypersonic aircraft, constructing a feature parameter set, generating a trajectory library under different initial variable conditions, and taking feature recognition parameter information contained in the trajectory library as training data; establishing a hypersonic aircraft maneuvering state recognition model based on a multi-channel attention mechanism convolution long-short term memory network, and training the model; and collecting feature recognition parameter information, inputting the feature recognition parameter information into the trained model, and outputting a maneuvering state recognition result of the hypersonic aircraft. The problems that a near space hypersonic flight vehicle is high in flight speed, the space range needed by maneuvering state conversion is wide, feature extraction is difficult, and a conventional maneuvering state recognition method is difficult to be directly applied to the hypersonic flight vehicle and recognize the maneuvering state of the hypersonic flight vehicle in a high-precision mode are solved.

Description

technical field [0001] The invention relates to the technical field of aircraft maneuvering state identification, in particular to a method for identifying a maneuvering state of a hypersonic aircraft. Background technique [0002] Hypersonic aircraft refers to weapons with a speed of Mach 5 or more and a flight airspace between 20 and 100 kilometers. With its high maneuverability and large range, it has broken the traditional balance of offense and defense, and has caused subversive effects on the existing air defense defense system. Influence, opened up new fields and new forms of military struggle. The world's military powers are engaged in a fierce arms race around the research and development of hypersonic vehicles. The continuous development of hypersonic vehicles poses a serious threat to the aerospace security of various countries, and the tracking, prediction and defense of hypersonic vehicles has become a research hotspot. The research on the recognition of the m...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that in the prior art, hypersonic aircraft in near space fly fast, the space range required for maneuver state transition is wide, and feature extraction is difficult. The problem of maneuvering state provides a hypersonic vehicle maneuvering state identification method, using deep learning theory to study the hypersonic vehicle maneuvering state identification, it first establishes a hypersonic vehicle maneuvering model, analyzes the maneuvering control parameters, according to the high The horizontal and vertical maneuvering characteristics of supersonic aircraft are divided into maneuver categories. Then, the conversion steps from radar observations to feature recognition parameters are deduced to realize the extraction of feature parameters. Finally, the convolution length and short-term Online identification of hypersonic vehicle maneuvering state by memory network

Images