Unmanned aerial vehicle obstacle avoidance method and device based on 3D camera

Abstract

Provided are an unmanned aerial vehicle (UAV) obstacle avoidance method and device based on a 3D camera. The device comprises a 3D camera, a visual analysis module, a flight control main module and an inertial navigation module. The 3D camera is responsible for collecting image data in real time; collected image data and the UAV attitude information provided by the inertial navigation module are sent to the visual analysis module for analysis through an algorithm, and an analysis result is timely fed back to the flight control main module; the flight control main module adjusts flight parameters according to flight line adjusting parameters. The obstacle avoidance technology is based on the passive image analysis technology, and thereby the UAV obstacle avoidance device has the advantages of small size, light weight, low cost, small consumption, rapid response and high success rate, and can be applied to unmanned helicopters, and various rotor wing and fixed wing UAVs.

Description

Technical field: [0001] The present invention relates to the field of intelligent aircraft systems, in particular to a 3D camera-based unmanned aerial vehicle obstacle avoidance technology, which can be applied to automatic navigation and obstacle avoidance of various unmanned aerial vehicles. Background technique: [0002] The research of Unmanned Aerial Vehicles (UAV) has made great progress in recent years, and its application fields are becoming wider and wider. It can be used for communication relay, meteorological detection, disaster monitoring, pesticide spraying, geological survey, map surveying, Traffic control, border control and many other fields. [0003] In the field of aviation navigation applications, a set of navigation control points (waypoints) of the UAV are usually given offline, and the aircraft flies straight between the waypoints. The short-range route planning between the waypoints needs to consider the visibility between the planning points. Will it...

AI Technical Summary

Problems solved by technology

Ultrasonic and laser radar are both technologies for actively detecting obstacles. The advantage is that the technology is relatively mature and the detection distance accuracy is high; the disadvantage is that the load is large, the power consumption is high, and there is a detection blind spot, which requires multiple probes to perform range scanning.

However, visual obstacle avoidance is mainly based on the analysis of 2D images, and the distance information of obstacles cannot be known, or the image is sent back to the ground station for algorithm analysis, and the real-time performance of obstacle avoidance is poor.

Images