Unmanned aerial vehicle target tracking system and method based on 5G and depth image information

Abstract

The invention relates to an unmanned aerial vehicle target tracking system and method based on 5G and depth image information, and the method comprises the steps: combining the universality of satellite navigation and the perception of a visual graph for a scene through a GPS and depth image information integrated navigation and positioning system; an MPII data set is utilized, improved Face recovery and an improved open-pose algorithm are fused, deep learning training is carried out to obtain an efficient model, and the efficient model is transplanted into an industrial control module to carry out target recognition and attitude estimation in real time; and the acquired target attitude and other data are transmitted in a mode of combining a leading-edge 5G technology with a WiFi spread spectrum technology and the like. Compared with the prior art, the method has the advantages that the defects of inaccurate positioning and high cost of a single transmission mode are avoided, the attitude estimation is accurate, the transmission rate is high and the like.

Description

technical field [0001] The invention relates to the technical field of UAV target tracking, in particular to a UAV target tracking system and method based on 5G and depth image information. Background technique [0002] UAV is an unmanned aircraft, which has the characteristics of simple structure, low cost, good maneuverability, and strong adaptability to complex environments. With the continuous development of high and new technologies such as embedded technology and automation technology, drones are gradually being used in various aspects, such as search and rescue, military reconnaissance and circuit inspection. [0003] At present, in the aspect of UAV tracking, the airborne camera device is the main application method of UAV perception and detection. For example, military combat UAVs lock and track combat targets, and in search and rescue missions, image processing methods are used to determine the targets that need to be rescued. Wait. Traditional target tracking de...

AI Technical Summary

Problems solved by technology

However, while the sensor continues to expand the scope of information search, it also increases the complexity of information and the difficulty of processing information.

[0004] At present, most of the navigation systems of drones are based on satellite navigation. Satellite navigation is generally global and one-way broadcasting, and the cost of transmitting data is relatively high. Under certain conditions, such as when Doppler noise is large, it cannot effectively The attitude information of the aircraft is sent back to the ground, and in the application of human identification, the accuracy rate is not very high. In addition, there are problems such as unreliable transmission links and loss of satellite signals in low altitude or complex environments, making it impossible to locate and navigate.

Images