A sky-ground integrated citrus picking method

Abstract

A method for automatic citrus picking. The method divides the picking into two steps in the air and on the ground: use a high-precision satellite positioning device to determine the exact position of the drone, and use the binocular vision positioning device carried on the drone to monitor each planting area. The information of each citrus tree is collected, and then image processing method is used to obtain the spatial three-dimensional coordinates of each citrus tree trunk and mature citrus. In a short time, the information of all citrus trees and mature citrus in the planting area can be collected and initially positioned. Before picking, the preliminary positioning coordinates are imported and stored in the ground picking system. The ground picking system moves along the interval between rows of fruit trees and stops in front of each citrus tree, driving the picking manipulator to move within the 0.2-meter spatial range of the three-dimensional coordinates of mature citrus. Start the same binocular vision positioning device carried on the manipulator as the drone, and only need to search within a small range (±0.25m) around the three-dimensional coordinates of the stored ripe citrus to find and confirm the ripe citrus. Precise positioning and automatic picking, until all ripe citrus is picked.

Description

technical field [0001] The invention relates to an automatic picking method for citrus, in particular to a method for picking citrus integrated with heaven and earth. Background technique [0002] When manually picking the fruits of fruit trees such as citrus, due to the different shapes of fruit trees growing under natural conditions, the branches are well developed, and the canopy exceeds the height of the human body, the fruits grow unevenly between the branches and the front and back. The efficiency is low, and there is a certain risk. It is the general trend to replace manual picking with automated picking. Automatic picking on the ground requires the use of machine vision positioning. Machine vision is seriously affected by natural light. The angle of sunlight and light intensity are constantly changing during continuous operations. Machine vision is interfered by background light, and often cannot identify and locate citrus, making the picking operation unable to proc...

AI Technical Summary

Problems solved by technology

[0002] When manually picking the fruits of fruit trees such as citrus, due to the different shapes of fruit trees growing under natural conditions, the branches are well developed, and the canopy exceeds the height of the human body, the fruits grow unevenly between the branches and the front and back. The efficiency is low and there is a certain risk. Automatic picking instead of manual picking is the general trend

Automatic ground picking requires the use of machine vision positioning. Machine vision is seriously affected by natural light. The angle of sunlight and light intensity change continuously during continuous operations. Machine vision is interfered by background light, and often cannot identify and locate citrus, making the picking operation unable to proceed smoothly.

UAVs are used to pick citrus. In aerial picking, due to the drift of the aircraft affected by the wind, it is impossible to accurately grab citrus, and the drone stays in the air for a short time, so it cannot continue to work for a long time.

Images