Computer vision-based unmanned aerial vehicle precise landing system

Abstract

The invention discloses a computer vision-based unmanned aerial vehicle precise landing system. The computer vision-based unmanned aerial vehicle precise landing system comprises an optical system, acomputer navigation processing system and a flight control system, wherein the optical system is used for acquiring image information of a landing area and transferring the image information to the computer navigation processing system; the computer navigation processing system is used for calculating the image information, outputting the position and posture of an unmanned aerial vehicle relativeto a landing target and transferring the position and posture to the flight control system; the flight control system is used for calculating an expected flight speed vector through relative positions and postures of the unmanned aerial vehicle and the landing target, outputting a corresponding motor speed controller driving voltage vector, and carrying out posture adjustment on a cloud deck through a deflection angle between an optical axis of the optical system and the center of the landing target; a propeller rotation speed combination changes the position and posture of the unmanned aerial vehicle, so as to influence the position and posture of the optical system; the change of the angle of the cloud deck influences the posture of the optical system; and a field of view of the opticalsystem is influenced by the propeller rotation speed combination and the angle of the cloud deck, so that image information updating is provided for the computer navigation processing system and theunmanned aerial vehicle is assisted to complete autonomous landing.

Description

technical field [0001] The invention relates to the technical field of drone landing control systems, in particular to a computer vision-based precise landing system for drones. Background technique [0002] With the rapid development of drone technology, more and more drones have entered people's lives. UAVs are available day and night, simple in structure, easy to use, low in cost, high in efficiency, and do not have to worry about casualties. Therefore, UAV operations are increasingly favored in high-level environments. It can be used for scene monitoring, meteorological investigation, highway inspection, surveying and mapping, flood monitoring, aerial photography, traffic management, forest fire, etc., and has extremely broad application prospects. [0003] During the mission of the UAV, the recovery process is a very important and prone to failure stage. The navigation system needs to accurately control the attitude and trajectory of the aircraft. Realizing the automa...

AI Technical Summary

Problems solved by technology

[0004] At present, the navigation technologies used for automatic landing of drones mainly include inertial navigation, global positioning system navigation, differential global positioning system navigation and other technologies, but the error in inertial navigation diverges with time, so it cannot be used independently. System navigation relies entirely on navigation satellites, and in the case of interference, it is easy to lose the signal