A Neural Network-Based Trajectory Optimization Method for Aircraft's Ascent Section

Abstract

The present invention proposes a trajectory optimization method based on a neural network for the ascent section of an aircraft, which is used to solve the technical problems of poor real-time performance and adaptability in the prior art, including the following steps: Step 1: Establishing an aircraft under the launch inertial system The continuous optimal control problem of the ascending segment; Step 2: Obtain the continuous two-point boundary value problem of the aircraft in the vacuum flight segment in the launch inertial coordinate system; Step 3: Obtain the nominal parameters and non-nominal parameters of the aircraft; Step 4: Calculate the launch inertia Offline solution of the continuous optimal control problem in the ascent segment of the aircraft in the coordinate system; Step 5: Solve the continuous two-point boundary value problem in the vacuum flight segment of the aircraft in the launch inertial coordinate system; Step 6: Construct a neural network and perform offline Training; Step 7, obtaining the trajectory optimization result of the ascent section of the aircraft online.

Description

technical field [0001] The invention belongs to the technical field of aircraft guidance and control, and relates to an aircraft ascending segment trajectory optimization method, in particular to a neural network-based aircraft ascending segment trajectory optimization method. Background technique [0002] For launch vehicles, ballistic missiles, aerospace aircraft and other aircraft that fly through dense atmosphere and enter vacuum flight, the ascent section refers to the working interval from when the aircraft leaves the launch pad to when the main engine stops working, which can be further divided into atmospheric flight section and The terminal accuracy of the trajectory in the vacuum flight stage and the ascent stage directly affects the success or failure of subsequent flight missions. However, the strong nonlinearity of the dynamic system, the uncertainty of the engine model and the complex flight environment, and complex flight constraints such as dynamic pressure, ...

AI Technical Summary

Problems solved by technology

[0007] The object of the present invention is to overcome the defect that above-mentioned prior art exists, proposes a kind of trajectory optimization method based on neural network for the ascent section of the aircraft, for solving the poor technical problems of adaptability and real-time performance existing in the prior art

Images