Unmanned aerial vehicle cluster control method based on machine vision

Abstract

The invention discloses an unmanned aerial vehicle cluster control method based on machine vision. The method comprises the following steps of: S1, performing formation flight on an unmanned aerial vehicle cluster; S2, establishing unmanned aerial vehicle cluster resource sharing: S21, establishing a real-time data resource sharing mechanism; S22, establishing a service resource sharing mechanism;S23, constructing a cluster unmanned aerial vehicle data backup and recovery mechanism; S24, establishing a cluster cooperative fault detection and processing mechanism; S3, establishing an unmannedaerial vehicle cluster layered anti-collision system: S31, constructing a layered model; S32, detecting layered conflict; S33, conducting layered collision avoidance; S4, controlling the unmanned aerial vehicle cluster; and S5, performing precise landing based on machine vision. Before the task is executed, training of a deep learning algorithm SSD is carried out by utilizing the collected targetdata, a target platform can be ensured to be detected in a video containing the target platform, and the unmanned aerial vehicle cluster can be assisted to precisely land on the specified platform.

Description

technical field [0001] The invention belongs to the field of unmanned aerial vehicle clusters, and in particular relates to a control method for unmanned aerial vehicle clusters based on machine vision. Background technique [0002] Unmanned aircraft, referred to as unmanned aerial vehicle, or UAV in English, is an unmanned aircraft controlled by radio remote control equipment and its own program control device. From a technical point of view, it can be divided into: unmanned helicopter, unmanned fixed-wing aircraft, unmanned multi-rotor aircraft, unmanned airship, unmanned parawing aircraft, etc. [0003] You can refer to the Chinese patent whose authorized announcement number is CN107340784A, which discloses a UAV cluster control method. The method includes the following steps: constructing the internal drone cluster group through the spatial position, speed and acceleration of the members of the drone cluster group. The member model; the acceleration control function of ...

AI Technical Summary

Problems solved by technology

[0004] The above-mentioned UAV swarm control method has strong flexibility in controlling swarm movement, good consistency, and remarkable control and obstacle avoidance effects; but the above-mentioned UAV swarm control method still has some shortcomings, such as: 1. The single anti-collision method in the existing technology cannot simultaneously solve the problem of collision avoidance and obstacle avoidance between individuals in the cluster; two, it is inconvenient to accurately locate the landing position of the drone cluster

Images