Nano type drone target tracking method based on concentric circles

Abstract

The invention provides a nano type drone target tracking method based on concentric circles, and belongs to the field of target tracking technology for drones. The method comprises first installing acamera, a simulated picture transmission and a battery on a nano type drone; setting a black and white concentric circle as the tracking target of the nano drone; after the nano type drone takes off from the ground, the camera captures the image in real time and transmits the image to the ground computer through the analog image transmission; the ground computer uses the target recognition algorithm to identify the received image: if the concentric circle target is successfully identified in the image, the relative pixel distances in the three directions of the concentric circle target and thedrone are respectively calculated. The visual control algorithm is used to obtain the control command of the drone; the drone moves according to the control instructions to achieve tracking of concentric targets. The method can be applied to a nano type drone platform with the target tracking capability and a great application prospect in the national defense military and civilian technology.

Description

technical field [0001] The invention relates to the technical field of UAV target tracking, in particular to a nano-UAV target tracking method based on concentric circle features. Background technique [0002] In recent years, UAV technology has developed rapidly, and with the increasing maturity of computer vision technology, vision-based target tracking technology has gradually become an important means for UAVs to perceive the environment and analyze tasks. In disaster relief, military It has broad application prospects in many fields such as reconnaissance and land survey. [0003] In 2011, D. Eberli et al. used the double concentric circle mark as the target to guide the UAV to land. However, in this method, the target needs to be in a static state, and the UAV cannot automatically land on the dynamic target. In 2015, Li Yongjian from South China University of Technology and others used the ellipse fitting algorithm to control the landing of the drone by allowing the d...

AI Technical Summary

Problems solved by technology

[0003] In 2011, D.Eberli et al. used the double concentric circle mark as the target to guide the UAV to land, but in this method, the target needs to be in a static state, and the UAV cannot automatically land on a dynamic target.

In 2015, Li Yongjian from South China University of Technology and others used the ellipse fitting algorithm to control the landing of the drone by allowing the drone to recognize the concentric circle mark, but this method did not take advantage of the inclusion relationship between the inner and outer circles of the concentric circle mark. As a result, the amount of calculation is increased, which affects the real-time performance of the method, and ultimately fails to apply the method to the UAV platform

In 2016, Fan Long of Harbin Institute of Technology and others used a target containing multiple concentric circles (more than two) to guide the UAV to land, but the accuracy of this method is not ideal, and the concentric circle targets in this method must also be stationary on the ground , failing to track the dynamic target

Due to the very limited processing capacity of the UAV’s onboard processor in this method, the real-time processing is low, and the processed image size is only 60*80 pixels, which increases the difficulty of recognition. The experimental results show that the average tracking error has reached 34cm

In addition, because the color information is used to identify the target, red objects cannot appear in the tracking scene of this method, otherwise it will be misidentified, resulting in tracking failure

At present, due to the limitations of volume, energy consumption, and processing and computing power, it is still a difficult problem to achieve accurate and stable target tracking on the nano-UAV platform.

Images