Crop leaf beating control method, device, equipment, system and storage medium

Abstract

Embodiments of the present application disclose a crop leafing control method, device, equipment, system and storage medium. Obtain the to-be-detected point cloud data of the to-be-leafed crop; perform target identification on the to-be-detected point cloud data to determine whether there is a target structure; if yes, determine a target stem section according to the to-be-detected point cloud data and the target structure, and determine a target path for leafing the to-be-leafed crop according to the to-be-detected point cloud data corresponding to the target stem section; control a manipulator to perform pole-holding type circling at a starting point position of the target path, and control the manipulator to rotate a blade in the manipulator while moving along the target path, so as to realize leafing of the target stem section. The embodiments of the present application improve the identification accuracy of leafing operation and improve the automation of leafing operation.

Description

Technical Field

[0001] This application relates to agricultural automation technology, and more particularly to a method, apparatus, equipment, system, and storage medium for controlling crop leaf removal. Background Technology

[0002] As crops grow, their leaves gradually age and their photosynthetic function weakens. Furthermore, as the fruit ripens, the demand for nutrients increases. Therefore, it is necessary to remove some leaves in a timely manner to reduce nutrient consumption.

[0003] In existing technologies, leaf removal operations rely on the precise identification of crop petioles to achieve accurate leaf removal.

[0004] However, due to the overlapping and interlacing growth of leaves, the accuracy of petiole identification cannot be guaranteed, resulting in poor automation of leaf removal operations. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, system, and storage medium for controlling crop leaf removal, so as to improve the recognition accuracy of leaf removal operations and enhance the automation of leaf removal operations.

[0006] In a first aspect, embodiments of this application provide a method for controlling crop leaf loss, the method comprising: Acquire the point cloud data of the crop to be defoliated; Target identification is performed on the point cloud data to be detected to determine whether a target structure exists; If so, the target stem segment is determined based on the point cloud data to be detected and the target structure, and the target path for leaf removal of the crop to be removed is determined based on the point cloud data to be detected corresponding to the target stem segment. The robotic arm is controlled to perform a pole-hugging circling motion at the starting point of the target path, and the blades in the robotic arm are rotated while the robotic arm moves along the target path to remove leaves from the target stem segment.

[0007] Secondly, embodiments of this application also provide a crop leaf-removal control device, which includes: The point cloud data acquisition module is used to acquire the point cloud data of the crop to be defoliated. The target structure determination module is used to identify targets in the point cloud data to be detected and determine whether a target structure exists. The target path determination module is used to determine the target stem segment based on the point cloud data to be detected and the target structure if the target is the target, and to determine the target path for leaf removal of the crop based on the point cloud data to be detected corresponding to the target stem segment. The target stem segment leaf-removing module is used to control the robotic arm to perform a pole-holding circling motion at the starting position of the target path, and to control the blades in the robotic arm to rotate while moving along the target path, so as to remove leaves from the target stem segment.

[0008] Thirdly, embodiments of this application also provide a crop leaf-removal control device, which includes: One or more processors; Storage device for storing one or more programs; When one or more programs are executed by one or more processors, the one or more processors implement any of the crop leaf removal control methods provided in the embodiments of this application.

[0009] Fourthly, this application also provides a crop leaf-removing control system, including the crop leaf-removing control device in the above embodiments; The system also includes a robotic arm, a robotic hand, and a vision information acquisition module; The crop leaf removal control equipment is connected to the robotic arm, robotic arm, and vision information acquisition module. The visual information acquisition module is used to acquire the point cloud data of the crop to be leafed and send it to the crop leafing control device. The crop leaf-removal control equipment is used to acquire the detection point cloud data of the crop to be leafed; to perform target identification on the detection point cloud data to determine whether a target structure exists; if so, to determine the target stem segment based on the detection point cloud data and the target structure, and to determine the target path for leaf removal based on the detection point cloud data corresponding to the target stem segment; to send a movement command to the robotic arm to control the robotic arm to move to the starting position of the target path and return to the starting point signal; after receiving the starting point signal, to send a closing command to the robotic arm to control the robotic arm to perform a pole-holding circling at the starting position of the target path and return to the closing signal; after receiving the closing signal, to send a start movement command to the robotic arm to control the robotic arm to move along the target path, and to send a leaf-removal command to the robotic arm to control the blades in the robotic arm to rotate while moving along the target path, so as to remove the leaves from the target stem segment; The robotic arm is used to receive a closing command and, in response to the closing command, to perform a lever-like circling motion at the starting position of the target path and return to the closing signal; it also receives a leaf-cutting command and, in response to the leaf-cutting command, controls the blade to rotate while the robotic arm moves along the target path. The robotic arm is used to receive movement commands and, in response to the movement commands, move to the starting position of the target path and return a signal indicating arrival at the starting point; it also receives a start movement command and, in response to the start movement command, moves along the target path.

[0010] Fifthly, embodiments of this application also provide a storage medium including computer-executable instructions, which, when executed by a computer processor, are used to perform any of the crop leaf-removal control methods provided in embodiments of this application.

[0011] Sixthly, embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements any of the crop leaf-removal control methods provided in embodiments of this application.

[0012] This application acquires point cloud data of the crop to be leafed, and uses this point cloud data to obtain three-dimensional data, providing multi-dimensional data support for target structure identification and target path determination. It then performs target identification on the point cloud data to determine the presence of a target structure, reducing the difficulty and improving the accuracy of identification by identifying specific target structures. If a target structure is present, the application identifies the target stem segment based on the point cloud data and the target structure, and determines the target path for leaf removal based on the corresponding point cloud data. Determining the target path by using the target stem segment reduces the difficulty of path determination and improves the accuracy of the target path. The application controls a robotic arm to perform a pole-hugging circling motion at the starting point of the target path, and controls the blades in the robotic arm to rotate while moving along the target path to remove leaves from the target stem segment. This pole-hugging circling motion enables leaf removal based on the target path, eliminating the need for precise identification of the leaf petioles, reducing the difficulty of target identification during leaf removal, improving the accuracy of leaf removal operations, and automatically controlling the robotic arm based on the target path, thus improving the automation of leaf removal operations. Therefore, the technical solution of this application solves the problem that the accuracy of petiole identification cannot be guaranteed due to the overlapping and interlacing growth of leaves, resulting in poor automation of leaf removal operations. It achieves the effect of improving the identification accuracy and automation of leaf removal operations. Attached Figure Description

[0013] Figure 1a This is a flowchart of a crop leaf-removal control method according to Embodiment 1 of this application; Figure 1b This is a schematic diagram of a robotic arm structure according to Embodiment 1 of this application; Figure 1c This is a schematic diagram of a robotic arm structure according to Embodiment 1 of this application; Figure 1d This is a schematic diagram of a robotic arm closure structure according to Embodiment 1 of this application; Figure 1e This is a schematic diagram of a robotic arm's stem-hugging and pole-encircling structure according to Embodiment 1 of this application; Figure 2aThis is a flowchart of a crop leaf-removal control method according to Embodiment 2 of this application; Figure 2b This is a schematic diagram of a robotic arm unfolded in a straight line, as shown in Embodiment 2 of this application; Figure 3 This is a flowchart of a crop leaf-removal control method according to Embodiment 3 of this application; Figure 4 This is a schematic diagram of the structure of a crop leaf-removing control device according to Embodiment 4 of this application; Figure 5 This is a schematic diagram of the structure of an electronic device according to Embodiment 5 of this application; Figure 6a This is a schematic diagram of a crop leaf-removing control system according to Embodiment Six of this application; Figure 6b This is a schematic diagram of the structure of a robotic arm according to Embodiment Six of this application; Figure 6c This is a schematic diagram of a double-arm crop leaf-removing control system according to Embodiment Six of this application; Figure 6d This is a schematic diagram of a double-arm crop leaf-removing control system according to Embodiment Six of this application; Figure 6e This is a front view of a double-arm crop leaf-removing control system according to Embodiment Six of this application; Figure 6f This is a side view of a double-arm crop leaf-removing control system according to Embodiment Six of this application; Figure 6g This is a top view of a double-arm crop leaf-removing control system according to Embodiment Six of this application; Figure 6h This is a schematic diagram of the installation angle of the two arms of a double-arm crop leaf-cutting control system according to Embodiment Six of this application. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0015] It should be noted that the terms "first" and "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] Example 1 Figure 1a This is a flowchart of a crop leaf-removal control method provided in Embodiment 1 of this application. This embodiment is applicable to the situation of leaf removal for vertically planted crops. The method can be executed by a crop leaf-removal control device, which can be implemented in software and / or hardware and specifically configured in an electronic device, such as a leaf-removal device.

[0017] See Figure 1a The crop leaf removal control method shown includes the following steps: S110. Obtain the point cloud data of the crop to be defoliated.

[0018] The crop to be leafed can be any crop that requires leaf removal. This crop can be vertically grown, such as a tomato. The point cloud data to be detected can be point cloud data of the crop to be leafed, acquired through an image acquisition device, used for target recognition and target path localization. For example, the point cloud data to be detected can be acquired using a binocular camera. The point cloud data to be detected is three-dimensional data; each point contains X, Y, and Z three-dimensional coordinate information, facilitating subsequent determination of the target path for leaf removal.

[0019] S120. Perform target identification on the point cloud data to be detected to determine whether a target structure exists.

[0020] The target structure can be a pre-defined crop structure that requires leaf removal, used to determine whether leaf removal should be performed. For example, since the target structure required for leaf removal may differ at different growth stages of the crop, the target structure can be set in stages by professional technicians according to the leaf removal requirements of the crop to be removed; this application does not specifically limit this. Target recognition can identify pre-defined target structures in the point cloud data to be detected. For example, target recognition can be performed using a deep learning model. For instance, target recognition can be performed using a convolutional neural network.

[0021] Taking tomatoes as an example, the target structures for leaf removal can include fruit enlargement and fruit coloring. Under normal circumstances, a tomato plant grows about 3 leaves per week, while the optimal number of functional leaves for a mature tomato plant is 9-11 in winter and 13-15 in summer. Over time, tomato leaves gradually age, reducing their photosynthetic function; therefore, aging leaves should be removed promptly to reduce nutrient consumption. When the first cluster of fruit begins to enlarge, begin removing the first set of old leaves, removing 2-3 leaves each time. When the tomato begins to change color, all leaves below the first cluster of fruit should be removed. Removing old leaves must be done on a sunny day, and the petioles should be cut cleanly to prevent disease infection from the wounds.

[0022] S130. If so, then based on the point cloud data to be detected and the target structure, determine the target stem segment, and based on the point cloud data to be detected corresponding to the target stem segment, determine the target path for leaf removal of the crop to be removed.

[0023] If so, meaning a target structure is identified, then leaf removal is required. The target stem segment can be the main stem portion containing the leaves to be removed. The target path can be the movement path of the equipment performing the leaf removal operation. For example, the target path can be a line segment formed by the center of gravity of the target stem segment.

[0024] Based on the target structure, the stem portion requiring leaf removal can be determined. The main stem structure is identified using the point cloud data to be detected, and the corresponding target stem segment is determined accordingly. For example, when the identified target structure is a ripening tomato fruit, the main stem below the first cluster of fruit is identified as the stem segment to be removed. The stem below the first cluster of fruit is segmented using the point cloud data to be detected, and this portion of the stem is identified as the target stem segment.

[0025] Because the stem has a certain shape, in order to accurately determine the target path, it is necessary to determine the target path according to certain rules. For example, cross-sections can be selected at certain intervals along the growth direction of the stem, and the geometric center or centroid of each cross-section can be determined as the target path based on the point cloud data of each cross-section.

[0026] Taking tomatoes as an example, extensive research in tomato production bases revealed that vertically planted tomatoes have a high degree of standardization, with the main stem growing vertically or tilted in one direction. This eliminates complex agronomical issues such as shading and breakage, and the main stems are relatively thick, allowing for high accuracy in main stem identification. Therefore, by determining the target path for leaf removal based on the point cloud data corresponding to the target stem segment, the identification of small targets such as petioles can be avoided, reducing the difficulty of leaf removal and improving efficiency and stability.

[0027] S140. Control the robotic arm to perform a pole-hugging circling motion at the starting position of the target path, and control the blade in the robotic arm to rotate while moving along the target path, so as to remove leaves from the target stem segment.

[0028] A robotic arm can be an automated operating device with transmission and cutting functions in crop leaf-pruning control equipment, used to perform leaf-pruning operations. The robotic arm can be a single-bar gripper type, employing a rope-driven, multi-stage rotary gear interconnection drive mode.

[0029] like Figure 1b This is a schematic diagram of a robotic arm structure. Figure 1b 11 is the rope-driven motor, 12 is the drive rope, 13 is the blade-driven motor, 14 is the integrated cutting transmission gear, 15 is the protective bracket, and 16 is the connecting component. (For example...) Figure 1b As shown, each rotating blade of the robotic arm has serrated teeth that perform both cutting and transmission functions. Driven by a rope-driven motor, the robotic arm opens and closes via a drive rope. When the robotic arm is closed, five gears mesh together. When the first gear rotates under the drive of the motor, it drives the other gears to rotate as well, thus cutting the leaf stalk. This pole-mounted cutting method eliminates the need for leaf stalk identification to perform leaf cutting. Figure 1c This is a schematic diagram of a robotic arm structure. Figure 1b and Figure 1c These are schematic diagrams of the structure from different perspectives. Figure 1c In the diagram, 17 represents a closed magnetic attraction A, and 18 represents a closed magnetic attraction B. For example... Figure 1d This is a schematic diagram of a robotic arm's closed structure. Figure 1d To and Figure 1c A schematic diagram of the structure when closed, viewed from the same perspective. When the robotic arm closes... Figure 1c Closed magnetic locator A and closed magnetic locator B are connected to form a fixed closed structure.

[0030] A pole-mounted encircling mechanism allows a robotic arm to encircle and close the stem for leaf removal operations. For example... Figure 1e This is a schematic diagram of a robotic arm that performs a stem-hugging, looping motion. Controlling the robotic arm to perform this looping motion at the starting point of the target path can be achieved by opening and closing the robotic arm at the starting point, thus looping the stem at that point. Then, as the robotic arm moves along the target path, the gears within it rotate under the drive of a motor, causing the blades on the gears to rotate as well, cutting the leaf stalks to remove leaves from the target stem segment.

[0031] Crops and leaf stalks present complex agronomical problems such as shading, breakage, tight contact, and small target size. By using a pole-holding operation method, only the stem needs to be identified, avoiding the problem of identifying the position of the stalk in complex environments. Moreover, this method only requires cutting along the main stem, avoiding the problem of arm path planning under complex plant growth conditions, reducing the identification difficulty in the leaf removal process, improving the identification accuracy of leaf removal operations, and thus improving the automation of leaf removal operations.

[0032] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant regions.

[0033] The technical solution of this embodiment acquires the point cloud data of the crop to be leafed, and obtains three-dimensional data from the point cloud data to provide multi-dimensional data support for the identification of target structures and the determination of target paths. It then performs target identification on the point cloud data to determine the presence of target structures, reducing the difficulty of identification and improving the accuracy of identification by identifying specific target structures. If a target structure is found, the target stem segment is determined based on the point cloud data and the target structure, and the target path for leafing the crop is determined based on the point cloud data corresponding to the target stem segment. Determining the target path by identifying the target stem segment reduces the difficulty of target path determination and improves the accuracy of target path determination. The robot arm is controlled to perform a pole-holding circling motion at the starting position of the target path, and the blades in the robot arm rotate while moving along the target path to achieve leafing of the target stem segment. This pole-holding circling motion enables leafing operation based on the target path, eliminating the need for precise identification of the leaf petioles, reducing the difficulty of target identification during leafing, improving the accuracy of leafing operation, and automatically controlling the robot arm based on the target path, thus improving the automation of leafing operation. Therefore, the technical solution of this application solves the problem that the accuracy of petiole identification cannot be guaranteed due to the overlapping and interlacing growth of leaves, resulting in poor automation of leaf removal operations. It achieves the effect of improving the identification accuracy and automation of leaf removal operations.

[0034] Example 2 Figure 2a This is a flowchart of a crop leaf-removal control method provided in Embodiment 2 of this application. The technical solution of this embodiment is further refined based on the above technical solution.

[0035] Furthermore, the concept of "controlling the robotic arm to perform a pole-hugging circling motion at the starting point of the target path, and controlling the blades in the robotic arm to rotate while moving along the target path to achieve leaf removal from the target stem segment" is refined as follows: "Controlling the rope-driven module in the robotic arm to pull the multi-stage rotating gears to unfold in a straight line, and when the center of the robotic arm after closing coincides with the starting point of the target path, controlling the rope-driven module to pull the multi-stage rotating gears to close in a pole-hugging circling motion; determining the target rotation speed based on the petiole hardness and blade diameter of the crop to be leafed; controlling the blades in the robotic arm to rotate at the target rotation speed while moving along the target path to achieve leaf removal from the target stem segment," thus realizing automatic leaf removal from the target stem segment.

[0036] See Figure 2a The method for controlling leaf drop in crops, as shown, includes: S210. Obtain the point cloud data of the crop to be defoliated.

[0037] S220. Perform target identification on the point cloud data to be detected to determine whether a target structure exists.

[0038] S230. If so, then based on the point cloud data to be detected and the target structure, determine the target stem segment, and based on the point cloud data to be detected corresponding to the target stem segment, determine the target path for leaf removal of the crop to be removed.

[0039] S240: Control the rope drive module in the manipulator to pull the multi-stage rotating gears to unfold in a straight line. When the center of the manipulator after closing coincides with the starting point of the target path, control the rope drive module to pull the multi-stage rotating gears to close in a lever-like manner, so as to achieve lever-like circling.

[0040] The robotic arm employs a rope-driven, multi-stage rotating gear interconnection drive system, with blades mounted on the rotating gears. Internally, the robotic arm features a rope-driven, multi-stage rotating gear interconnection drive structure, enabling multi-gear transmission within the confined space of a plant. Each rotating gear simultaneously performs both cutting and transmission functions: when the gripper opens or closes, its movement is controlled by a motor in the rope-driven module via a steel cable. After the robotic arm closes around the target stem segment, the internal rotating gears close and interlock; at this point, the motor drives the first gear to rotate, synchronously driving the remaining gears to rotate in unison, thereby driving all blades to rotate at high speed to cut off the encountered petioles. A schematic diagram of the robotic arm is shown below. Figure 1b .

[0041] The core principle of the robotic arm operation is that the robotic arm uses a multi-functional gripper to hold the main stem of the crop plant and move upwards or downwards along the main stem. High-speed rotating blades around it cut off the petioles that it encounters, thus achieving the leaf removal function. When this robotic arm is operating, it only needs to identify the approximate position of the main stem, fruit, and leaves, without needing to identify the fruit stalks and petioles to complete the leaf removal operation. This avoids the two major challenges of recognizing the pose of small targets such as fruit, leaf, and petioles and planning the path of the arm in unstructured environments.

[0042] The rope drive motor of the robot's rope drive module can control the steel wire rope to pull a multi-stage rotating gear that unfolds in a straight line. For example... Figure 2b This is a schematic diagram of a robotic arm that unfolds in a straight line. After unfolding, the robotic arm moves towards the starting point of the target path. When the center of the robotic arm after closing coincides with the starting point of the target path, the control rope drive module pulls the multi-stage rotating gears to close in a lever-like manner, achieving lever-like circling.

[0043] S250. Determine the target rotation speed based on the petiole hardness of the crop to be leafed and the blade diameter.

[0044] The blade diameter can be the diameter of the blade in the robotic arm, used to determine the target rotation speed. The petiole stiffness can be the stiffness of the petiole of the crop to be leafed, used to determine the target rotation speed.

[0045] The blades used in the robotic arm for cutting are made of high-hardness materials, enabling efficient cutting of the petioles of crops to be stripped of leaves. Specifically, the blade diameter can be determined based on the diameter of the petiole. For example, when stripping tomatoes, a blade diameter of 45 mm is suitable. The target rotation speed is the rotational speed used during the stripping operation to ensure that the leaves are knocked off.

[0046] For example, the formula for calculating the target rotation speed is as follows: ; in, The target rotational speed; It is the cutting line speed; It is the diameter of the blade.

[0047] For example, when cutting leafy tomato leaves, a 45mm diameter blade is selected. Generally, the blade's edge speed for cutting cork is 50-80m / s. Since tomato stems are relatively soft, a saw blade edge linear speed of 60m / s is chosen. The target rotational speed of the motor is then calculated using the formula above. It is 424.4 .

[0048] S260. While the robot arm moves along the target path, the blades in the robot arm rotate at the target rotation speed to remove leaves from the target stem segment.

[0049] By controlling the robotic arm to move along the target path while the blades in the robotic arm rotate at the target rotation speed, all the petioles on the target stem segment corresponding to the target path can be knocked off, thus achieving leaf removal from the target stem segment.

[0050] In one optional embodiment, controlling the robotic arm to move along the target path while the blade in the robotic arm rotates at a target rotation speed to remove leaves from the target stem segment includes: determining the target power of the rotary drive motor based on the target rotation speed, blade diameter, total cutting force at the blade edge, mechanical transmission efficiency, and mechanical transmission efficiency of the rotary drive motor driving the robotic arm; and controlling the rotary drive motor to operate at the target power while the robotic arm moves along the target path to remove leaves from the target stem segment.

[0051] The blade is driven by a frameless motor, and the motor's driving force is transmitted to the next stage through a multi-stage rotating gear system, ensuring torque and speed during the cutting process. By determining the target power of the rotary drive motor, the blade can be guaranteed to rotate at the target speed to achieve the cutting operation.

[0052] Torque required for blade cutting The torque of the rotary drive motor is related to the material hardness and cutting resistance. The calculation formula is as follows: ; in, This represents the total cutting force at the blade edge. Where is the blade radius; For mechanical transmission efficiency.

[0053] For example, when preparing flavorful tomatoes for leaf removal, a texture analyzer is used to test the shear force of the tomato stalk (early stage of maturity). If the average shear force of the distal stalk is 9.66 N and that of the proximal stalk is 17.90 N, then the maximum value of 17.90 N can be taken. The selection factor is typically set between 0.8 and 0.95. It is 0.9; The value is 0.0225 meters; therefore, the torque required for a single gear to operate is calculated using the torque calculation formula. It is 0.4475N.

[0054] The formula for calculating the petiole fracture cutting force of the crop to be processed is as follows: ; in, The cutting force that causes the petiole to break. Material shear strength; This represents the cutting area.

[0055] Among them, cutting area The calculation formula is as follows: ; in, The thickness of the material into which the blade cuts; The cutting width is determined by the blade diameter.

[0056] Gear drive power of a rotary drive motor driving a single gear The calculation formula is as follows: .

[0057] For example, when processing flavored tomatoes with defoliated leaves, the motor speed is calculated earlier. It is 424.4 drive motor torque The mechanical transmission efficiency is 0.4475 N. If it is 0.9, then based on the gear drive power The calculation formula yields a power of 0.022 kW required to drive one gear. Considering gear resistance and multi-stage transmission losses, the gear transmission efficiency can be set to φ=0.5. Therefore, the power required for all five gears to work simultaneously is 5* / φ=0.22 .

[0058] While controlling the robotic arm to move along the target path, the rotary drive motor is controlled to run at the target power to ensure that the petioles of the target stem segment are knocked off as the robotic arm moves along the target path, thus achieving leaf removal from the target stem segment.

[0059] The target power of the rotary drive motor is determined based on the target rotation speed, blade diameter, total cutting force at the blade edge, mechanical transmission efficiency, and the mechanical transmission efficiency of the rotary drive motor driving the robot arm. This ensures that the petioles of the target stem segment can be knocked off. The robot arm is controlled to move along the target path while the rotary drive motor operates at the target power to knock off the leaves from the target stem segment, ensuring accurate leaf knocking.

[0060] The technical solution of this embodiment controls the rope-driven module in the robotic arm to pull the multi-stage rotating gears in a straight line. When the center of the robotic arm after closing coincides with the starting point of the target path, the rope-driven module pulls the multi-stage rotating gears in a lever-like closing motion to achieve lever-like circling. This lever-like circling is accurately achieved at the starting point of the target path, ensuring the accuracy of subsequent leaf-removing operations. The target rotation speed is determined based on the petiole hardness and blade diameter of the crop to be leafed, ensuring efficient removal of the petioles. While the robotic arm moves along the target path, the blades in the robotic arm rotate at the target rotation speed to remove leaves from the target stem segment, ensuring the accuracy of leaf removal operations on the target stem segment. Furthermore, through lever-like circling and the determination of the target stem segment, there is no need for precise identification of small targets such as petioles, reducing the difficulty of leaf removal operations and improving the automation and recognition accuracy of leaf removal operations.

[0061] Example 3 Figure 3 This is a flowchart of a crop leaf-removal control method provided in Embodiment 3 of this application. The technical solution of this embodiment is further refined based on the above technical solution.

[0062] Furthermore, the statement "If so, then based on the point cloud data to be detected and the target structure, determine the target stem segment, and based on the point cloud data to be detected corresponding to the target stem segment, determine the target path for leaf removal of the crop to be leafed" is refined to: "If so, then the structural type of the target structure determines the corresponding target stem segment; segment the target point cloud data corresponding to the target stem segment from the point cloud data to be detected; determine the centroid of the target stem segment based on the target point cloud data, and use the centroid of the target stem segment as the target path" to determine the target path.

[0063] See Figure 3 The method for controlling leaf drop in crops, as shown, includes: S310. Obtain the point cloud data of the crop to be defoliated.

[0064] S320. Perform target identification on the point cloud data to be detected to determine whether a target structure exists.

[0065] S330. If so, the structural type of the target structure determines the corresponding target stem segment.

[0066] The structure type can be the type of the target structure used to determine the corresponding target stem segment. For example, the structure type can be set by a professional technician based on the leaf-removal requirements of the crop to be treated; this application does not specifically limit this. For instance, when the crop to be leaf-removed is a tomato, the structure type can include fruit enlargement and color-changing fruit, etc. The leaf-removal requirements differ at different growth stages of the crop; therefore, different structure types correspond to different target stem segments.

[0067] Taking tomatoes as an example, when the first cluster of fruits is swelling, the first removal of old leaves begins, removing 2-3 leaves each time, and the main stem section below the first cluster of fruits is designated as the target stem segment. When the tomatoes begin to ripen, all leaves below the first cluster of fruits should be removed, and the main stem section below the first cluster of fruits is again designated as the target stem segment. Therefore, it is necessary to determine the corresponding target stem segment based on the type of target structure.

[0068] S340. Segment the target point cloud data corresponding to the target stem segment from the point cloud data to be detected.

[0069] The target point cloud data can be the point cloud data corresponding to the target stem segment in the point cloud data to be detected, used to determine the target path. For example, the target point cloud data corresponding to the target stem segment in the point cloud data to be detected can be segmented using an image recognition model. The image recognition model can be a deep learning model, such as a convolutional neural network. The image recognition model can identify the positions of the fruit, main stem, and leaf canopy, and then, based on the target stem segment, determine the target point cloud data corresponding to the target stem segment from the identified main stem point cloud data.

[0070] S350. Determine the centroid of the target stem segment based on the target point cloud data, and use the centroid of the target stem segment as the target path.

[0071] The location of the target stem segment can be determined using target point cloud data, and further, its center of gravity can be identified. Since the main stem may grow with some tilt, using the center of gravity of the target stem segment as the target path ensures that the robotic arm will not damage the main stem when moving along that path. Specifically, after determining the center of gravity of the target stem segment based on the target point cloud data, coordinate transformation can be used to convert the point cloud data from the camera coordinate system to the robotic arm coordinate system, facilitating the robotic arm's spatial positioning.

[0072] S360: Control the robotic arm to perform a pole-hugging circling motion at the starting position of the target path, and control the blades in the robotic arm to rotate while moving along the target path, so as to remove leaves from the target stem segment.

[0073] The technical solution of this embodiment determines the corresponding target stem segment by identifying the structural type of the target structure if it is true, thus ensuring the accuracy of the leaf-removal position; it segments the target point cloud data corresponding to the target stem segment from the point cloud data to be detected; it determines the centroid of the target stem segment based on the target point cloud data, and uses the centroid of the target stem segment as the target path. The determination of the target path based on the centroid of the target stem segment takes into account the possibility that the main stem of the crop to be leafed may have a certain degree of tilt, ensuring the safety of the main stem, ensuring the accuracy of subsequent leaf-removal operations, and protecting the safety of the crop.

[0074] Example 4 Figure 4 The diagram shown is a structural schematic of a crop leaf-removing control device according to Embodiment 4 of this application. This embodiment is applicable to the situation of leaf removal for vertically planted crops. The specific structure of the crop leaf-removing control device is as follows: The point cloud data acquisition module 410 is used to acquire the point cloud data of the crop to be defoliated. The target structure judgment module 420 is used to perform target recognition on the point cloud data to be detected and determine whether a target structure exists. The target path determination module 430 is used to determine the target stem segment based on the point cloud data to be detected and the target structure if the target is so, and to determine the target path for leaf removal of the crop to be removed based on the point cloud data to be detected corresponding to the target stem segment. The target stem segment leaf-removing module 440 is used to control the robot arm to perform a pole-holding circling motion at the starting position of the target path, and to control the blades in the robot arm to rotate while moving along the target path, so as to remove leaves from the target stem segment.

[0075] The technical solution of this embodiment acquires the point cloud data of the crop to be leafed, and obtains three-dimensional data from the point cloud data to provide multi-dimensional data support for the identification of target structures and the determination of target paths. It then performs target identification on the point cloud data to determine the presence of target structures, reducing the difficulty of identification and improving the accuracy of identification by identifying specific target structures. If a target structure is found, the target stem segment is determined based on the point cloud data and the target structure, and the target path for leafing the crop is determined based on the point cloud data corresponding to the target stem segment. Determining the target path by identifying the target stem segment reduces the difficulty of target path determination and improves the accuracy of target path determination. The robot arm is controlled to perform a pole-holding circling motion at the starting position of the target path, and the blades in the robot arm rotate while moving along the target path to achieve leafing of the target stem segment. This pole-holding circling motion enables leafing operation based on the target path, eliminating the need for precise identification of the leaf petioles, reducing the difficulty of target identification during leafing, improving the accuracy of leafing operation, and automatically controlling the robot arm based on the target path, thus improving the automation of leafing operation. Therefore, the technical solution of this application solves the problem that the accuracy of petiole identification cannot be guaranteed due to the overlapping and interlacing growth of leaves, resulting in poor automation of leaf removal operations. It achieves the effect of improving the identification accuracy and automation of leaf removal operations.

[0076] Optionally, the robotic arm adopts a rope-driven multi-stage rotary gear interconnection drive mode, with blades mounted on the rotary gears; Correspondingly, the target stem segment leaf removal module 440 includes: The robot control unit is used to control the rope drive module in the robot to pull the multi-stage rotating gears to unfold in a straight line, and when the center of the robot after closing coincides with the starting point of the target path, it controls the rope drive module to pull the multi-stage rotating gears to close in a lever-like manner, so as to achieve lever-like circling. The target rotation speed determination unit is used to determine the target rotation speed based on the petiole hardness and blade diameter of the crop to be leafed. The target stem segment leaf-removing unit is used to control the robot arm to move along the target path while the blades in the robot arm rotate at the target rotation speed to remove leaves from the target stem segment.

[0077] Optionally, the target stem segment leaf-removing unit includes: The target power determination subunit is used to determine the target power of the rotary drive motor based on the target rotation speed, blade diameter, total cutting force at the blade edge, mechanical transmission efficiency, and mechanical transmission efficiency of the rotary drive motor driving the robot. The target power control subunit is used to control the robot arm to move along the target path while controlling the rotary drive motor to run at the target power, so as to achieve leaf removal from the target stem segment.

[0078] Optionally, the target path determination module 430 includes: The target stem segment determination unit is used to determine the corresponding target stem segment if the target structure type is specified. The target point cloud data determination unit is used to segment the target point cloud data corresponding to the target stalk segment from the point cloud data to be detected. The target path determination unit is used to determine the centroid of the target stem segment based on the target point cloud data, and to use the centroid of the target stem segment as the target path.

[0079] The crop leaf-removing control device provided in this application embodiment can execute the crop leaf-removing control method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the crop leaf-removing control method.

[0080] According to embodiments of the present invention, the present invention also provides an electronic device, a readable storage medium, and a computer program product.

[0081] Example 5 Figure 5 This is a schematic diagram of the structure of a crop leaf-removing control device provided in Embodiment 5 of this application, as shown below. Figure 5 As shown, the crop leaf-removing control device includes a processor 510, a memory 520, an input device 530, and an output device 540; the number of processors 510 in the crop leaf-removing control device can be one or more. Figure 5Taking a processor 510 as an example; the processor 510, memory 520, input device 530, and output device 540 in the electronic device can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0082] The memory 520, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the crop leaf-removal control method in this embodiment (e.g., the point cloud data acquisition module 410, the target structure judgment module 420, the target path determination module 430, and the target stem segment leaf-removal module 440). The processor 510 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 520, thereby implementing the above-mentioned crop leaf-removal control method.

[0083] The memory 520 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, the memory 520 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 520 may further include memory remotely located relative to the processor 510, which can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0084] The input device 530 can be used to receive input character information and generate key signal inputs related to user settings and function control of the crop leaf-removing control equipment. The output device 540 may include a display device such as a display screen.

[0085] Example 6 Figure 6a This is a schematic diagram of a crop leaf-removal control system provided in Embodiment Six of this application, as shown below. Figure 6a As shown, the crop leaf removal control system includes a crop leaf removal control device 610; the system also includes a robotic arm 620, a robotic arm 630, and a visual information acquisition module 640; the crop leaf removal control device 610 is communicatively connected to the robotic arm 620, the robotic arm 630, and the visual information acquisition module 640.

[0086] The visual information acquisition module 640 is used to acquire the point cloud data of the crop to be leafed and send it to the crop leafing control device 610; the visual information acquisition module 640 can be a binocular camera.

[0087] The crop leaf-removing control device 610 is used to acquire the detection point cloud data of the crop to be leafed; to perform target recognition on the detection point cloud data to determine whether a target structure exists; if so, to determine the target stem segment based on the detection point cloud data and the target structure, and to determine the target path for leaf removal of the crop based on the detection point cloud data corresponding to the target stem segment. A movement command is sent to the robotic arm 630 to control the robotic arm 630 to move to the starting position of the target path and return to the starting point signal; after receiving the starting point signal, a closing command is sent to the robotic hand 620 to control the robotic hand 620 to perform a pole-holding circling motion at the starting position of the target path and return to the closing signal; after receiving the closing signal, a start movement command is sent to the robotic arm 630 to control the robotic arm 630 to move along the target path, and a leaf-removing command is sent to the robotic hand 620 to control the blades in the robotic hand 620 to rotate while moving along the target path, thereby removing leaves from the target stem segment.

[0088] The movement command can be a control command sent by the crop leaf-trimming control device 610 to the robotic arm 630, instructing the robotic arm 630 to move towards the starting position of the target path. The movement command includes the coordinates of the starting position. The arrival signal can be a signal returned by the robotic arm 630 to the crop leaf-trimming control device 610, indicating that the robotic arm 630 has moved to the starting position of the target path after the robotic arm closes, so that the crop leaf-trimming control device 610 can determine to send a closing command to the robotic arm 620. For example, before sending the movement command to the robotic arm 630, an unfolding command is sent to the robotic arm 620 to instruct the robotic arm to unfold in a straight line, avoiding damage to other leaves during the movement of the robotic arm 630, and preparing for the closing of the robotic arm 620. The closing command is a control command sent by the crop leaf-trimming control device 610 to the robotic arm 620, instructing the multi-stage rotating gears in the robotic arm 620 to close under the pull of the drive rope, preparing for leaf trimming. The start-move command is a control command sent by the crop leaf-pruning control device 610 to the robotic arm 630, instructing the robotic arm 630 to move along a target path. The start-move command includes the target path. The leaf-pruning command is a control command sent by the crop leaf-pruning control device 610 to the robotic arm 620, instructing the multi-stage rotary gears in the robotic arm 620 to rotate to perform the leaf-pruning operation. The leaf-pruning command includes the target power of the rotary drive motor, etc. The method for determining the relevant parameters in the movement commands and start-move commands sent by the crop leaf-pruning control device 610 has been described in the above embodiments and will not be repeated here.

[0089] The robotic arm 620 is used to receive a closing command and, in response to the closing command, to perform a pole-like circling at the starting position of the target path and return to the closing signal; it also receives a leaf-cutting command and, in response to the leaf-cutting command, controls the blade to rotate while the robotic arm 630 moves along the target path.

[0090] In one alternative embodiment, the robotic arm 620 includes: a drive rope 621, a multi-stage rotary gear 622, a rope-driven motor 623, and a rotary drive motor 624; each rotary gear 624 is equipped with a rotatable blade; the rope-driven motor 623 is communicatively connected to the drive rope 621, and the multi-stage rotary gear 622 is communicatively connected to the rotary drive motor 624.

[0091] The rope-driven motor 623 is used to receive a closing command and, in response to the closing command, stretch the drive rope 621 to control the multi-stage rotating gear 622 to perform a pole-like loop at the starting position of the target path, and after completing the closing, returns a closing signal to the crop leaf-removing control device 610; the rotary drive motor 624 is used to receive a leaf-removing command and, in response to the leaf-removing command, controls the rotatable blades mounted on each rotary gear 624 to rotate.

[0092] The robotic arm 630 is used to receive movement commands and, in response to the movement commands, move to the starting position of the target path and return a signal indicating arrival at the starting point; it also receives a start movement command and, in response to the start movement command, moves along the target path.

[0093] In one alternative embodiment, the robotic arm 630 is a foldable multi-degree-of-freedom robotic arm. The robotic arm 630 includes a foldable mechanical part and a vertical slide. The robotic arm 630 is mounted on the vertical slide and can move up and down in the vertical direction to meet the operational requirements of a high vertical space within the greenhouse.

[0094] The robotic arm 630 features a folding and unfolding function: when not performing leaf-removing operations, the joints of the robotic arm 630 can be folded and retracted, allowing the entire arm to be compactly stacked together, occupying minimal lateral space. This facilitates the robot's movement or parking within narrow tomato planting rows. When starting work, the joints of the robotic arm 630 unlock and unfold sequentially, extending to the working position as needed, achieving a larger working range within a limited space. Because the robotic arm 630 employs a foldable design, it occupies little space and does not interfere with each other when both arms are working simultaneously, effectively solving the collision and interference problems that may occur when two robotic arms are operating in confined spaces.

[0095] like Figure 6b This is a schematic diagram of the structure of a robotic arm. Figure 6b21 is the first joint of the folding arm, 22 is the second joint of the folding arm, 23 is the third joint of the folding arm, 24 is the fourth joint of the folding arm, 25 is the rotary joint of the folding arm, 26 is the lifting slide, and 27 is the robotic arm. Figure 6b As shown, the robotic arm is a multi-degree-of-freedom, large-stroke, 6-DOF foldable robotic arm, consisting of a vertical sliding section, a 4-stage folding arm, and a rotating wrist. The robotic arm can adapt to work scenarios with limited space, large stroke requirements, and high flexibility requirements, and has the following characteristics: it possesses 6 degrees of freedom, enabling it to perform complex spatial operations; it features multi-stage folding, occupying little space and having a large working stroke; The first part is the vertical sliding section 26. The sliding module consists of a lead screw module, which achieves smooth linear movement through a high-precision roller lead screw. The roller lead screw module slide is tightly connected to the rotary platform through a connecting rod.

[0096] The second part, the multi-stage folding arm (21-24), addresses the need for large robotic arms to operate in situations where crops, particularly those with fruits, have a wide height span and narrow row spacing, making them unsuitable for large robotic arms. Therefore, a folding robotic arm with a telescopic arm consisting of four foldable sections is designed. This folding robotic arm offers advantages such as a large operating range and minimal space requirements, effectively solving the problems of large spans and narrow row spacing for targets like fruits.

[0097] Part Three: Rotating Wrist 26. This rotating joint is used to address the issue of end flexibility when the main stem of the plant is tilted in different directions.

[0098] The robotic arm needle features a large stroke, multiple degrees of freedom, foldability, and high flexibility, making it well-suited to the agronomic characteristics of vertically grown crops in multi-span greenhouses. It can be used for harvesting, pruning, and removing old leaves from crops such as tomatoes and cucumbers, effectively solving the problems of limited working space and large height differences in vertical planting scenarios such as tomatoes.

[0099] Based on the structure of the robotic arm, the kinematics of the DH coordinate system are modeled and analyzed using the DH parameter method. The DH coordinate system, proposed by Dienergetic Hartenberg in 1955, establishes a link coordinate system and uses four basic parameters—joint offset (d), joint distance (a), joint angle (θ), and link torsion angle (α)—to accurately describe the spatial geometric relationships between the joints of the robotic arm. Through these parameters, a homogeneous transformation matrix between the links of the robotic arm is systematically established, thus comprehensively describing the pose relationship of the end effector relative to the base coordinate system. Table 1 shows the DH kinematic parameters of a six-axis robotic arm.

[0100] Table 1. Parameters of the Six-Axis Robotic Arm (DH) Figure 6cThis is a schematic diagram of a leaf-removing control system for a double-arm crop. Figure 6c As shown, a dual-arm collaborative crop leaf-cutting control system can be further constructed. The crop leaf-cutting control system comprises two symmetrically arranged foldable robotic arms, each with a lever-type leaf-cutting manipulator at its end. Both robotic arms are mounted on a vertical slide and can move synchronously along the vertical direction. During operation, under the command of the crop leaf-cutting control device, the dual-arm crop leaf-cutting control system moves to the side of the crop plant. The two robotic arms extend from both sides of the plant, and the manipulators open and wrap around the main stem from both sides. Subsequently, the dual-arm crop leaf-cutting control system drives the vertical slide to move the robotic arms along the main stem of the crop from bottom to top, while simultaneously starting the blades to rotate and continuously cut leaves. The two robotic arms work in coordination; while one arm moves upward to cut leaves, the other arm can also perform leaf-cutting tasks on adjacent plants without interference, thus greatly improving the overall operating efficiency. Figure 6d This is a schematic diagram of a double-arm crop leaf-removing control system. Figure 6e A front view of a double-arm crop leaf-removing control system. Figure 6f This is a side view of a double-arm crop leaf-removing control system. Figure 6g This is a top view of a double-arm crop leaf-removing control system. Figure 6h This is a schematic diagram showing the installation angle of the two arms in a dual-arm crop leaf-removing control system.

[0101] like Figure 6h As shown, the arm is mounted at a 45-degree angle to the vehicle body, ensuring sufficient space for the robotic arm to work without interference. Throughout the leaf-removal process, the robot uses visual recognition to determine the plant's location and the area requiring treatment, eliminating the need for individual inspection of each petiole. As the lever-type robotic arm slides along the main stem, any petioles it touches are automatically cut off by the high-speed rotating serrated blades. In the vertical planting environment of a glass greenhouse, the dual-arm leaf-removal robot can efficiently complete automatic leaf removal tasks: a single up-and-down slide removes all leaves within a section of the main stem. This system utilizes the coordinated movement of both arms and the vertical sliding table to achieve rapid and comprehensive leaf removal of crop plants, significantly improving the automation and efficiency of greenhouse production management.

[0102] Example 7 Embodiment 7 of this application also provides a storage medium containing computer-executable instructions. When executed by a computer processor, the computer-executable instructions are used to execute a crop leaf-removal control method. The method includes: acquiring detection point cloud data of the crop to be leafed; performing target recognition on the detection point cloud data to determine whether a target structure exists; if so, determining a target stem segment based on the detection point cloud data and the target structure, and determining a target path for leaf removal of the crop based on the detection point cloud data corresponding to the target stem segment; controlling a robotic arm to perform a pole-hugging circling motion at the starting position of the target path, and controlling the blade in the robotic arm to rotate while moving along the target path, so as to remove leaves from the target stem segment.

[0103] Of course, the computer-executable instructions provided in the embodiments of this application are not limited to the method operations described above, but can also perform related operations in the crop leaf removal control method provided in any embodiment of this application.

[0104] Based on the above description of the implementation methods, those skilled in the art can clearly understand that this application can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0105] It is worth noting that in the embodiments of the crop leaf-removing control device described above, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of this application.

[0106] Note that the above are merely preferred embodiments and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for controlling leaf drop in crops, characterized in that, include: Acquire the point cloud data of the crop to be defoliated; The point cloud data to be detected is used for target identification to determine whether a target structure exists; If so, then based on the point cloud data to be detected and the target structure, the target stem segment is determined, and based on the point cloud data to be detected corresponding to the target stem segment, the target path for leaf removal of the crop to be removed is determined; The robotic arm is controlled to perform a pole-hugging circling motion at the starting point of the target path, and the blades in the robotic arm are controlled to rotate while moving along the target path, so as to remove leaves from the target stem segment.

2. The method according to claim 1, characterized in that, The robotic arm adopts a rope-driven multi-stage rotary gear interconnection drive mode, and the rotary gear is equipped with blades. Correspondingly, the control robot arm performs a pole-hugging circling maneuver at the starting position of the target path, and controls the blades in the robot arm to rotate while moving along the target path, so as to remove leaves from the target stem segment, including: The rope drive module in the robotic arm is controlled to pull the multi-stage rotating gears to unfold in a straight line. When the center of the robotic arm after closing coincides with the starting point of the target path, the rope drive module is controlled to pull the multi-stage rotating gears to close in a pole-holding manner, so as to achieve pole-holding circling. The target rotation speed is determined based on the petiole hardness of the crop to be leafed and the blade diameter. While the robotic arm moves along the target path, the blades in the robotic arm rotate at the target rotation speed to remove leaves from the target stem segment.

3. The method according to claim 2, characterized in that, The control of the robotic arm to move along the target path while the blades in the robotic arm rotate at the target rotation speed to remove leaves from the target stem segment includes: The target power of the rotary drive motor is determined based on the target rotation speed, blade diameter, total cutting force at the blade edge, mechanical transmission efficiency, and mechanical transmission efficiency of the rotary drive motor driving the robot arm. While controlling the robotic arm to move along the target path, the rotary drive motor is controlled to operate at the target power to remove leaves from the target stem segment.

4. The method according to claim 1, characterized in that, If so, then based on the point cloud data to be detected and the target structure, a target stem segment is determined, and based on the point cloud data to be detected corresponding to the target stem segment, a target path for leaf removal of the crop to be removed is determined, including: If so, the structural type of the target structure determines the corresponding target stem segment; Segment the target point cloud data corresponding to the target stem segment from the point cloud data to be detected; The centroid of the target stem segment is determined based on the target point cloud data, and the centroid of the target stem segment is used as the target path.

5. A crop leaf-removal control device, characterized in that, include: The point cloud data acquisition module is used to acquire the point cloud data of the crop to be defoliated. The target structure determination module is used to identify targets in the point cloud data to be detected and determine whether a target structure exists. The target path determination module is used to determine the target stem segment based on the point cloud data to be detected and the target structure if the target structure is so, and to determine the target path for removing leaves from the crop to be removed based on the point cloud data to be detected corresponding to the target stem segment. The target stem segment leaf-removing module is used to control the robotic arm to perform a pole-holding circling motion at the starting position of the target path, and to control the blades in the robotic arm to rotate while moving along the target path, so as to remove leaves from the target stem segment.

6. A crop leaf-removal control device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the crop leaf removal control method as described in any one of claims 1-4.

7. A crop leaf-removal control system, characterized in that, include: The crop leaf removal control device as described in claim 6; The system also includes a robotic arm, a robotic hand, and a vision information acquisition module; The crop leaf removal control device is communicatively connected to the robotic arm, the robotic arm, and the visual information acquisition module; The visual information acquisition module is used to acquire the point cloud data of the crop to be leafed and send it to the crop leafing control device. The crop leaf removal control device is used to acquire the point cloud data of the crop to be removed; The point cloud data to be detected is used for target identification to determine whether a target structure exists; If so, the target stem segment is determined based on the point cloud data to be detected and the target structure, and the target path for leaf removal is determined based on the point cloud data to be detected corresponding to the target stem segment; a movement command is sent to the robotic arm to control the robotic arm to move to the starting position of the target path and return to the starting point signal; after receiving the starting point signal, a closing command is sent to the robotic arm to control the robotic arm to perform a pole-holding circling at the starting position of the target path and return to the closing signal; after receiving the closing signal, a start movement command is sent to the robotic arm to control the robotic arm to move along the target path, and a leaf removal command is sent to the robotic arm to control the blades in the robotic arm to rotate while moving along the target path, so as to remove the leaves from the target stem segment; The robotic arm is used to receive the closing command and, in response to the closing command, to perform a pole-holding circling motion at the starting position of the target path and return to the closing signal; it also receives the leaf-cutting command and, in response to the leaf-cutting command, controls the blade to rotate while the robotic arm moves along the target path. The robotic arm is configured to receive the movement command, and in response to the movement command, move to the starting position of the target path and return a signal indicating arrival at the starting point; receive the start movement command, and in response to the start movement, move according to the target path.

8. The system according to claim 7, characterized in that, The robotic arm includes: a drive rope, a multi-stage rotary gear, a rope-driven motor, and a rotary drive motor; each of the rotary gears is equipped with a rotatable blade; The rope drive motor is communicatively connected to the drive rope, and the multi-stage rotary gear is communicatively connected to the rotary drive motor; The rope-driven motor is used to receive the closing command and, in response to the closing command, stretch the drive rope to control the multi-stage rotating gear to perform a pole-holding loop at the starting position of the target path, and after completing the closing, return to the closing signal of the crop leaf-removing control device. The rotary drive motor is used to receive the leaf-beating command and, in response to the leaf-beating command, to rotate the rotatable blades mounted on each of the gears.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the crop leaf removal control method as described in any one of claims 1-4.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the crop leaf removal control method as described in any one of claims 1-4.

Images