A dynamic programming method for picking sequence of spherical fruits

Abstract

The invention discloses a dynamic planning method for picking sequence of spherical fruits, and designs a dynamic harvesting sensing method that plans sensing points and picking points together. After the robotic arm picks or perceives each time, the weight is updated through judgment, and the weight update rule of the fruit depth information is added. The dynamic harvest perception picking sequence planning method proposed by the present invention combines the picking perception points with the picking points on the basis of the dynamic programming method, and is compared with the traditional picking sequence planning method based on experience, the global perception method and the batch perception method. Compared with that, the planned picking sequence not only has a short picking path, but also has strong real-time performance, which improves the picking efficiency and picking success rate of the picking robot. The designed dynamic harvest perception picking sequence planning method was improved, and the fruit picking sequence planning based on three-dimensional space was added, which avoided the fruit damage caused by the picking sequence and reduced the fruit picking damage rate.

Description

technical field [0001] The invention belongs to the field of robot fruit picking, in particular to the picking of spherical fruits, and is used in fields such as plant greenhouses and plant factories where the fruit quantity is large and the fruit distribution is uneven, and picking robots are required to independently complete picking tasks. Background technique [0002] With the increase in the scale of plant greenhouses and plant factories and the increase in labor costs, the intelligentization of fruit and vegetable picking and harvesting has become an urgent demand for agriculture in various countries. With technological breakthroughs in the fields of computer and image processing, fruit and vegetable picking robots have developed rapidly. For example, the citrus picking robot developed by the University of Florida in the United States, the bell pepper picking robot developed by Ben Ha-Gurion University of the Negev in Israel and applied to commercial greenhouses, and t...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the problem of low picking efficiency and easy fruit damage in the currently commonly used picking method based on experience from top to bottom and from left to right, the present invention improves the overall fruit harvesting speed, improves the success rate of fruit picking, and reduces fruit damage. From the perspective of efficiency, a dynamic programming method for the picking sequence of spherical fruits is proposed.

Images