Green plant pruning track control system and method based on depth camera

Abstract

The invention discloses a green plant pruning track control system and method based on a depth camera, the system comprises a depth camera, a PC, a control programming mechanism and an execution mechanism, and the depth camera is in data connection with the PC, the control programming mechanism and the execution mechanism; the depth camera is used for collecting and storing a three-dimensional point cloud of a single tree and sending the obtained point cloud data to the PC for analysis and processing; the PC is used for performing analysis and three-dimensional reconstruction on the three-dimensional point cloud acquired by the depth camera, and sending analysis and three-dimensional reconstruction results to the control programming mechanism; the control programming mechanism is used for programming a green plant pruning track according to the analysis and three-dimensional reconstruction result, setting pruning parameters and sending the pruning parameters to the execution mechanism; and the execution mechanism is used for moving and trimming along the green plant trimming track according to the command sent by the control programming mechanism. According to the invention, the operation object can be accurately identified, and the operation production efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of mechanical control, and in particular to a green plant pruning trajectory control system and method based on a depth camera. Background Art

[0002] With the rapid development of China's economy and the acceleration of urbanization, urban road greening, as an indispensable part of urban ecological construction, is constantly expanding its construction scale. The green lungs of the city - the green hedges in the middle of the road, not only play a role in separating traffic, reducing noise and air pollution, but also greatly beautify the urban environment and improve the quality of life of citizens.

[0003] In the maintenance and management of hedge isolation belts, hedge trimmers, as a professional garden machinery tool, play a vital role. Compared with traditional manual pruning methods, hedge trimmers have greatly reduced the labor intensity of manual work, improved labor efficiency, and significantly reduced maintenance costs. However, although hedge trimmers bring many conveniences, their problems in practical applications cannot be ignored. For example, the pruning accuracy is not high, the function is single, it is impossible to prune hedges of different shapes, and the degree of intelligence is low. Due to the limitations of mechanical operation itself, hedge trimmers often find it difficult to achieve the level of fineness of manual pruning when pruning hedges, which to a certain extent affects the beauty and neatness of hedges. At present, most hedge trimmers on the market can only perform simple straight-line pruning, and often cannot meet the needs of pruning hedges of different shapes and heights. This not only limits the scope of use of hedge trimmers, but also increases the complexity of hedge maintenance. In addition, there is still a lot of room for improvement in the intelligence of current hedge trimmers, such as the inability to realize autonomous navigation and intelligent recognition of hedge shapes, which to a certain extent limits the efficiency and convenience of hedge trimmers. Summary of the invention

[0004] The object of the present invention is to provide a green plant pruning trajectory control system and method based on a depth camera, so as to solve the problems of low intelligence level and low pruning accuracy of existing hedge trimmers.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A green plant pruning trajectory control system based on a depth camera, comprising a depth camera, a PC, a control programming mechanism and an actuator, wherein the depth camera is communicatively connected with the PC, the control programming mechanism and the actuator;

[0007] The depth camera is used to collect and store the three-dimensional point cloud of a single tree, and send the acquired point cloud data to a PC for analysis and processing;

[0008] The PC is used to analyze and reconstruct the three-dimensional point cloud collected by the depth camera, and send the analysis and three-dimensional reconstruction results to the control programming mechanism;

[0009] The control programming mechanism is used to program the green plant pruning trajectory according to the analysis and three-dimensional reconstruction results, set the pruning parameters, and send them to the execution mechanism;

[0010] The actuator is used to move and trim along the green plant trimming trajectory according to the command sent by the control programming mechanism.

[0011] Furthermore, the three-dimensional reconstruction process includes:

[0012] According to the 3D point cloud data information of the trees obtained by the depth camera, the Bundle Fusion algorithm is used to fuse and align the depth and color of the trees to achieve real-time 3D reconstruction.

[0013] Further, the PC includes a three-dimensional reconstruction module and a crown center visualization module;

[0014] The 3D reconstruction module is used to optimize the 3D point cloud information collected by the depth camera and generate a 3D model of the target tree through Poisson reconstruction;

[0015] The crown center visualization module is used to perform fitting analysis on the three-dimensional model of the target tree generated by the three-dimensional reconstruction module.

[0016] Furthermore, the crown center visualization module includes a circle fitting unit and a space straight line fitting unit;

[0017] The circle fitting unit is used to perform hierarchical fitting on the three-dimensional model of the target tree, and uses the least squares circle and RANSAC hybrid algorithm to find the center point of each layer;

[0018] The space straight line fitting unit is used to perform space straight line fitting on the circle center point generated by the circle fitting unit to obtain a fitting straight line with the minimum error.

[0019] Furthermore, the process of performing point cloud denoising on the three-dimensional model of the target tree is as follows:

[0020] The MATLAB platform is used to filter the noise points and segment the unnecessary background in the three-dimensional model of the target trees, and the radius filtering method is used to perform point cloud denoising on the three-dimensional model of the target trees.

[0021] Furthermore, the control programming mechanism programs the green plant pruning trajectory through LabVIEW graphical programming, transmits the center line data of the spatial straight line fitting unit to the control parameter end, and sets the green plant pruning trajectory, speed, acceleration and starting point parameters.

[0022] Furthermore, the actuator includes a return-to-origin module, a fixed-length acceleration / deceleration module, a multi-axis linear interpolation module, and a two-axis circular interpolation module;

[0023] The homing module is used to perform homing movement on the linear module, receive a reset command from the control programming mechanism, and run the linear module back to the origin;

[0024] The fixed-length acceleration and deceleration module is used to perform fixed-length movement on the linear module, and stops moving after running to the running length input by the control programming mechanism;

[0025] The multi-axis linear interpolation module is used to perform linked linear motion in the XYZ directions on the linear module;

[0026] The two-axis circular interpolation module is used to perform circular motion on any two axes of the linear module, and receives the interpolation start and end point coordinate position information output by the control programming mechanism.

[0027] Furthermore, the depth camera is installed on the actuator, the depth camera is connected to the linear module through an L-shaped connecting plate, the linear module is supported by a support frame, the linear module is connected to a stepper motor, the stepper motor drives the linear module to perform linear motion, and the depth camera dynamically scans the target, thereby driving the actuator to complete trimming.

[0028] Furthermore, the green plant pruning trajectory includes a cylindrical pruning trajectory, a conical pruning trajectory and a spherical pruning trajectory.

[0029] A green plant pruning trajectory control method based on a depth camera, comprising:

[0030] Collect and store 3D point clouds of individual trees;

[0031] Analyze and reconstruct the collected 3D point cloud to obtain analysis and 3D reconstruction results;

[0032] Program the plant pruning trajectory and set pruning parameters based on the analysis and 3D reconstruction results;

[0033] According to the set pruning parameters, it moves and prunes along the green plant pruning trajectory.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The present invention provides a green plant pruning trajectory control system based on a depth camera, which uses a depth camera to scan target trees, reconstructs an environmental model of the target trees through a PC, analyzes the model data, and fits the center line of the model crown; then, the center line data is input into a control programming mechanism, and the operation parameters are set. After the linear module workbench receives the target object position signal, it generates a PWM wave signal to control the movement direction and speed of the stepper motor, and the actuator is controlled to return to the center line position with the set parameters, and starts to move according to a preset trajectory. After completing the pruning operation around the target tree, it can be reset and continue to repeat the above pruning operation. The present invention is based on a depth camera, and by acquiring target point cloud information and modeling, it can help it accurately identify the operation object and improve the operation production efficiency. At the same time, it realizes the informatization, intelligence and precision of mechanical equipment, and can be used for highway green belts, park greening pruning, etc., to achieve the operation requirements of the motion state regulation of the pruning device.

[0036] Furthermore, the PC uses a hybrid algorithm of RANSAC and LSC to visualize the center lines of trees and reduce the impact of the noisy point cloud on the fitting accuracy.

[0037] Furthermore, the control programming organization uses LabVIEW graphical programming to develop a human-computer interaction interface that is simple to understand and has low operating difficulty.

[0038] Furthermore, the actuator adopts an XYZ space workbench, which greatly reduces the development difficulty and cost compared to the traditional model, is simple and reliable to operate, improves the working efficiency of the operating device, and achieves the requirement of precise operation by controlling the trajectory of the actuator. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 It is a structural schematic diagram of the green plant pruning trajectory control system based on the depth camera of the present invention.

[0041] Figure 2 This is a diagram of the experimental platform device of the present invention.

[0042] Figure 3 This is a flow chart of the error-free fitting algorithm in the Z direction of the present invention.

[0043] Among them: 1-depth camera, 2-L-type connecting plate, 3-XYZ linear module, 4-support frame. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0045] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0047] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0048] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0049] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0051] See also Figure 1 The present invention provides a green plant pruning trajectory control system based on a depth camera, including a depth camera, a PC, a control programming mechanism and an execution mechanism. The depth camera is connected to the PC, the control programming mechanism and the execution mechanism through data to issue control instructions and receive relevant data results.

[0052] The depth camera is used to collect and store the three-dimensional point cloud of a single tree, and send the acquired three-dimensional point cloud data to a PC through a data connection for analysis and processing.

[0053] The PC is used to analyze and reconstruct the 3D point cloud collected by the depth camera, and send the analysis and 3D reconstruction results to the control programming organization through data.

[0054] The control programming mechanism is used to program the green plant pruning trajectory according to the analysis and three-dimensional reconstruction results, set the pruning shape, speed, starting point and other parameters, and send them to the actuator through the data connection.

[0055] The actuator moves and trims the green plants along the trimming trajectory according to the commands sent by the control programming mechanism.

[0056] like Figure 2 In the simulation experimental platform shown, the depth camera 1 is connected to the XYZ linear module 3 through the L-shaped connecting plate 2, the XYZ linear module 3 is supported by the support frame 4, the XYZ linear module 3 is connected to the stepper motor, and the depth camera 1 is installed on the actuator at the same position as the pruning tool. After the experimental platform is built, the experimental platform is placed above the target tree, that is, the target tree is in the experimental platform. Driven by the stepper motor, the XYZ linear module 3 is driven to move in a straight line, thereby driving the depth camera 1 to dynamically scan the target, the PC completes the modeling task, and finally the stepper motor drives the actuator to complete the pruning.

[0057] The present invention obtains scene data information of target trees through a depth camera, performs 3D reconstruction on the 3D point cloud of the target trees, and uses the Bundle Fusion algorithm to fuse and register the depth and color of the trees to achieve high-quality real-time 3D reconstruction.

[0058] After obtaining the target tree 3D model, point cloud denoising is performed. The MATLAB platform is used to filter the noise points and segment the unnecessary background in the target tree 3D model, and the radius filtering method is used to perform point cloud denoising on the target tree 3D model.

[0059] The PC includes a 3D reconstruction module and a crown center visualization module:

[0060] The 3D reconstruction module is used to optimize the 3D point cloud information collected at depth and generate a 3D model of the target tree through Poisson reconstruction.

[0061] The crown center visualization module is used to perform fitting analysis on the 3D model of the target tree generated by the 3D reconstruction module. It includes a circle fitting unit and a space straight line fitting unit:

[0062] The circle fitting unit is used to perform hierarchical fitting on the three-dimensional model of the target tree, and the least squares circle (LSC) and RANSAC hybrid algorithm is used to find the center point of each layer.

[0063] The spatial straight line fitting unit is used to perform spatial straight line fitting on the center of the circle generated by the circle fitting unit to find a fitting straight line with the minimum error.

[0064] For spatial straight line fitting, the present invention proposes a Z-direction error-free spatial straight line fitting algorithm, which projects the point cloud of the space onto a two-dimensional plane (XoY plane), calculates the average value of the scattered points in the X and Y directions, and uses X mean and Y mean Indicates, as the coordinates of the fitting lines in the X and Y directions, respectively. The straight line fitting method has no error in the Z direction, and the obtained fitting line intersects the XoY plane perpendicularly. Figure 3 As shown, the center point of the fitting circle of each cross section is extracted, the obtained center point is detected for outliers, the center coordinates of the center point are calculated using the weighted average method, and a straight line fitting is performed.

[0065] The error-free spatial straight line fitting algorithm in the Z direction adopts a distance-based outlier detection method. This method considers that the outliers are far away from the normal points. Therefore, the K-nearest neighbor distance (or average distance) of each data point can be calculated and compared with the threshold. If the distance is greater than the threshold, it is considered to be an outlier and removed. The specific steps are as follows:

[0066] (1) Set the outlier detection parameters and the distance threshold T. Points exceeding this distance are considered outliers.

[0067] (2) Assume that there are two points A(x1, y1) and B(x2, y2) on a two-dimensional plane. The Euclidean distance between them is:

[0068]

[0069] (3) For a data point P(x3,y3) in a data set, its average Euclidean distance to all other data points is:

[0070]

[0071] Where N represents the total number of data points in the data set, (x 2,i,y 2,i ) represents other data points in the data set;

[0072] (4) Determine outliers based on the threshold. When , the point is considered as an outlier and is removed.

[0073] In the processing of the circle center point after removing the outliers, the error-free spatial straight line fitting algorithm in the Z direction uses the weighted average method to calculate the center point coordinates.

[0074] Calculate the mean of the center points of the circle. The specific calculation expression is as follows:

[0075]

[0076] Where x0 and y0 represent the coordinate values ​​of the mean point O respectively.

[0077] Calculate the distance from each point to the center point, and use its reciprocal as the weight. The specific calculation expression is as follows:

[0078]

[0079] The weighted average center is calculated according to the weight. The specific calculation expression is as follows:

[0080]

[0081] The weighted center point coordinates are This point is used as the value of the fitting line in the X and Y directions. The fitting line expression is:

[0082]

[0083] After processing, the position of the center line of the target tree can be obtained, and the center line of the target tree can be visualized by combining the fitted spatial straight line with the three-dimensional model of the target tree.

[0084] The control programming mechanism is used to program the green plant pruning trajectory. LabVIEW graphical programming is used to input the center line data of the spatial linear fitting unit to the control parameter terminal to set the motion trajectory, speed, acceleration, starting point and other parameters. The green plant pruning trajectory can prune three types of trajectories, including cylindrical pruning trajectory, conical pruning trajectory and spherical pruning trajectory.

[0085] Cylindrical pruning trajectory trees are pruned by circular interpolation of the XY axis and fixed-length motion of the Z axis. The specific steps are as follows:

[0086] (1) Set the initial arc radius, that is, set the coordinates of the starting point and the end point, and perform arc interpolation on the XY axis;

[0087] (2) After the first circle is drawn, the Z axis moves downward a set distance and starts drawing a circle again;

[0088] (3) Repeat the above steps.

[0089] The pruning of trees with conical pruning trajectories is similar to that of trees with cylindrical pruning trajectories. The difference is that the radius of the circular interpolation motion gradually increases from top to bottom during pruning. To address this change, the following methods can be used:

[0090] Enter the minimum trimming radius r on the control interface min With the maximum trimming radius r max , the trimming distance DZ and trimming density N in the Z direction are programmed through the Math Script node to calculate the distance r that the radius of the trimming circle increases each time d and the falling distance d Z , the specific calculation expression is as follows:

[0091] r d =(r max -r min ) / N

[0092] d Z =D Z / N

[0093] Input trimming parameters and transmit the output nodes to the control program. Each time a complete circular trajectory is executed, the actuator drops d z distance, so that the radius increases to (r min +n×r d ), where n is the number of times in the current loop, and the above operation is repeated until the loop ends.

[0094] The pruning of trees with spherical pruning trajectories is also achieved through arc interpolation of the XY axis and fixed-length motion of the Z axis. After completing a round of pruning, the tree moves a certain distance on the sphere through fixed-length motion of the Z axis before the next round of pruning. The dynamic adjustment of the pruning radius is similar to that of trees with conical pruning trajectories, but the spherical pruning trajectory control program uses the MATLAB Script node method to fit the maximum inscribed sphere in space to the tree model point cloud data to obtain the maximum inscribed sphere radius.

[0095] In the control program, the radius of the maximum fitting sphere is used as parameter r max Input, the rest of the process is similar to the conical pruning trajectory hedge pruning. After each cycle, update the current radius, that is, increase the radius of the pruning circle. By repeating the above steps, the radius of the pruning circle is gradually expanded until the maximum radius r is reached. max Then it gradually decreases to r min , completing the entire spherical pruning.

[0096] The actuator is used to receive commands from the control programming mechanism and move along the green plant pruning trajectory. It includes a return to origin module, a fixed-length acceleration and deceleration module, a multi-axis linear interpolation module and a two-axis circular interpolation module.

[0097] The homing module is used to return the linear module to the origin, receive the reset command from the control programming mechanism, and run the linear module back to the origin;

[0098] The fixed-length acceleration and deceleration module is used to perform fixed-length movement on the linear module. It can select T-type acceleration and deceleration curve or S-type acceleration and deceleration curve. It stops moving after running to the running length input by the control programming mechanism.

[0099] The multi-axis linear interpolation module is used to perform linked linear motion in the XYZ directions on the linear module;

[0100] The two-axis circular interpolation module is used to perform circular motion on any two axes of the linear module, and receives the interpolation start and end point coordinate position information output by the control programming mechanism.

[0101] The green plant pruning trajectory control system based on the depth camera of the present invention obtains the three-dimensional point cloud information modeling of the target tree by using the depth camera, and displays the modeling information and the center fitting information on the PC through the crown center visualization module for the staff to observe and refer to intuitively, and then cooperates with the control programming mechanism and the execution mechanism to realize the control of the green plant pruning trajectory, thereby improving the pruning accuracy and pruning efficiency.

[0102] The green plant pruning trajectory control method based on the depth camera of the present invention is as follows:

[0103] When preparing for pruning operations, first use a depth camera to scan the target trees and store the 3D point cloud of a single tree; use a PC to analyze and reconstruct the collected 3D point cloud to fit the center line of the model crown; then, input the center line data into the control programming mechanism, which can program the green plant pruning trajectory according to the analysis and 3D reconstruction results to set the pruning operation parameters; after the XYZ linear module workbench receives the target object position signal, it generates a PWM wave signal to control the movement direction and speed of the stepper motor to set the pruning parameter movement to control the pruning tool to return to the center line position, and start moving according to the preset green plant pruning trajectory. After completing the pruning operation around the target tree, it resets and continues to repeat the above pruning operation.

[0104] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A green plant pruning trajectory control system based on a depth camera, characterized in that: It includes a depth camera, a PC, a control programming mechanism and an execution mechanism, wherein the depth camera is communicatively connected with the PC, the control programming mechanism and the execution mechanism; The depth camera is used to collect and store the three-dimensional point cloud of a single tree, and send the acquired point cloud data to a PC for analysis and processing; The PC is used to analyze and reconstruct the three-dimensional point cloud collected by the depth camera, and send the analysis and three-dimensional reconstruction results to the control programming mechanism; The control programming mechanism is used to program the green plant pruning trajectory according to the analysis and three-dimensional reconstruction results, set the pruning parameters, and send them to the execution mechanism; The actuator is used to move and trim along the green plant trimming trajectory according to the command sent by the control programming mechanism.

2. According to the green plant pruning trajectory control system based on a depth camera according to claim 1, it is characterized in that: The three-dimensional reconstruction process includes: According to the 3D point cloud data information of the trees obtained by the depth camera, the Bundle Fusion algorithm is used to fuse and align the depth and color of the trees to achieve real-time 3D reconstruction.

3. The green plant pruning trajectory control system based on a depth camera according to claim 1, characterized in that: The PC includes a three-dimensional reconstruction module and a crown center visualization module; The 3D reconstruction module is used to optimize the 3D point cloud information collected by the depth camera and generate a 3D model of the target tree through Poisson reconstruction; The crown center visualization module is used to perform fitting analysis on the three-dimensional model of the target tree generated by the three-dimensional reconstruction module.

4. The green plant pruning trajectory control system based on a depth camera according to claim 3, characterized in that: The crown center visualization module includes a circle fitting unit and a space straight line fitting unit; The circle fitting unit is used to perform hierarchical fitting on the three-dimensional model of the target tree, and uses the least squares circle and RANSAC hybrid algorithm to find the center point of each layer; The space straight line fitting unit is used to perform space straight line fitting on the circle center point generated by the circle fitting unit to obtain a fitting straight line with the minimum error.

5. The green plant pruning trajectory control system based on a depth camera according to claim 3, characterized in that: The process of performing point cloud denoising on the three-dimensional model of the target tree is as follows: The MATLAB platform is used to filter the noise points and segment the unnecessary background in the three-dimensional model of the target trees, and the radius filtering method is used to perform point cloud denoising on the three-dimensional model of the target trees.

6. The green plant pruning trajectory control system based on a depth camera according to claim 1, characterized in that: The control programming mechanism programs the green plant pruning trajectory through LabVIEW graphical programming, transmits the center line data of the spatial straight line fitting unit to the control parameter end, and sets the green plant pruning trajectory, speed, acceleration and starting point parameters.

7. The green plant pruning trajectory control system based on a depth camera according to claim 1, characterized in that: The actuator includes a return-to-origin module, a fixed-length acceleration / deceleration module, a multi-axis linear interpolation module, and a two-axis circular interpolation module; The homing module is used to perform homing movement on the linear module, receive a reset command from the control programming mechanism, and run the linear module back to the origin; The fixed-length acceleration and deceleration module is used to perform fixed-length movement on the linear module, and stops moving after running to the running length input by the control programming mechanism; The multi-axis linear interpolation module is used to perform linked linear motion in the XYZ directions on the linear module; The two-axis circular interpolation module is used to perform circular motion on any two axes of the linear module, and receives the interpolation start and end point coordinate position information output by the control programming mechanism.

8. The green plant pruning trajectory control system based on a depth camera according to claim 1, characterized in that: The depth camera is installed on the actuator, the depth camera is connected to the linear module through an L-shaped connecting plate, the linear module is supported by a support frame, the linear module is connected to a stepper motor, the stepper motor drives the linear module to perform linear motion, and the depth camera dynamically scans the target, thereby driving the actuator to complete trimming.

9. The green plant pruning trajectory control system based on a depth camera according to claim 1, characterized in that: The green plant pruning trajectory includes a cylindrical pruning trajectory, a conical pruning trajectory and a spherical pruning trajectory.

10. A green plant pruning trajectory control method based on a depth camera, characterized in that: include: Collect and store 3D point clouds of individual trees; Analyze and reconstruct the collected 3D point cloud to obtain analysis and 3D reconstruction results; Program the plant pruning trajectory and set pruning parameters based on the analysis and 3D reconstruction results; According to the set pruning parameters, it moves and prunes along the green plant pruning trajectory.

Images