A Distributed Visual Active Sensing Method and Its Application of String Fruits

Abstract

The invention discloses a distributed visual active perception method for string-shaped fruits and an application thereof. The present invention adopts a large field of view camera to obtain a large field of view image of a fruit tree to sense a large field of view target, obtain the picking priority of each string-shaped fruit target, and determine each picking preparation point for secondary perception; then drive the manipulator to reach a After picking the preparatory point, the small field of view camera follows the movement of the manipulator to perceive and observe the string-shaped fruits at close range, and obtain dynamic small field of view images to realize active perception of small field of view targets and intelligent decision-making for picking. The invention can reduce the technical problems of insufficient decision-making of picking operation information and redundant operation activities of manipulators, and effectively improve the picking efficiency of string-shaped fruits.

Description

technical field [0001] The invention relates to the field of fruit intelligent picking, in particular to a distributed visual active sensing method for string-shaped fruits and an application thereof. Background technique [0002] my country is the main producer of string fruits such as longan, grape, litchi, etc. The total planting and production volume ranks among the top in the world, and it has huge economic potential. But at present, the harvest of these string fruits basically depends on manual picking, which is labor-intensive and expensive for picking. The manual picking process involves judging the ripeness and quality of the fruit with the naked eye, and then combining tools for manual picking. Studies have shown that during the entire fruit planting and production process, the labor spent on fruit picking accounts for 35% to 45% of the total fruit planting labor force. With the standardization and scale of orchard development, the cost of manual picking will bec...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the shortcomings existing in the prior art, and provide a distributed visual active perception method for cluster fruits, which can realize the joint cooperation of large field of view target perception and small field of view picking decision-making in the picking scene, and effectively solve the picking operation Insufficient information decision-making and redundant manipulator activities can improve the recognition rate of string-shaped fruits, effectively reduce the uncertainty of picking occluded or overlapping fruits, and improve the picking efficiency of string-shaped fruits

Images