Tea picking robot and fresh leaf grading mechanism thereof

Abstract

The application discloses a tea picking robot and a fresh leaf grading mechanism thereof, and belongs to the field of tea picking robots. The fresh leaf grading mechanism comprises a negative pressure adsorption pipeline assembly of the tea picking robot and a collector arranged below a discharge port of the negative pressure adsorption pipeline assembly. The collector comprises a flat container with an open top. The flat container is provided with symmetrically arranged slope type conveying belts at the top end. The two sets of slope type conveying belts jointly enclose a V-shaped conveying channel. The narrowest part of the V-shaped conveying channel is equal to the width of the inner cavity of the flat container. The tea picking robot adopting the grading mechanism is used for grading fresh tea leaves just picked in a tea picking process. The application can quickly and efficiently grade fresh leaves in a simple way in the picking process, improves the single tooth acquisition efficiency, and can quickly and intuitively evaluate the quality of the single tooth.

Description

Technical Field

[0001] This invention relates to tea-picking robots, specifically to a tea-picking robot and its fresh leaf grading mechanism. Background Technology

[0002] Currently, some intelligent tea-picking robots have been applied in standardized tea plantations. More advanced examples include tea-picking robots that integrate artificial intelligence deep learning, machine vision positioning, and robotic end-effectors. For instance, the existing literature CN115280970B discloses a negative pressure adsorption vertical lifting human-hand-like tea-picking device and method, which includes a direct-acting component, a negative pressure adsorption pipe assembly, and a picking mechanism. The direct-acting component pushes the picking mechanism to slide on the negative pressure adsorption pipe, and the picking mechanism clamps the tea leaves. The upward lifting action of the picking mechanism is achieved through the cooperation of gears and racks one and two, thus picking the tea leaves. A negative pressure field is created at the inlet of the negative pressure pipe, causing the picked tea leaves to be sucked into the pipe for collection. The tea leaves are picked by simulating the two main picking actions of human hand picking: "pinching" and "pulling."

[0003] Tea grades are generally divided into ten levels based on the freshness and tenderness of the tea leaves: Special Grade, Grade 1, Grade 2, Grade 3, and so on. Some Special Grade teas are further subdivided into Grade 1, Grade 2, and Grade 3 based on the shape of the tea buds (Lotus Heart (single bud or single bud), Flag Spear (one bud and one leaf just unfolding, the bud is larger than the spear)). Currently, before using Special Grade fresh leaves to prepare high-end tea, single buds are usually picked manually or graded using a tea sieving machine. However, both manual picking and grading by tea sieving machine have the problem of low efficiency in obtaining single buds. More importantly, it is difficult to quickly and intuitively assess the quality of single buds, making it inconvenient to quickly further screen them (in the preparation of high-end tea, taking a traditional drum sieve as an example, after separating single buds with the help of the through holes of the drum, a lot of manpower and time are still needed to further screen single buds within the standard length range. Each 10,000 single buds requires about 40 minutes per person for further screening, and it is easy to cause visual fatigue). In addition, the high-frequency impact of the tea sieving machine can also affect the taste of the tea. In addition, although the aforementioned tea-picking robots can efficiently pick fresh leaves of different grades, they cannot quickly and efficiently grade the fresh leaves in a simple way during the picking process, nor are they conducive to quickly and intuitively assessing the quality of individual leaves. Summary of the Invention

[0004] In view of the technical problems mentioned in the background art, the purpose of this invention is to provide a tea-picking robot and its fresh leaf grading mechanism.

[0005] The present invention adopts the following technical solution.

[0006] A fresh leaf grading mechanism for a tea-picking robot includes a negative pressure adsorption pipe assembly and a collector installed below the outlet of the negative pressure adsorption pipe assembly. The collector includes a flat container with an open top. The width of the inner cavity of the flat container is 1.2 to 1.3 times the diameter of a single tooth. Two symmetrically arranged inclined conveyor belts are located at the top of the flat container. A cylindrical container is positioned below one end of each inclined conveyor belt. The two inclined conveyor belts together form a V-shaped transmission channel, the narrowest part of which is equal to the width of the inner cavity of the flat container. Single teeth falling from the outlet of the negative pressure adsorption pipe assembly first enter the V-shaped transmission channel and then fall directly into the inner cavity of the flat container. Non-single teeth falling from the outlet of the negative pressure adsorption pipe assembly enter the V-shaped transmission channel and are then laterally conveyed to the cylindrical container. This design enables rapid and efficient grading of fresh leaves during the tea-picking process.

[0007] To enable a quick and intuitive assessment of single-tooth quality, the flat container is made of plexiglass with a wall thickness of no more than 5 mm.

[0008] Furthermore, several flat containers are arranged side by side below the inclined conveyor belt. All flat containers are movably clamped in the strip groove. An intermittent pushing part is provided behind the flat container on the tail side. When a single flat container is filled with a single tooth, the intermittent pushing part pushes the flat container forward by a standard distance, so that an empty flat container moves to the bottom of the V-shaped conveyor channel.

[0009] Furthermore, the intermittent feeding section includes a screw drive structure located on the side of the strip groove. The movable seat of the screw drive structure abuts against the rear end of the flat container on the tail side. The screw drive structure is connected to the control system, and the operation of the screw drive structure is controlled by the control system.

[0010] Furthermore, a set of through holes is provided on the upper part of the flat container, and a through-beam photoelectric sensor is provided on the side of the flat container located at the receiving station. The transmitting end and receiving end of the photoelectric sensor are located on both sides of the through hole, respectively. When there is no single tooth blocking the through hole in the flat container, the beam emitted by the transmitting end of the photoelectric sensor can just pass through the through hole and be received by the receiving end.

[0011] Furthermore, the flat container has several ventilation holes.

[0012] A tea-picking robot employing the aforementioned grading mechanism is used to grade freshly picked tea leaves during the tea-picking process.

[0013] To enable faster assessment of single-tooth quality, a camera terminal is installed on the side of the strip groove. The camera terminal is used to acquire images of fresh leaves inside the flat container and upload the images to the control system. The control system includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: S1, reads fresh leaf images from the camera terminal in real time; S2, preprocess the obtained fresh leaf image and extract the single tooth contour from the image; S3, obtain the length of the obtained single tooth contour, and compare the obtained length value with the preset threshold range. If the obtained length value is less than the minimum value of the threshold range, then mark the fresh leaf image as having an unqualified single tooth.

[0014] Beneficial effects: The solution of this invention can not only quickly and efficiently grade fresh leaves in a simple way during the picking process, improving the efficiency of single leaf acquisition, but also quickly and intuitively evaluate the quality of single leaves, making it easy to identify single leaves that do not meet the requirements. Further screening of 10,000 single leaves takes less than five minutes and requires no human intervention, which is conducive to the production of higher standard high-grade tea. More importantly, the single leaves will not be subjected to high-frequency and high-speed impacts during the grading process. Attached Figure Description

[0015] Figure 1 , Figure 2 This is a schematic diagram of the three-dimensional structure of the tea-picking robot in the embodiment; Figure 3 This is a schematic diagram of the main structure of the tea-picking robot in the embodiment; Figure 4 This is a partial schematic diagram of the lead screw transmission structure of the tea-picking robot in the embodiment; Figure 5 This is a schematic diagram of the grading mechanism of the tea-picking robot in the embodiment; Figure 6 This is a schematic diagram showing the location of the flat container of the tea-picking robot in the embodiment. Figure 7 This is a schematic cross-sectional view of the grading mechanism of the tea-picking robot in the embodiment; Figure 8 This is a schematic diagram of a flat container containing fresh leaves obtained in the embodiment. Figure 9 This is a schematic diagram of the processed image of fresh leaves obtained in the embodiment. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example 1

[0017] Combination Figures 1 to 7 As shown, a fresh leaf grading mechanism 2 for a tea-picking robot includes a negative pressure adsorption pipe assembly 1 (the main body of the picking robot and its negative pressure adsorption pipe assembly 1 are existing technologies; specifically, the intelligent tea-picking robot developed by Zhejiang University of Technology can be used), and a collector installed below the discharge port 3 of the negative pressure adsorption pipe assembly 1. The collector includes a flat container 12 with an open top. The width of the inner cavity 18 of the flat container 12 is 1.3 times the diameter of a single tooth (this structure allows single teeth to fall into the inner cavity 18 of the flat container 12 and prevents non-single tooth fresh leaves from falling). The flat container 12 is made of plexiglass with a wall thickness of 3mm. The bottom wall of the flat container 12 is provided with several ventilation holes. The top of the flat container 12 is provided with two sets of symmetrically arranged inclined conveyor belts 16. A cylindrical container is provided below one end of the inclined conveyor belt 16. The two sets of inclined conveyor belts 16 together form a V-shaped transmission channel 17. The narrowest part of the V-shaped transmission channel 17 is equal to the width of the inner cavity 18 of the flat container 12. The single tooth falling from the outlet 3 of the negative pressure adsorption pipe assembly 1 first falls into the V-shaped transmission channel 17, and then falls directly into the inner cavity 18 of the flat container 12. The non-single tooth falling from the outlet 3 of the negative pressure adsorption pipe assembly 1 falls into the V-shaped transmission channel 17 and is then horizontally conveyed to the cylindrical container 15.

[0018] In this embodiment, several flat containers 12 are arranged side by side below the inclined conveyor belt 16. All flat containers 12 are movably locked in a strip groove, which is the space between two spaced plates 11. A baffle 10 is provided at one end of the strip groove, and an intermittent pushing part is provided behind the flat containers 12 on the tail side. Figure 3 (The right side is the tail side). When a single flat container 12 is filled with a single tooth, the intermittent pusher pushes the flat container 12 forward a standard distance, so that an empty flat container 12 moves directly below the V-shaped transmission channel 17. The intermittent pusher includes a screw drive structure 19 located on the side of the strip groove. The movable seat 9 of the screw drive structure 19 abuts against the rear end of the flat container 12 on the tail side. The screw drive structure 19 is connected to the control system, which controls the operation of the screw drive structure 19.

[0019] In this embodiment, a set of through holes is provided on the upper part of the flat container 12, and a through-beam photoelectric sensor is provided on the side of the flat container 12 located at the receiving station. The emitting end 14 and the receiving end 13 of the photoelectric sensor are respectively located on both sides of the through holes. When there is no single tooth blocking the through hole in the flat container 12, the light beam emitted by the emitting end 14 of the photoelectric sensor can just pass through the through hole and be received by the receiving end 13.

[0020] In this embodiment, a tea-picking robot employing the grading mechanism 2 described in this example is used to grade freshly picked tea leaves during the tea-picking process. A camera terminal is positioned on the side of the strip-shaped trough. The camera terminal is used to acquire images of the fresh leaves inside the flat container 12 and upload these images to the control system. The control system includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: S1, reads fresh leaf images from the camera terminal in real time; S2, preprocess the obtained fresh leaf image (including contrast adjustment, brightness adjustment and saturation adjustment, etc.) and extract the single tooth contour in the image; S3, obtain the length of the obtained single tooth contour, and compare the obtained length value with the preset threshold range. If the obtained length value is less than the minimum value of the threshold range, then mark the fresh leaf image as having an unqualified single tooth.

[0021] Taking a small number of fresh leaves (single buds) of a certain variety as an example (e.g.) Figure 8 As shown), the obtained fresh leaf images are preprocessed, and then the length of the single tooth contour is measured (e.g., Figure 9 As shown in the image, the maximum length of a single tusk outline is 12cm, and the minimum length is 9mm. Based on the pre-set threshold range of 8.5-13mm according to tea grade, all the tusks in the image meet the size requirements and can be directly processed into the tea processing procedure. During use, approximately 600 tusks in a single image are processed by the computer control system to determine if they meet the requirements, taking only about ten seconds. Further screening of every 10,000 tusks takes an average of less than five minutes and requires no manual operation.

[0022] The key features of this solution include: using a transparent, flat container 12 with a specific structure to collect the picked tea buds, ensuring that the buds are arranged in a single, flat, and tightly packed layer within the container 12. This results in no ghosting, obstruction, or overlap of buds in the acquired images, allowing personnel to easily identify substandard buds and facilitating rapid and intuitive quality assessment by the tea-picking robot. The flat container 12, in conjunction with the V-shaped transmission channel 17, enables rapid separation of non-bud buds from single bud buds. This allows for quick and efficient grading of fresh leaves during the picking process (a fundamental difference from existing methods that involve sending picked leaves to a tea processing workshop for grading using a sieve machine). This improves the efficiency of bud acquisition, facilitating the production of higher-standard, high-grade tea. Crucially, the grading process avoids high-frequency and high-speed impacts on the buds. Furthermore, this solution allows for accurate and rapid assessment of the tea hairs within the same batch of buds, a function that existing solutions cannot achieve.

Claims

1. A fresh leaf grading mechanism for a tea-picking robot, comprising a negative pressure adsorption pipe assembly of the picking robot, and a collector installed below the discharge port of the negative pressure adsorption pipe assembly, characterized in that: The collector includes a flat container with an open top. The width of the inner cavity of the flat container is 1.2 to 1.3 times the diameter of a single tooth. Two symmetrically arranged inclined conveyor belts are provided at the top of the flat container. A cylindrical container is located below one end of each inclined conveyor belt. The two inclined conveyor belts together form a V-shaped transmission channel. The narrowest part of the V-shaped transmission channel is equal to the width of the inner cavity of the flat container. Single teeth falling from the outlet of the negative pressure adsorption pipe assembly first fall into the V-shaped transmission channel and then fall directly into the inner cavity of the flat container. Non-single teeth falling from the outlet of the negative pressure adsorption pipe assembly fall into the V-shaped transmission channel and are then transversely conveyed to the cylindrical container.

2. The fresh leaf grading mechanism according to claim 1, characterized in that: The flat container is made of plexiglass with a wall thickness of no more than 5 mm.

3. The fresh leaf grading mechanism according to claim 2, characterized in that: Several flat containers are arranged side by side below the inclined conveyor belt. All flat containers are movably clamped in the strip groove. An intermittent pusher is set behind the flat container on the tail side. When a single flat container is filled with a single tooth, the intermittent pusher pushes the flat container forward by a standard distance, so that an empty flat container moves to the bottom of the V-shaped conveyor channel.

4. The fresh leaf grading mechanism according to claim 3, characterized in that: The intermittent feeding section includes a screw drive structure located on the side of the strip groove. The movable seat of the screw drive structure abuts against the rear end of the flat container on the tail side. The screw drive structure is connected to the control system, which controls the operation of the screw drive structure.

5. The fresh leaf grading mechanism according to claim 4, characterized in that: A set of through holes is provided on the upper part of the flat container. A through-beam photoelectric sensor is provided on the side of the flat container located at the receiving station. The transmitting end and receiving end of the photoelectric sensor are located on both sides of the through hole. When there is no single tooth blocking the through hole in the flat container, the light beam emitted by the transmitting end of the photoelectric sensor can just pass through the through hole and be received by the receiving end.

6. The fresh leaf grading mechanism according to claim 5, characterized in that: The flat container has several ventilation holes.

7. A tea-picking robot employing a grading mechanism as described in any one of claims 1-6, characterized in that: Used to grade freshly picked tea leaves during the tea-picking process.

8. The tea-picking robot according to claim 7, characterized in that: A camera terminal is installed on the side of the strip groove. The camera terminal is used to acquire images of fresh leaves inside the flat container and upload the images to the control system. The control system includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: S1, reads fresh leaf images from the camera terminal in real time; S2, Extract the single bud contour from the obtained fresh leaf image; S3, obtain the length of the obtained single tooth contour, and compare the obtained length value with the preset threshold range. If the obtained length value is less than the minimum value of the threshold range, then mark the fresh leaf image as having an unqualified single tooth.

Images