Rotor unmanned aerial vehicle automatic cruise method and system based on deep learning

Abstract

The present invention discloses a rotor unmanned aerial vehicle automatic cruise method and system based on deep learning. The method can avoid barriers along the line based on the deep learning, namely add eyes for the unmanned aerial vehicle to assist in the ultrasonic ranging method and timely avoid the barriers in the close range. The system comprises an image collection module, a barrier identification module, an ultrasonic wave distance detection module, a flight control module and a location module based on a Beidou satellite. The rotor unmanned aerial vehicle automatic cruise method and system based on deep learning are especially suitable for the logistics system and can realize the point-to-point unmanned aerial vehicle automatic cargo transport function so as to greatly save the manpower.

Description

technical field [0001] The invention relates to a deep learning-based automatic cruising method and system for a rotor unmanned aerial vehicle, belonging to the field of unmanned aerial vehicle automatic cruising. Background technique [0002] UAV automatic cruising refers to a system in which the UAV automatically plans the route and finds the destination under the condition of a given destination. The planning of the flight route of the UAV uses the visual sensing technology based on deep learning to identify objects that remain unchanged for a long time during the navigation process, and after multiple trainings, the automatic cruising of the UAV is realized. [0003] The invention patent of "UAV obstacle avoidance system and control method based on ultrasonic distance detection" is to use ultrasonic detection technology to detect obstacles in the flying environment of UAVs, and cooperate with specific obstacle avoidance algorithms to achieve the purpose of obstacle avoid...

AI Technical Summary

Problems solved by technology

The obstacle avoidance method adopted by this system is single. When the UAV is flying in a complex and dense environment, the ultrasonic signal will be weakened due to the multiple reflections of the ultrasonic wave, making the detection inaccurate and the obstacle avoidance failure.

[0004] The invention patent of "A Method and Device for Obstacle Avoidance of UAV Based on 3D Camera" uses a 3D camera to collect image data in real time, and sends the data to the visual analysis module for calculation and analysis to achieve the purpose of obstacle avoidance, but in this invention The planning of the flight route of the UAV depends on the external navigation system, and the automatic cruise function of the UAV is not realized.

Images