Efficient grape harvesting system

Abstract

The invention provides an efficient grape harvesting system, and relates to the field of agricultural intelligent machinery, the efficient grape harvesting system comprises a chassis, a walking mechanism, a rotating plate, a three-coordinate mechanism, a tail end executing mechanism, a visual assembly, a navigation system and an electric control box, and a specific harvesting method of the system comprises the steps that 1, the harvesting system moves to a to-be-harvested area for target recognition; step 2, extracting a navigation path in real time through a Beidou antenna, and running the path; in the navigation process, starting, stopping and steering control of the harvesting system is achieved by recognizing obstacles between rows; and step 3, starting the three-coordinate mechanism and the end execution mechanism to cut and harvest the target in real time. According to the system, interference of surrounding fruits of target grapes or interference of grape vines in the harvesting process can be avoided, the flexibility and accuracy in the harvesting process are ensured, and accurate identification and positioning of the grapes and rapid, efficient and low-cost harvesting are achieved.

Description

[0001] This invention is a divisional application of patent application number 2025100629793, entitled "A three-degree-of-freedom grape harvesting robot and harvesting method". Technical Field

[0002] This invention relates to the field of agricultural intelligent machinery technology, specifically to a high-efficiency grape harvesting system. Background Technology

[0003] Grapes include table grape varieties (such as Kyoho, Red Globe, Shine Muscat, and Seedless White), wine grape varieties (such as Cabernet Sauvignon, Chardonnay, and Pinot Noir), dried grape varieties (such as Seedless White and Muscat), and processing grape varieties (used for juice, canning, etc., requiring high juice yield and stable color). Currently, grape harvesting methods mainly include manual harvesting and mechanized harvesting. Manual harvesting involves precisely cutting the grape bunches with hand tools, and manually sorting and transporting them throughout the process, thereby minimizing damage to the bunches and berries and preserving the original quality of the variety. Mechanized harvesting involves using mechanical vibration devices to clamp the grapevines / branches, generating high-frequency, low-amplitude vibrations, causing the grape bunches to detach from the vines under inertia and fall into a collection device below. While manual harvesting can reduce fruit damage and preserve fruit quality, it is time-consuming, inefficient, and costly, failing to achieve the goal of rapid, efficient, and low-cost harvesting for wine grape varieties. Mechanized harvesting can quickly collect grapes, but it easily causes significant damage to the grapes and leads to juice leakage. It is only suitable for dried or processed varieties such as raisins and whole grapes. For wine grape varieties that need to be processed into whole bunches, it cannot maximize the integrity of the grapes after harvesting, resulting in the loss of flavor during the winemaking process.

[0004] Furthermore, existing technologies face difficulties in locating and identifying grapes during the grape harvesting process, resulting in a low success rate and an increased probability of fruit damage. This not only affects the utilization rate of grapes after harvesting but also makes it easy for the robotic arm to get stuck during the harvesting process (such as being entangled by vines or surrounded by fruit), thereby affecting the efficiency of grape harvesting and missing the best time for grape harvesting. Summary of the Invention

[0005] To address the problems existing in the prior art, the present invention aims to provide a high-efficiency grape harvesting system. This system adopts a three-degree-of-freedom structural design, which not only simplifies the structure of the harvesting machinery and avoids interference from surrounding fruits or grapevines during the harvesting process, but also effectively reduces the complexity and cost of the system, ensures flexibility and accuracy during the harvesting process, and achieves accurate grape identification, positioning, and rapid, efficient, and low-cost harvesting.

[0006] The objective of this invention is achieved through the following technical solution: A high-efficiency grape harvesting system includes a chassis, a walking mechanism, a rotating plate, a three-coordinate mechanism, an end effector, a vision component, a navigation system, and an electrical control box. The walking mechanism, which is a tracked walking component, is mounted on the bottom surface of the chassis and is used to control the operation of the entire harvesting system. A rotating plate is rotatably mounted on the end face of the chassis, and a three-coordinate mechanism is mounted on the end face of the rotating plate. The three-coordinate mechanism includes a Y-axis component, an X-axis component, and a Z-axis component, with the Z-axis component located on the end face of the rotating plate, the Y-axis component mounted on the Z-axis component, and the X-axis component located between the Y-axis components. The end effector is located at the end of the X-axis component away from the Y-axis component, and the vision component is located at the end of the X-axis component corresponding to the end effector and above the end effector. The navigation system is located on the side of the chassis, and the electrical control box is located on the end face of the chassis. The electrical control box is offset from the three-coordinate mechanism and is electrically connected to the control devices of the walking mechanism, the rotating plate, the three-coordinate mechanism, the end effector, the navigation system, and the vision component.

[0007] Based on further optimization of the above scheme, the Z-axis assembly includes two sets of horizontal guide rails, two horizontal lead screws, two horizontal sliders, and two horizontal slide blocks. The two sets of horizontal guide rails are arranged parallel to each other on both sides of the end face of the rotating plate, and a rotating horizontal lead screw is set on each side of the two sets of horizontal guide rails that are close to each other. The outer wall of the horizontal lead screw is threaded with a horizontal slider. The horizontal slide block is slidably set on the horizontal guide rail and is fixedly connected to the corresponding horizontal slider. The Y-axis assembly includes a connecting crossbar, two sets of vertical guide rails, two vertical lead screws, two vertical sliders, and two vertical slide blocks. The bottom surface of the connecting crossbar is fixedly connected to the end faces of the two horizontal sliders and the two horizontal slide blocks. Two parallel vertical guide rails are fixedly installed on both sides of the end face of the connecting crossbar. A rotating vertical screw is installed on the side of the two sets of vertical guide rails that are far apart from each other, and a vertical slider is threaded onto the outer wall of the vertical screw. A vertical slide block is slidably installed on the outer wall of the vertical guide rail, and the vertical slide block is fixedly connected to the corresponding vertical slider. The X-axis assembly includes a positioning crossbar, a moving screw, and a linkage bracket. The two ends of the positioning crossbar are fixedly connected to the inner side wall of the corresponding vertical slide block, and a rotating moving screw is installed on one side of the positioning crossbar. The linkage bracket is perpendicular to the positioning crossbar, and the bottom surface of the linkage bracket is threadedly connected to the moving screw and slidably connected to the positioning crossbar. The end effector includes two servo motors, two adjusting ropes, a support plate, connecting hoses, a universal ball cage, a high-speed motor, and a three-sided cutting tool. The support plate is mounted on the bottom surface of the end of the linkage bracket furthest from the positioning plate via connecting hoses. Multiple connecting hoses are evenly distributed around the outer ring of the support plate's end face. Two servo motors are fixedly mounted on the end face of the linkage bracket furthest from the positioning plate, and each servo motor's output end is wound with an adjusting rope. The end of the adjusting rope furthest from the servo motor passes through a fixed pulley on the end face of the linkage bracket, then through the linkage bracket and connects with the support plate. The end face is connected, and the adjusting ropes are symmetrically distributed on both sides of the support plate and the adjusting ropes are set in opposite positions to the connecting hose; between the linkage bracket and the support plate and in the inner circle of the connecting hose, a universal ball cage is set with the support plate as the same axis as the support plate; a high-speed motor is fixedly installed on the bottom surface of the support plate, and the output end of the high-speed motor is fixedly sleeved with a three-sided cutting tool through a coupling. The main body of the three-sided cutting tool is a disc-shaped structure, which extends from the base with three main cutting surfaces (that is, three slots are evenly opened on the outer circle of the disc-shaped structure, dividing the outer circle of the disc-shaped structure into three cutting surfaces), and each cutting surface is provided with a biomimetic angular groove resembling a dinosaur tooth.

[0008] Based on further optimization of the above scheme, the inner ring of the three-sided cutting tool is provided with arc-shaped grooves corresponding to the three cutting surfaces.

[0009] Based on further optimization of the above scheme, the vision component includes a camera bracket and a vision camera, with the vision camera fixedly mounted on the end face of the linkage bracket via the camera bracket. The navigation system includes two sets of antenna brackets and two sets of BeiDou antennas, with the two sets of antenna brackets respectively located at the front and rear ends of the chassis, and the BeiDou antennas fixedly mounted on the antenna brackets.

[0010] A harvesting method for a high-efficiency grape harvesting system includes: Step 1, the harvesting system moves to the harvesting area, and a vision camera is activated to capture images and videos to complete target recognition; Step 2, the navigation path is extracted in real time using a Beidou antenna, and the harvesting system is started to move towards the target identified in Step 1 according to the navigation path; during navigation, the start, stop, and steering control of the harvesting system are achieved by identifying obstacles between rows; Step 3, the three-coordinate mechanism and the end effector are activated to cut and harvest the target in real time.

[0011] Based on further optimization of the above scheme, step one uses an improved YOLOv10 model to achieve target recognition. Specifically, it involves: first, obtaining an image dataset D={ containing the main grapevine, fruiting branches, and fruit stalks. I 1 ,I 2 ,…,I N}, I i (i=1,2,…,N) represents a single image; the image dataset is divided into training, validation and test sets; then, a labeling tool is used to add bounding boxes and category labels to each cluster of fruits; Next, an improved YOLOv10 model was constructed. The YOLOv10 network model includes a deep learning network structure consisting of a Backbone, Neck, and Head. This network model starts with image input, extracts features through the Backbone, performs feature fusion through the Neck, and finally performs regression and classification prediction through the Head. Therefore, in the improved YOLOv10 model, the FasterBlock module replaces the Bottleneck module in the C2f function of the Backbone to optimize computational efficiency and improve the accuracy of small object detection. ; In the formula: H, W Indicates the size of the feature map. k Indicates convolution and size. C p Indicates the number of channels calculated; To address the detection requirements of different scale targets such as grapevines and fruiting branches, the LSKA attention mechanism is used to replace the SPPF module in the backbone. LSKA reduces computational complexity and improves the model's long-range dependency characteristics and multi-scale adaptability by decomposing the 2D convolutional kernel into horizontal and vertical 1D convolutional kernels. LSKA outputs at each channel... Z c for: ; In the formula: W kx1and W 1xk These represent the convolution kernels in the horizontal and vertical directions, respectively. F c Indicates the convolutional kernel weights generated based on the content; To improve the detection accuracy of small targets such as grape stems, a CARAFE module was introduced into the Neck section, and the upsampling operation was optimized for a given target location. Its corresponding upsampling feature is : ; In the formula: This represents the convolutional kernel weights generated based on the content. r Indicates the radius of the neighborhood region; This indicates that after the convolution operation, the value located at ( i+n,j+m Elements on the output feature map; Finally, the improved YOLOv10 model was trained on the labeled training set. After training, the model performance was evaluated using the validation set. The trained improved YOLOv10 model was then deployed to the grape harvesting system to achieve the goal of quickly and accurately detecting the main grape vine, fruiting branches, and fruit stalks.

[0012] Based on further optimization of the above scheme, the specific method for extracting the navigation path in step two is as follows: First, the YOLOv8-SEG model is used to identify the roads between the grape rows, and the left and right edge points of the roads are extracted. x l ,L l ), ( x r ,L r The left and right grapevine row lines were fitted using the least squares method: ; In the formula: This represents the slope and intercept of the line on the left. This represents the slope and intercept of the line on the right. Then, based on the rows of grapevines on the left and right, extract the reference points for the navigation line. P i :( x i ,y i ), i=1,2,…,n : ; In the formula: ( x li ,Lli ） Represents the coordinates of any point on the left-hand line, ( x ri ,L ri ） Indicates and ( x li ,L li ） The coordinates of the corresponding point on the right side of the same horizontal line; β Indicates the position parameters of the navigation line; Set position parameter threshold ,like If , then the navigation line is considered to be located in the center of the grape row; if If the navigation line is on the left row, then... If the navigation line is located on the right side, the harvesting system can be moved to the left or right by controlling the position parameters, thereby achieving grape harvesting on the left or right side.

[0013] Based on further optimization of the above scheme, the specific method for controlling the start, stop, and steering of the harvesting system by identifying obstacles between rows in step two is as follows: First, the orchard navigation line is fitted using the least squares method: ; Then, select m points at equal intervals along the orchard navigation line, that is, from the bottom of the image to the top, with... H The width is determined by sequentially cutting through m horizontal lines along the selected orchard navigation line. m Each point is recorded as :( ); After that, with Based on this, obtain the left boundary point for the safe operation of the harvesting system. P slj :( ) and the right boundary point P srj :( ): ; In the formula: This indicates the ratio between the actual width of the road and the actual width of the harvesting system. express RoI The width of the lowest inner road in pixel coordinates; Finally, the YOLOv8-SEG model was used to identify obstacles on the road during the journey, and the location of the obstacles was obtained. RoI The bounding rectangle is obtained from the inner rectangle; and the coordinates of the leftmost bottom vertex of the bounding rectangle are obtained respectively. P obl:( x obl ,y obl ) and the coordinates of the rightmost bottom vertex P obr :( x obr ,y obr If satisfied or If the obstacle is determined to be within the safe driving area of ​​the harvesting system, the harvesting system will turn left or right to avoid the obstacle; otherwise, the harvesting system will continue to move forward.

[0014] The following are the technical effects of the present invention: This invention, through a three-coordinate mechanism design, avoids the drying caused by environmental factors such as vines, adjacent grapes, and grape leaves during grape harvesting, which could lead to jamming of the entire harvesting system or damage to the grapes. This prevents problems such as reduced harvesting efficiency and quality due to environmental interference. Simultaneously, the structural design of the end effector not only effectively improves the flexibility and adaptability of the cutting tools, ensuring efficient operation of the harvesting system in complex harvesting environments and reducing or even eliminating interference with adjacent fruits, vines, and grape leaves during harvesting, but also enhances the cutting tolerance through cooperation with the three-sided cutting tool. This ensures precise and effective cutting even when there are errors in the fruit axis posture angle or spatial plane, and also helps to remove debris during cutting, preventing jamming caused by cutting debris. Furthermore, by improving the image recognition model YOLOv10, this invention enables the identification of grapes during the harvesting process. This not only improves the monitoring accuracy of the main vine, fruit value, and fruit stem, thereby accurately identifying and locating grape fruits and reducing the probability of missed or incorrect picking, but also enhances its stability in complex harvesting environments, improving the level of automation and operational safety during the grape harvesting process. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall structure of the harvesting system in an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of the Z-axis component of the harvesting system in an embodiment of the present invention.

[0017] Figure 3 This is a schematic diagram of the Y-axis and X-axis components of the harvesting system in an embodiment of the present invention.

[0018] Figure 4 This is a schematic diagram of the X-axis component and end effector of the harvesting system in an embodiment of the present invention.

[0019] Figure 5 This is a schematic diagram of the structure of the three-sided cutting tool of the harvesting system in an embodiment of the present invention.

[0020] Figure 6 This is a network structure diagram of the improved YOLOv10 in an embodiment of the present invention.

[0021] Figure 7 This is a flowchart of the process for extracting navigation paths in an embodiment of the present invention.

[0022] Figure 8 This is a schematic diagram illustrating the analysis of inter-row obstacle identification in an embodiment of the present invention.

[0023] Figure 9 This is a schematic diagram of the low-collision ROI of the fruit stalk in an embodiment of the present invention.

[0024] Among them, 10 is the chassis; 20 is the walking mechanism; 30 is the rotating plate; 411 is the horizontal guide rail; 412 is the horizontal lead screw; 413 is the horizontal slider; 414 is the horizontal slide block; 421 is the connecting crossbar; 422 is the vertical guide rail; 423 is the vertical lead screw; 424 is the vertical slider; 425 is the vertical slide block; 431 is the positioning crossbar; 432 is the moving lead screw; 433 is the linkage bracket; 4330 is the fixed pulley; 51 is the servo motor; 52 is the adjusting rope; 53 is the support plate; 54 is the connecting hose; 55 is the universal ball cage; 56 is the high-speed motor; 57 is the three-sided cutting tool; 61 is the camera bracket; 62 is the vision camera; 71 is the antenna bracket; 72 is the Beidou antenna. Detailed Implementation

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0026] Example 1: A high-efficiency grape harvesting system includes a chassis 10, a walking mechanism 20, a rotating plate 30, a three-coordinate mechanism, an end effector, a vision component, a navigation system, and an electrical control box. The walking mechanism 20 is mounted on the bottom surface of the chassis 10, and the walking mechanism 20 adopts a tracked walking component (e.g., Figure 1 As shown), it is used to control the operation of the entire harvesting system. A rotating plate 30 is rotatably mounted on the end face of the chassis 10, and a three-coordinate mechanism (such as...) is mounted on the end face of the rotating plate 30. Figure 1(As shown); the three-axis mechanism includes a Y-axis assembly, an X-axis assembly, and a Z-axis assembly, with the Z-axis assembly positioned on the end face of the rotating plate, the Y-axis assembly mounted on the Z-axis assembly, and the X-axis assembly positioned between the Y-axis assemblies. Specifically, the Z-axis assembly includes two sets of horizontal guide rails 411, two horizontal lead screws 412, two horizontal sliders 413, and two horizontal slide blocks 414. The two sets of horizontal guide rails 411 are arranged parallel to each other on both sides of the end face of the rotating plate 30 (e.g., Figure 2 As shown), and a rotating horizontal lead screw 412 is respectively installed on the side of the two sets of horizontal guide rails 411 that are close to each other (as shown). Figure 2 As shown, the two ends of the horizontal lead screw 412 are mounted on the end faces of the rotating plate 30 on the side of the horizontal guide rail 411 that are close to each other via rotating brackets, and a motor controlling its rotation is installed at one end of the horizontal lead screw 412. Horizontal sliders 413 are threaded onto the outer wall of the horizontal lead screw 412. A horizontal slide block 414 is slidably mounted on the horizontal guide rail 411, and the horizontal slide block 414 is fixedly connected to the corresponding horizontal slider 413 (thus, the horizontal slider 413 drives the horizontal slide block 414 to slide on the horizontal guide rail 411). The Y-axis assembly includes a connecting crossbar 421, two sets of vertical guide rails 422, two vertical lead screws 423, two vertical sliders 424, and two vertical slide blocks 425. The bottom surface of the connecting crossbar 421 is fixedly connected to the end faces of the two horizontal sliders 413 and the two horizontal slide blocks 414, and two parallel vertical guide rails 422 are fixedly mounted on both sides of the end face of the connecting crossbar 421 (e.g., ...). Figure 3 As shown), a rotating vertical lead screw 423 is respectively installed on the side of the two sets of vertical guide rails 422 that are far apart from each other (as shown). Figure 3 As shown, the two ends of the vertical lead screw 423 are mounted on the side wall of the vertical guide rail 422 away from each other via rotating brackets, and the end face of the connecting crossbar 421 at the bottom of the vertical lead screw 423 is equipped with a motor to control its rotation. Vertical sliders 424 are threaded onto the outer wall of the vertical lead screw 423. A vertical slide block 425 is slidably mounted on the outer wall of the vertical guide rail 422, and the vertical slide block 425 is fixedly connected to the corresponding vertical slider 424 (thus, the vertical slider 424 drives the vertical slide block 425 to rise and fall on the vertical guide rail 422). The X-axis assembly includes a positioning cross plate 431, a moving lead screw 432, and a linkage bracket 433. The two ends of the positioning cross plate 431 are fixedly connected to the inner side wall of the corresponding vertical slide block 425 (e.g., ...). Figure 3 (As shown) and a rotating movable lead screw 432 is provided on one side of the positioning plate 431 (as shown). Figure 3As shown: The movable lead screw 432 is mounted on the back of the positioning horizontal plate 431 via rotating brackets at both ends, and a motor controlling its rotation is mounted at one end of the movable lead screw 432. The linkage bracket 433 is perpendicular to the positioning horizontal plate 431, and its bottom surface is threadedly connected to the movable lead screw 432 and slidably connected to the positioning horizontal plate 431. The end effector is located at the end of the X-axis assembly away from the Y-axis assembly (specifically, the end of the linkage bracket 433 away from the positioning horizontal plate 431, such as...). Figure 4 As shown), the end effector includes two servo motors 51, two adjusting ropes 52, a support plate 53, connecting hoses 54, a universal ball cage 55, a high-speed motor 56, and a three-sided cutting tool 57. The support plate 53 is mounted on the bottom surface of the end of the linkage bracket 433 furthest from the positioning plate 431 via the connecting hoses 54. Multiple connecting hoses 54 are evenly distributed around the outer ring of the end face of the support plate 53 (the number of connecting hoses 54 is generally 3 to 8, depending on the actual situation, such as...). Figure 4 As shown, in this embodiment, four evenly distributed connecting hoses 54 are used; two servo motors 51 are fixedly installed on the end face of the linkage bracket 433 away from the positioning cross plate 431, and an adjusting rope 52 is wound around the output end of each of the two servo motors 51. The end of the adjusting rope 52 away from the servo motor 51 passes through the fixed pulley 4330 set on the end face of the linkage bracket 433, passes through the linkage bracket 433, and connects to the end face of the support plate 53. The adjusting ropes 52 are symmetrically distributed on both sides of the support plate 53, and the adjusting ropes 53 and the connecting hoses 54 are set in opposite positions; a universal ball cage 55 coaxial with the support plate 53 is set between the linkage bracket 433 and the support plate 53 and located in the inner circle of the connecting hose 54; a high-speed motor 56 is fixedly installed on the bottom surface of the support plate 53, and the output end of the high-speed motor 56 is fixedly sleeved with a three-sided cutting tool 57 through a coupling. The main body of the three-sided cutting tool 57 is a disc-shaped structure (e.g., Figure 5 As shown), it extends from the base with three main cutting surfaces (i.e., three slots are evenly opened on the outer ring of the disc-shaped structure, dividing the outer ring of the disc into three cutting surfaces, such as...). Figure 5 (As shown) and each cutting surface is equipped with biomimetic angular grooves resembling dinosaur teeth (such as... Figure 5As shown in the enlarged view, the inner ring of the three-sided cutting tool 57 has arc-shaped grooves corresponding to the three cutting surfaces. To reduce the weight of the three-sided cutting tool 57 while ensuring its cutting performance, it is made of three different materials: the cutting edge is made of high-hardness, high-wear-resistant materials such as high-speed steel; the main body is made of lightweight materials such as alloy steel or aluminum alloy; and the internal parts are made of low-density materials such as carbon steel or its composite materials. The vision component is located at one end of the X-axis assembly corresponding to the end effector and is situated on the upper side of the end effector. The vision component includes a camera bracket 61 and a vision camera 62. The vision camera 62 is fixedly mounted on the end face of the linkage bracket 433 via the camera bracket 61. The navigation system is located on the side of the chassis 10. The navigation system includes two sets of antenna brackets 71 and two sets of Beidou antennas 72. The two sets of antenna brackets 71 are respectively located at the front and rear ends of the chassis 10, and the Beidou antennas 72 are fixedly mounted on the antenna brackets 71 (e.g., ...). Figure 1 (As shown). The electrical control box is located on the end face of the chassis 10. The electrical control box is located separately from the coordinate measuring machine and is electrically connected to the control devices of the walking mechanism 20 and the rotating plate 30, the coordinate measuring machine, the end effector, the navigation system, and the vision component.

[0027] Example 2: As a preferred embodiment of the present invention, based on the scheme of embodiment 1, in order to protect the three-sided blade 57 and prevent the three-sided blade 57 from damaging the fruit, vines, etc. during the movement, a protective shell is provided on the bottom surface of the support plate 53 and on the outer ring of the high-speed motor 56. The bottom surface of the protective shell is provided with a cutting shell with a front opening corresponding to the three-sided blade 57. The three-sided blade 57 is located inside the cutting shell and its front end protrudes from the opening of the cutting shell.

[0028] Example 3: As a preferred embodiment of the present invention, based on the scheme of embodiment 1, a collection frame (which can be a net-type collection device or a hard collection device, depending on the actual situation) is set on the lower side of the end effector and away from the positioning horizontal plate 431 of the linkage bracket 433, for collecting the grape bunches after cutting by the three-sided blade 57.

[0029] Example 4: A harvesting method for a high-efficiency grape harvesting system, using any one of the harvesting systems in Examples 1 to 3, includes: Step 1: The harvesting system moves to the harvesting area and activates the vision camera to capture images and videos, completing target recognition. An improved YOLOv10 model is used for target recognition, specifically: First, an image dataset D={ containing the main grapevine, fruiting branches, and fruit stalks is obtained. I 1 ,I 2 ,…,IN}, I i (i=1,2,…,N) represents a single image; and the image dataset is divided into a training set, a validation set, and a test set (in this embodiment, the ratio of the training set, validation set, and test set is 8:1:1); then, a labeling tool is used to add bounding boxes and category labels to each bunch of fruits; Next, an improved YOLOv10 model was constructed, referring to... Figure 6 As shown, the YOLOv10 network model includes a deep learning network structure of Backbone, Neck, and Head. This network model starts with image input, extracts features through the Backbone, performs feature fusion through the Neck, and finally performs regression and classification prediction through the Head. Therefore, the improved YOLOv10 model replaces the Bottleneck module in the C2f function of the Backbone with the FasterBlock module to optimize computational efficiency and improve the accuracy of small object detection. ; In the formula: H, W Indicates the size of the feature map. k Indicates convolution and size. C p Indicates the number of channels calculated; To address the detection requirements of different scale targets such as grapevines and fruiting branches, the LSKA attention mechanism is used to replace the SPPF module in the backbone. LSKA reduces computational complexity and improves the model's long-range dependency characteristics and multi-scale adaptability by decomposing the 2D convolutional kernel into horizontal and vertical 1D convolutional kernels. LSKA outputs at each channel... Z c for: ; In the formula: W kx1 and W 1xk These represent the convolution kernels in the horizontal and vertical directions, respectively. F c Indicates the convolutional kernel weights generated based on the content; To improve the detection accuracy of small targets such as grape stems, a CARAFE module was introduced into the Neck section, and the upsampling operation was optimized for a given target location. Its corresponding upsampling feature is : ; In the formula: This represents the convolutional kernel weights generated based on the content.r Indicates the radius of the neighborhood region; This indicates that after the convolution operation, the value located at ( i+n,j+m Elements on the output feature map; Finally, the improved YOLOv10 model was trained on the labeled training set. After training, the model performance was evaluated using the validation set. The trained improved YOLOv10 model was then deployed to the grape harvesting system to achieve the goal of quickly and accurately detecting the main grape vine, fruiting branches, and fruit stalks.

[0030] Step 2: Extract the navigation path in real time using the BeiDou antenna, and start the harvesting system to move towards the target identified in Step 1 according to the navigation path, referring to... Figure 7 As shown, the specific method for extracting the navigation path is as follows: First, the YOLOv8-SEG model is used to identify the roads between the grape rows, and the left and right edge points of the roads are extracted. x l ,L l ), ( x r ,L r The left and right grapevine row lines were fitted using the least squares method: ; In the formula: This represents the slope and intercept of the line on the left. This represents the slope and intercept of the line on the right. Then, based on the rows of grapevines on the left and right, extract the reference points for the navigation line. P i :( x i ,y i ), i=1,2,…,n : ; In the formula: ( x li ,L li ） Represents the coordinates of any point on the left-hand line, ( x ri ,L ri ） Indicates and ( x li ,L li ） The coordinates of the corresponding point on the right side of the same horizontal line; β Indicates the position parameters of the navigation line; Set position parameter threshold (Location parameter threshold) The settings are configured according to the actual situation; in this embodiment... ),like If , then the navigation line is considered to be located in the center of the grape row; if If the navigation line is on the left row, then... If the navigation line is located on the right side, the harvesting system can be moved to the left or right by controlling the position parameters, thereby achieving grape harvesting on the left or right side.

[0031] During navigation, the start, stop, and steering control of the harvesting system are achieved by identifying obstacles between rows. Specifically, the least squares method is used to fit the orchard navigation line: ; Then, select m points at equal intervals along the orchard navigation line, that is, from the bottom of the image to the top, with... H The width is determined by sequentially cutting through m horizontal lines along the selected orchard navigation line. m Each point is recorded as :( ); Then, refer to Figure 8 As shown, with Based on this, obtain the left boundary point for the safe operation of the harvesting system. P slj :( ) and the right boundary point P srj :( ): ; In the formula: This indicates the ratio between the actual width of the road and the actual width of the harvesting system. express RoI The width of the lowest inner road in pixel coordinates; Finally, the YOLOv8-SEG model was used to identify obstacles on the road during the journey, and the location of the obstacles was obtained. RoI The inner bounding rectangle (e.g.) Figure 8 As shown, the distance from the bottom edge of the circumscribed rectangle to the bottom edge of the image is... H s ); and obtain the coordinates of the leftmost bottom vertex of the circumscribed rectangle respectively. P obl :( x obl ,y obl ) and the coordinates of the rightmost bottom vertex P obr :(x obr ,y obr If satisfied or If the obstacle is determined to be within the safe driving area of ​​the harvesting system, the harvesting system will turn left or right to avoid the obstacle; otherwise, the harvesting system will continue to move forward.

[0032] Step 3: Activate the coordinate measuring machine and the end effector to cut and harvest the target in real time; The specific method for obtaining picking points is as follows: For the fruit stalks and fruiting branches of the target grapes, a low-collision ROI (such as...) is constructed. Figure 9 As shown); where the width of the rectangle is used to limit the horizontal range of the low-collision ROI of the fruit stalk; the lowest point of the resultant branch in the current horizontal range is used as the upper limit of the ROI in the vertical direction, and the highest point of the fruit bunch rectangle is used as the lower limit of the ROI region in the vertical direction (e.g. Figure 9 As shown), the specific ROI region of the fruit stalk is: ; In the formula: Wid_roi Indicates the horizontal range limit of the ROI region; Hei_roi Indicates the horizontal range limit of the ROI region; P_Clutopl.x This represents the top left vertex of the fruit string detection box. x value, P_Clutopl.y This represents the top left vertex of the fruit string detection box. y value, P_Clutopr.x This represents the top right corner of the fruit string detection box. x value, P_Stembot.y This indicates the result branch's relationship to the lowest point within the current horizontal range. y value; If the current grape bunch has an ROI, then for the stem portion of the ROI, obtain its circumscribed rectangle, and use the centroid of the circumscribed rectangle as the picking point of the three-sided blade; if the current grape ROI does not exist, then the current grape cannot be picked.

Claims

1. A high-efficiency grape harvesting system, characterized in that: The system includes a chassis, a walking mechanism, a rotating platform, a three-coordinate measuring machine (CCM), an end effector, a vision unit, a navigation system, and an electrical control box. The walking mechanism, which uses tracked components, is mounted on the bottom surface of the chassis. A rotating platform is rotatably mounted on the end face of the chassis, and a three-coordinate measuring machine (CCM) is mounted on the end face of the rotating platform. The CCM includes a Y-axis assembly, an X-axis assembly, and a Z-axis assembly. The Z-axis assembly is located on the end face of the rotating platform, the Y-axis assembly is mounted on the Z-axis assembly, and the X-axis assembly is positioned between the Y-axis assemblies. The end effector is located at the end of the X-axis assembly furthest from the Y-axis assembly. The vision unit... The component is located at one end of the X-axis assembly corresponding to the end effector and is located on the upper side of the end effector; the navigation system is located on the side of the chassis and the electrical control box is located on the end face of the chassis. The electrical control box is displaced from the three-coordinate mechanism and is electrically connected to the walking mechanism, the control device of the rotary plate, the three-coordinate mechanism, the end effector, the navigation system, and the vision component respectively; the end effector includes two servo motors, two adjusting ropes, a support plate, a connecting hose, a universal ball cage, a high-speed motor, and a three-sided cutting tool, and the inner ring of the three-sided cutting tool has arc-shaped grooves on the three cutting surfaces respectively.

2. The grape high-efficiency harvesting system according to claim 1, characterized in that: The Z-axis assembly includes two sets of horizontal guide rails, two horizontal lead screws, two horizontal sliders, and two horizontal slide blocks. The two sets of horizontal guide rails are arranged parallel to each other on both sides of the rotating plate end face, with a rotating horizontal lead screw mounted on each side of the two sets of horizontal guide rails that are close to each other. Horizontal sliders are threaded onto the outer walls of the horizontal lead screws. Horizontal slide blocks are slidably mounted on the horizontal guide rails, and the horizontal slide blocks are fixedly connected to their corresponding horizontal sliders. The Y-axis assembly includes a connecting crossbar, two sets of vertical guide rails, two vertical lead screws, two vertical sliders, and two vertical slide blocks. The bottom surface of the crossbar is fixedly connected to two horizontal sliders and two horizontal slide block end faces, and two parallel vertical guide rails are fixedly installed on both sides of the crossbar end face. A rotating vertical lead screw is installed on each side of the two sets of vertical guide rails away from each other, and the outer wall of the vertical lead screw is threaded onto the vertical slider. A vertical slide block is slidably installed on the outer wall of the vertical guide rail, and the vertical slide block is fixedly connected to the corresponding vertical slider. The X-axis assembly includes a positioning cross plate, a moving lead screw, and a linkage bracket. Both ends of the positioning cross plate are fixedly connected to the inner side wall of the corresponding vertical slide block. A rotating movable screw is installed on one side of the positioning horizontal plate. The linkage bracket is perpendicular to the positioning horizontal plate, and the bottom surface of the linkage bracket is threadedly connected to the movable screw and slidably connected to the positioning horizontal plate. A support plate is installed on the bottom surface of the end of the linkage bracket away from the positioning horizontal plate through a connecting hose. There are multiple connecting hoses, which are evenly distributed on the outer ring of the end face of the support plate. Two servo motors are fixedly installed on the end face of the linkage bracket away from the positioning horizontal plate, and an adjustment rope is wound around the output end of each of the two servo motors. The end of the adjustment rope away from the servo motor passes through a fixed pulley on the end face of the linkage bracket, passes through the linkage bracket, and is connected to the end face of the support plate. The adjustment ropes are symmetrically distributed on both sides of the support plate, and the adjustment ropes and the connecting hoses are set in opposite positions. A universal ball cage coaxial with the support plate is installed between the linkage bracket and the support plate and in the inner ring of the connecting hose. A high-speed motor is fixedly installed on the bottom surface of the support plate, and a three-sided cutting tool is fixedly sleeved on the output end of the high-speed motor through a coupling. The main body of the three-sided cutting tool is a disc-shaped structure, which extends from the base with three main cutting surfaces, and each cutting surface is provided with a biomimetic angular groove resembling a dinosaur tooth.

3. A grape high-efficiency harvesting system according to claim 1 or 2, characterized in that: The vision component includes a camera bracket and a vision camera, with the vision camera fixedly mounted on the end face of the linkage bracket via the camera bracket; the navigation system includes two sets of antenna brackets and two sets of Beidou antennas, with the two sets of antenna brackets respectively located at the front and rear ends of the chassis and the Beidou antennas fixedly mounted on the antenna brackets.

4. A grape high-efficiency harvesting system according to claim 2 or 3, characterized in that: The bottom surface of the support plate and the outer ring of the high-speed motor are provided with a protective shell. The bottom surface of the protective shell is provided with a cutting shell with a front opening corresponding to the three-sided cutting tool. The three-sided cutting tool is located inside the cutting shell and its front end protrudes from the opening of the cutting shell.

5. A grape high-efficiency harvesting system according to claim 2 or 3, characterized in that: The linkage bracket is located away from the positioning plate and below the end effector, with a collection frame set thereon.

6. The grape high-efficiency harvesting system according to claim 3, characterized in that: The specific harvesting method of this system includes: Step 1, the harvesting system moves to the area to be harvested, and the vision camera is activated to capture images and videos to complete target recognition; Step 2, the navigation path is extracted in real time through the Beidou antenna, and the harvesting system is started to move towards the target identified in Step 1 according to the navigation path; during the navigation process, the start, stop and steering control of the harvesting system is realized by recognizing obstacles between rows; Step 3, the three-coordinate mechanism and the end effector are activated to cut and harvest the target in real time.

7. The grape high-efficiency harvesting system according to claim 6, characterized in that: In step one, an improved YOLOv10 model is used to achieve target recognition. Specifically, the process involves: first, obtaining an image dataset D containing the main grapevine, fruiting branches, and fruit stalks. I 1 ,I 2 ,…,I N }, I i (i=1,2,…,N) represents a single image; and the image dataset is divided into training set, validation set and test set; Then, a labeling tool was used to add bounding boxes and category labels to each bunch of fruit; Next, an improved YOLOv10 model was constructed. The YOLOv10 network model includes a deep learning network structure consisting of a Backbone, Neck, and Head. This network model starts with image input, extracts features through the Backbone, performs feature fusion through the Neck, and finally performs regression and classification prediction through the Head. Therefore, in the improved YOLOv10 model, the FasterBlock module replaces the Bottleneck module in the C2f function of the Backbone to optimize computational efficiency and improve the accuracy of small object detection. ； In the formula: H, W Indicates the size of the feature map. k Indicates convolution and size. C p Indicates the number of channels calculated; The SPPF module in the backbone section is replaced with an LSKA attention mechanism. LSKA is applied to the output of each channel. Z c for: ； In the formula: W kx1 and W 1xk These represent the convolution kernels in the horizontal and vertical directions, respectively. F c Indicates the convolutional kernel weights generated based on the content; The CARAFE module is introduced into the Neck section, and the upsampling operation is optimized for a given target position. Its corresponding upsampling feature is : ； In the formula: This represents the convolutional kernel weights generated based on the content. r Indicates the radius of the neighborhood region; This indicates that after the convolution operation, the value located at ( i+n,j+m Elements on the output feature map; Finally, the improved YOLOv10 model was trained on the labeled training set. After training, the model performance was evaluated using the validation set. The trained improved YOLOv10 model was then deployed to the grape harvesting system to achieve the goal of quickly and accurately detecting the main grape vine, fruiting branches, and fruit stalks.

8. The grape high-efficiency harvesting system according to claim 6, characterized in that: The specific method for extracting the navigation path in step two is as follows: First, the YOLOv8-SEG model is used to identify the roads between the grape rows, and the left and right edge points of the roads are extracted. x l ,L l ), ( x r ,L r The left and right grapevine row lines were fitted using the least squares method: ； In the formula: This represents the slope and intercept of the line on the left. This represents the slope and intercept of the line on the right. Then, based on the rows of grapevines on the left and right, extract the reference points for the navigation line. P i :( x i ,y i ), i=1,2,…,n : ； In the formula: ( x li ,L li ） Represents the coordinates of any point on the left-hand line, ( x ri ,L ri ） Indicates and ( x li ,L li ） The coordinates of the corresponding point on the right side of the same horizontal line; β Indicates the position parameters of the navigation line; Set position parameter threshold ,like If , then the navigation line is considered to be located in the center of the grape row; if If the navigation line is on the left row, then... If the navigation line is located on the right side, the harvesting system can be moved to the left or right by controlling the position parameters, thereby achieving grape harvesting on the left or right side.

9. A high-efficiency grape harvesting system according to claim 8, characterized in that: The specific method for controlling the start-up, shutdown, and steering of the harvesting system by identifying obstacles between rows in step two is as follows: First, the orchard navigation line is fitted using the least squares method: ； Then, select m points at equal intervals along the orchard navigation line, that is, from the bottom of the image to the top, with... H The width is determined by sequentially cutting through m horizontal lines along the selected orchard navigation line. m Each point is recorded as :( ); After that, with Based on this, obtain the left boundary point for the safe operation of the harvesting system. P slj :( ) and the right boundary point P srj :( ): ； In the formula: This indicates the ratio between the actual width of the road and the actual width of the harvesting system. express RoI The width of the lowest inner road in pixel coordinates; Finally, the YOLOv8-SEG model was used to identify obstacles on the road during the journey, and the location of the obstacles was obtained. RoI The bounding rectangle is obtained from the inner rectangle; and the coordinates of the leftmost bottom vertex of the bounding rectangle are obtained respectively. P obl :( x obl ,y obl ) and the coordinates of the rightmost bottom vertex P obr :( x obr ,y obr If satisfied or If the obstacle is determined to be within the safe driving area of ​​the harvesting system, the harvesting system will turn left or right to avoid the obstacle; otherwise, the harvesting system will continue to move forward.

Images