Aircraft intelligent trajectory planning method based on chaotic model and DNN

Abstract

The invention discloses an aircraft intelligent trajectory planning method based on a chaos model and a DNN, and belongs to the technical field of aircraft trajectory planning. The method comprises the following steps: obtaining a network mapping relation between a flight state quantity and a flight control quantity; converting the process constraint into flight height and heeling angle constraint; generating a final chaotic signal in the guidance instruction; calculating a heeling angle section median meeting task constraints and a heeling angle required in real time; the size of the heeling angle after chaos processing is obtained; determining a heeling angle direction by adopting a sight angle error corridor; and restraining the amplitude of the heeling angle and the amplitude of the heeling rate. According to the method, the model limitation during the traditional flight path design is overcome, and the unpredictability of the guide instruction and the flight path is increased; and the influence of uncertainty on task reachability is limited. Therefore, the chaos of the flight path is increased on the premise of ensuring the feasibility of the predetermined task, the process constraint is met in the guidance instruction calculation, and the method has a certain engineering practical value.

Description

technical field [0001] The invention relates to an aircraft intelligent trajectory planning method based on a chaotic model and DNN, and belongs to the technical field of aircraft trajectory planning. Background technique [0002] With the comprehensive development of land, sea, air and space detection technology, it is becoming more and more difficult for modern aircraft to achieve stealth flight. The development of new penetration technology under detection conditions is an inevitable trend in the future, and maneuvering flight is the most important and effective one. Penetration means. [0003] On the one hand, the application of artificial intelligence in aircraft is becoming more and more mature. Therefore, the accuracy and cost advantages brought by network learning can be used to predict and analyze the flight status, and provide necessary reference information for penetration decision-making and maneuver flight trajectory planning. [0004] On the other hand, the f...

AI Technical Summary

Problems solved by technology

[0004] On the other hand, the flight trajectory produced by the traditional trajectory planning method has strong regularity, and the existing fuzzy prediction and trajectory prediction technology has been able to achieve high accuracy.

Images