Method, device and equipment for controlling row-to-row garden patrolling and insect repelling of cherry tomato plant factory and medium

By constructing a grid map using 3D LiDAR and point cloud filtering algorithms, and combining adaptive Monte Carlo localization and improved path planning algorithms, the applicability and efficiency of the cherry tomato orchard patrol and pest control robot in the plant factory were solved, achieving efficient and safe navigation control.

CN118960735AInactive Publication Date: 2024-11-15SOUTH CHINA AGRICULTURAL UNIVERSITY +1

Patent Information

Application Number
CN202411010812.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The cherry tomato orchard patrol and pest control robot suffers from excessive point cloud noise in the environmental map construction between rows in the plant factory, resulting in poor applicability of path planning and seriously affecting operational efficiency and safety.

Method used

A grid map is constructed using 3D LiDAR combined with point cloud filtering algorithms. It is then combined with adaptive Monte Carlo positioning and an improved Dijkstra global path planning algorithm, along with TEB local path planning, to optimize path planning to adapt to complex environments.

Benefits of technology

It improves the navigation accuracy and efficiency of robots between rows in plant factories, reduces dependence on external equipment, reduces cumulative errors and costs, and optimizes the smoothness and applicability of paths.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118960735A_ABST
    Figure CN118960735A_ABST
Patent Text Reader

Abstract

The invention relates to a cherry tomato plant factory line-to-line garden patrolling and insect repelling control method, device and equipment and a medium, and the method comprises the steps: tracking the pose of a cherry tomato garden patrolling and insect repelling robot in real time according to an initial pose, odometer data and a grid map through employing a self-adaptive Monte Carlo positioning algorithm; the real-time positioning result of the cherry tomato garden patrol insect expelling robot is determined; determining each target desinsectization operation point, and determining a global optimal desinsectization operation path of the cherry tomato garden patrol desinsectization robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden patrol desinsectization robot and each target desinsectization operation point; and adopting a preset local path planning algorithm to carry out obstacle avoidance correction on the global optimal insect expelling operation path so as to determine a safe insect expelling operation path of the cherry tomato garden patrol insect expelling robot. The global path planned by the method has the characteristics of being smoother and straighter, and the effect of planning the path is remarkably optimized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer vision, and in particular to a method for controlling pests in a cherry tomato plant factory by patrolling between rows, a corresponding device, an electronic device, and a computer-readable storage medium. Background Art

[0002] my country is the world's largest tomato producer, with the total tomato planting area accounting for 1 / 3 of the world's total. With the continuous progress and development of the cherry tomato planting industry, its planting area has expanded rapidly across the country. At present, the planting area of ​​winter cherry tomatoes has exceeded 500,000 mu. With the expansion of the planting area of ​​cherry tomato plant factories and the increase in labor costs, the demand for large-scale autonomous operations has been valued and developed. The current mainstream operation mode in the cherry tomato plant factory environment is manual operation, and the operation efficiency is low. In recent years, agricultural patrol robots based on autonomous navigation technology have been developed, which have the characteristics of high autonomous navigation efficiency and high degree of automation. Ultrasonic pest control is a process in which humans use the high frequency, high energy and bionic characteristics of ultrasound to directly cause physical damage to pests or indirectly interfere with their auditory system. In recent years, it has been widely studied by many scholars. Therefore, the ultrasonic pest control equipment deployed on the robot can effectively improve the operation efficiency in the cherry tomato plant factory environment. It is of positive significance to design a cherry tomato patrol pest control robot suitable for autonomous operation in a large-scale plant factory environment.

[0003] The autonomous navigation working mode of the cherry tomato patrol and pest control robot is to move in fixed points and straight lines between rows. The navigation design using 3D lidar revolves around map construction, self-positioning and path planning. In the current mainstream navigation design, the constructed environment map has the problem of too many point cloud noise points, and the path planned by the traditional path planning algorithm has poor applicability between the rows of the plant factory where the cherry tomato patrol and pest control robot operates, which seriously reduces the efficiency and safety of the robot's operation.

[0004] To sum up, the environmental map constructed by the cherry tomato garden patrol and insect repellent robot in the prior art has a lot of point cloud noise, and the planned path has poor applicability between the rows of the plant factory where the cherry tomato garden patrol and insect repellent robot operates, which seriously reduces the efficiency and safety of the robot's operation. In order to solve this problem, the applicant has made corresponding explorations. Summary of the invention

[0005] The purpose of the present application is to solve the above-mentioned problems and to provide a method for controlling insect repellent in the rows of a cherry tomato plant factory, a corresponding device, an electronic device and a computer-readable storage medium.

[0006] In order to meet the various objectives of this application, this application adopts the following technical solutions:

[0007] A method for controlling insects in a cherry tomato plant factory between rows is proposed to meet one of the purposes of the present application, comprising:

[0008] Receive the task of patrolling the cherry tomato garden and removing insects, and obtain the lidar point cloud data from the lidar;

[0009] A preset point cloud map construction algorithm is used to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, a back-end loop test is performed on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and the point cloud map is converted into a grid map for navigation of the cherry tomato garden patrol insect repellent robot;

[0010] Determine the initial posture and odometer data corresponding to the cherry tomato garden patrol and insect repellent robot, and use an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot;

[0011] Determine each target insect repellent operation point, and determine a global optimal insect repellent operation path of the cherry tomato garden insect repellent robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden insect repellent robot, and each target insect repellent operation point;

[0012] A preset local path planning algorithm is used to perform obstacle avoidance correction on the globally optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol insect repellent robot, and the cherry tomato garden patrol insect repellent robot is controlled to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation, so as to complete the garden patrol insect repellent control of cherry tomatoes.

[0013] Optionally, a step of using a preset point cloud map construction algorithm to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, and performing a back-end loopback test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory includes:

[0014] In response to the point cloud map construction instruction, the laser radar point cloud data is input into the Cartographer algorithm, and the local mapping module in the Cartographer algorithm generates a sub-map of the inter-row environment of the plant factory, and the point cloud map construction algorithm is the Cartographer algorithm;

[0015] The global mapping module in the Cartographer algorithm performs a back-end loop test on the sub-map of the inter-row environment of the plant factory, and corrects the map by detecting the error between the sub-maps at adjacent moments to determine the sub-map after the loop test;

[0016] The sub-map after the loop test is subjected to posture optimization to generate a point cloud map of the inter-row environment of the plant factory.

[0017] Optionally, after the step of determining the point cloud map of the inter-row environment of the plant factory, the method further comprises:

[0018] A preset straight-through filtering algorithm is used to post-process the point cloud map of the inter-row environment of the plant factory, and based on the straight-through filtering algorithm and a preset height threshold, the ground point cloud noise points and the invalid point clouds above the motion range of the cherry tomato garden patrol insect repellent robot are filtered and optimized;

[0019] Determine the filter radius and the number of filtered neighboring point clouds, and set the Euclidean distance radius range for each point based on the preset radius filtering algorithm according to the filter radius and the number of filtered neighboring point clouds. If the number of point clouds within the range is less than the number of filtered neighboring point clouds, the point is filtered out as an outlier to determine the point cloud map of the inter-row environment of the plant factory.

[0020] Optionally, the step of using an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot includes:

[0021] In response to a self-positioning instruction of the cherry tomato garden patrol and insect repellent robot, a preset adaptive Monte Carlo positioning algorithm is used to generate particles obeying a Gaussian distribution near an initial position of the cherry tomato garden patrol and insect repellent robot, wherein the particles represent a possible position distribution of the cherry tomato garden patrol and insect repellent robot;

[0022] The corresponding position and posture of each particle is updated according to the odometer data of the cherry tomato garden patrol and insect repellent robot to determine the updated particle position and posture, the observation model of each particle is obtained by using the laser radar point cloud data and the likelihood domain model of the grid map, and the particle is assigned a corresponding position and posture weight;

[0023] The pose weight of each particle is corrected according to the measurement probability of the laser radar point cloud data, so as to obtain the particle pose and pose weight that are continuously iteratively updated during the movement of the cherry tomato garden patrol insect repellent robot;

[0024] According to the pose weights of the particles, the current pose corresponding to the cherry tomato garden patrol and insect repellent robot is estimated, and based on the current pose corresponding to the cherry tomato garden patrol and insect repellent robot, the laser radar point cloud data is converted into a point cloud data set in the same coordinate system as the grid map;

[0025] Each frame of the point cloud data set is input into a preset iterative closest point algorithm to determine the rotation matrix and translation matrix of the point cloud data set of the current frame relative to the grid map, and the real-time positioning result of the cherry tomato garden patrol and insect repellent robot in the grid map is determined based on the rotation matrix and the translation matrix.

[0026] Optionally, the global path planning algorithm is an improved Di jkstra global path planning algorithm; and the local path planning algorithm is a TEB local path planning algorithm.

[0027] Optionally, the step of determining each target expelling operation point, and determining a global optimal expelling operation path of the cherry tomato garden expelling insect robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden expelling insect robot and each target expelling insect operation point comprises:

[0028] Construct an open list and a closed list to store the path points to be determined and the determined path points respectively. Take the starting point as the origin to traverse and calculate the Euclidean distance and Manhattan distance to the nearby nodes. When the shortest path is found, put the node into the corresponding path list, and traverse and calculate again with the shortest path point as the center until the shortest path to the target point is found. This path is the global shortest path.

[0029] Determine a preset cost threshold of the grid map, introduce potential cost values ​​into the open list to perform secondary traversal of nodes, perform secondary cost value traversal of multiple directions adjacent to the current non-obstruction node, if the node cost values ​​of multiple directions of the current adjacent node exceed the preset cost threshold, move it to the overflow list and no longer perform subsequent traversal, if it does not exceed the preset cost threshold, put it into the retention list for subsequent path planning;

[0030] The generated path is traversed and expanded to increase the density of path points. The absolute distance between two adjacent points is calculated by traversing the path points. If the absolute distance is greater than a preset length threshold, bilinear interpolation optimization is performed on the original path.

[0031] Based on the normalization function, the angle values ​​of the front and rear paths are normalized to obtain the processed angle values, so that the angle range is 0 to 2π, and the path is re-traversed to calculate the normalized angle between the current node and the adjacent node in the list. If the calculated angle exceeds the preset angle threshold, the adjacent node is optimized according to the median interpolation method;

[0032] Extract the path list based on the preset JPS jump point search algorithm, take the initial node as the parent node, if the newly generated path connected to the second node after skipping the first node does not collide with the obstacle, and the potential cost value of the interpolated child nodes of the path is less than the preset cost threshold, then cut the first node to form a new path from the initial node to the second node, continue to traverse the third node until the path collides or the potential cost value is greater than or equal to the preset cost threshold, the path list is closed to form a new sub-path list, and the above steps are repeated with the next node of the end node as the parent node;

[0033] The global planning path after straightening and smoothing of curves is output to determine the global optimal insect repellent operation path of the cherry tomato garden insect repellent robot.

[0034] Optionally, the step of controlling the cherry tomato garden patrol and insect repellent robot to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation includes:

[0035] In response to the cherry tomato pest control operation instruction, determine a safe pest control operation path for the cherry tomato garden pest control robot, and reach each pest control operation target point based on the safe pest control operation path;

[0036] Acquire an image of a cherry tomato plant in a binocular camera at each of the target points of the insect repellent operation, and perform target detection on the image of the cherry tomato plant based on a preset tomato fruit detection model to determine a target detection frame corresponding to the cherry tomato;

[0037] The cherry tomato garden patrol and insect repellent robot is driven to perform ultrasonic insect repellent on the target tomato fruit according to the target detection frame according to a preset frequency threshold, so as to complete the garden patrol and insect repellent control of the cherry tomatoes.

[0038] A cherry tomato plant factory inter-row patrol insect control device adapted to another purpose of the present application comprises:

[0039] A data acquisition module is configured to receive a cherry tomato garden patrol and insect repellent task and obtain laser radar point cloud data from the laser radar;

[0040] A grid map determination module is configured to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data using a preset point cloud map construction algorithm, perform a back-end loop test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and convert the point cloud map into a grid map for navigation of the cherry tomato garden patrol insect repellent robot;

[0041] A positioning result determination module is configured to determine an initial posture and odometer data corresponding to the cherry tomato garden patrol and insect repellent robot, and use an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine a real-time positioning result of the cherry tomato garden patrol and insect repellent robot;

[0042] A global path determination module is configured to determine each target insect repellent operation point, and determine a global optimal insect repellent operation path of the cherry tomato garden insect repellent robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden insect repellent robot and each target insect repellent operation point;

[0043] The insect repellent operation control module is configured to use a preset local path planning algorithm to perform obstacle avoidance correction on the globally optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol insect repellent robot, and control the cherry tomato garden patrol insect repellent robot to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operations, so as to complete the cherry tomato garden patrol insect repellent control.

[0044] An electronic device provided to meet another purpose of the present application includes a central processing unit and a memory, wherein the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the inter-row patrol and insect control method for a cherry tomato plant factory described in the present application.

[0045] A computer-readable storage medium is provided to meet another purpose of the present application, which stores a computer program implemented according to the inter-row patrol and insect control method of the cherry tomato plant factory in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

[0046] Compared with the prior art, the present application aims at solving the problems that the environment map constructed by the cherry tomato garden patrol and insect repellent robot in the prior art has many point cloud noise points, the planned path has poor applicability between the rows of the plant factory where the cherry tomato garden patrol and insect repellent robot operates, and the efficiency and safety of the robot operation are seriously reduced. The present application includes but is not limited to the following beneficial effects:

[0047] First, the method for controlling pest control in a cherry tomato plant factory between rows is described in the present application. A 3D laser radar is combined with a point cloud filtering algorithm to finally construct a grid map required for navigation. The method has the advantages of stronger environmental applicability and richer perception information. The adaptive Monte Carlo positioning scheme that integrates robot posture and mileage information for multi-sensor fusion has the advantage of stronger robustness. The traversal path smoothing interpolation optimization method and the path optimization method based on JPS jump point search are introduced to improve the Dijkstra global path planning algorithm. The global path planned by the algorithm has the characteristics of being smoother and straighter. The local path planning is performed in combination with the TEB local path planning algorithm that meets the minimum turning radius constraint condition of the cherry tomato patrol pest control robot, thereby optimizing the effect of the planned path.

[0048] Secondly, the present application is suitable for driving navigation of a cherry tomato garden patrol and insect repellent robot in the inter-row operation scenario of a plant factory. The overall navigation method is designed using a 3D laser radar, and the algorithm is improved in environmental adaptability in map construction and path planning. The traditional navigation method has poor mapping and path planning effects in the inter-row environment of a plant factory, and finally provides a set of efficient and safe navigation methods suitable for cherry tomato garden patrol and insect repellent robot inter-row operations in plant factories;

[0049] Third, the Cartographer algorithm of the present application locates the position of the cherry tomato patrol and insect repellent robot while constructing the point cloud map of the inter-row environment of the plant factory. By performing a back-end loop test on the sub-map of the inter-row environment of the plant factory, the cumulative error of the algorithm can be significantly reduced. When constructing the point cloud map of the inter-row environment of the plant factory, it does not rely too much on external equipment, and a relatively accurate map can be run through a laser radar with lower precision, which greatly reduces the cost of the robot;

[0050] Fourthly, the adaptive Monte Carlo positioning algorithm of the present application can dynamically adjust the number and distribution of particles to adapt to different environmental conditions and accuracy requirements. In complex environments such as between rows in plant factories or when higher positioning accuracy is required, increasing the number of particles can improve the accuracy of positioning, and adjusting the resampling strategy and sensor model according to robot movement and environmental changes to improve the robustness and efficiency of positioning;

[0051] Fifth, the improved Di jkstra global path planning algorithm of the present application introduces a traversal path smoothing interpolation optimization method to smooth and optimize the curved path to reduce inflection points and turning angles, and introduces a path optimization method based on JPS jump point search to perform node pruning on straight sections to straighten the path, so that the global path planned by the cherry tomato garden patrol and insect repellent robot is more suitable for working between rows in plant factories. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0053] Figure 1 This is a flow chart of the method for controlling insect repellent in the cherry tomato plant factory rows in the embodiment of the present application;

[0054] Figure 2 This is a schematic diagram of an exemplary scenario used in the method for controlling insect repellent in the cherry tomato plant factory rows in the embodiment of the present application;

[0055] Figure 3 A schematic diagram of a process for determining a point cloud map of an inter-row environment of a plant factory in an embodiment of the present application;

[0056] Figure 4 This is a schematic diagram of a process for post-processing a point cloud map using a straight-through filtering algorithm and a radius filtering algorithm in an embodiment of the present application;

[0057] Figure 5 A schematic diagram of a process for determining the real-time positioning result of the cherry tomato garden patrol and insect repellent robot in an embodiment of the present application;

[0058] Figure 6 A schematic diagram of a process for determining a globally optimal insect repellent operation path of a cherry tomato garden insect repellent robot in an embodiment of the present application;

[0059] Figure 7 This is a schematic diagram of a process for controlling the cherry tomato garden patrol and insect repellent robot to reach each insect repellent operation target point to perform insect repellent operation according to a safe insect repellent operation path in an embodiment of the present application;

[0060] Figure 8 This is a schematic diagram of the kinematic analysis of the cherry tomato garden patrol and insect repellent robot using the Ackerman structure in the embodiment of the present application;

[0061] Fig. 9 This is a functional block diagram of a cherry tomato plant factory row patrol and insect repellent control device in an embodiment of the present application;

[0062] Fig.10 It is a schematic diagram of the structure of the computer device in the embodiment of the present application. DETAILED DESCRIPTION

[0063] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0064] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0065] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0066] It will be understood by those skilled in the art that the terms "client", "terminal" and "terminal device" as used herein include both devices with wireless signal receivers, which are devices with only wireless signal receivers without transmission capability, and devices with receiving and transmitting hardware, which are devices with receiving and transmitting hardware capable of two-way communication over a two-way communication link. Such devices may include: cellular or other communication devices such as personal computers, tablet computers, which have single-line displays or multi-line displays or cellular or other communication devices without multi-line displays; PCS (Personal Communications Services) which may combine voice, data processing, fax and / or data communication capabilities; PDA (Personal Digital Assistant) which may include a radio frequency receiver, pager, Internet / intranet access, web browser, notepad, calendar and / or GPS (Global Positioning System) receiver; conventional laptop and / or palmtop computers or other devices which have and / or include a radio frequency receiver. The "client", "terminal" and "terminal device" used herein may be portable, transportable, installed in a vehicle (air, sea and / or land), or adapted and / or configured to run locally, and / or in a distributed form, at any other location on the earth and / or in space. The "client", "terminal" and "terminal device" used herein may also be a communication terminal, an Internet terminal, a music / video playing terminal, such as a PDA, a MID (Mobile Internet Device) and / or a mobile phone with a music / video playing function, or may be a smart TV, a set-top box and other devices.

[0067] The hardware referred to by the names such as "server", "client", and "service node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer. It is a hardware device with the necessary components revealed by the von Neumann principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device. The computer program is stored in its memory, and the central processing unit calls the program stored in the external memory into the internal memory for execution, executes the instructions in the program, and interacts with the input and output devices to complete specific functions.

[0068] It should be pointed out that the concept of "server" referred to in this application can also be extended to the case of server clusters. According to the network deployment principle understood by those skilled in the art, the servers should be logically divided. In physical space, these servers can be independent of each other but can be called through interfaces, or integrated into a physical computer or a set of computer clusters. Those skilled in the art should understand this flexibility, and should not use it to restrict the implementation of the network deployment method of this application.

[0069] Unless expressly specified, one or more technical features of the present application can be deployed on a server for implementation and accessed by a client through a remote call to obtain an online service interface provided by the server, or can be directly deployed and run on a client for access.

[0070] The neural network models referenced or may be referenced in this application, unless expressly specified, can be deployed on a remote server and remotely called on the client, or can be deployed and directly called on a client with sufficient device capabilities. In some embodiments, when it runs on the client, its corresponding intelligence can be obtained through transfer learning to reduce the requirements for the client's hardware operating resources and avoid excessive occupation of the client's hardware operating resources.

[0071] Unless explicitly specified, the various data involved in this application can be stored remotely on a server or on a local terminal device, as long as it is suitable for being called by the technical solution of this application.

[0072] Those skilled in the art should be aware that, although the various methods of the present application are described based on the same concept and thus present commonality to each other, unless otherwise specified, these methods can be independently executed. Similarly, for each embodiment disclosed in the present application, they are all proposed based on the same inventive concept, therefore, concepts with the same expression, and concepts that are appropriately changed for convenience despite different expressions, should be understood as equivalent.

[0073] Unless the mutually exclusive relationship between the embodiments to be disclosed in this application is explicitly stated, the relevant technical features involved in each embodiment can be cross-combined to flexibly construct a new embodiment, as long as such combination does not deviate from the creative spirit of this application and can meet the needs of the prior art or solve certain deficiencies in the prior art. Those skilled in the art should be aware of this flexibility.

[0074] See also Figure 1 as well as Figure 2 In one embodiment, the inter-row patrol insect control method of the cherry tomato plant factory of the present application includes:

[0075] Step S10, receiving a cherry tomato garden patrol and insect repellent task, and obtaining laser radar point cloud data from the laser radar;

[0076] The cherry tomato garden patrol and insect repellent robot can receive cherry tomato insect repellent operation tasks issued by terminal devices, and the terminal devices can be mobile terminals or computer terminals, etc. When the cherry tomatoes in the plant factory are ripe, the managers in the plant factory can issue insect repellent operation instructions to the cherry tomato garden patrol and insect repellent robot through the mobile terminal or computer terminal. After receiving the cherry tomato garden patrol and insect repellent task, the robot scans the inter-row environment of the plant factory through a laser radar to obtain laser radar point cloud data from the laser radar, etc., and the laser radar point cloud data can be three-dimensional laser radar point cloud data.

[0077] Step S20, using a preset point cloud map construction algorithm to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, performing a back-end loop test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and converting the point cloud map into a grid map for navigation of the cherry tomato garden patrol insect repellent robot;

[0078] The point cloud map construction algorithm is a Cartographer algorithm; after obtaining the laser radar point cloud data from the laser radar, a preset Cartographer algorithm is used to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, and a back-end loop test is performed on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and the point cloud map is converted into a grid map for navigation of the cherry tomato garden patrol insect repellent robot;

[0079] In some embodiments, see Figure 3 The steps of using a preset point cloud map construction algorithm to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, and performing a back-end loopback test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory include:

[0080] Step S201, responding to a point cloud map construction instruction, inputting the laser radar point cloud data into the Cartographer algorithm, and generating a sub-map of the inter-row environment of the plant factory by a local mapping module in the Cartographer algorithm;

[0081] Step S203: the global mapping module in the Cartographer algorithm performs a back-end loop test on the sub-map of the inter-row environment of the plant factory, and corrects the map by detecting the error between the sub-maps at adjacent moments to determine the sub-map after the loop test;

[0082] Step S205: optimizing the posture of the sub-map after the loop test to generate a point cloud map of the inter-row environment of the plant factory.

[0083] Specifically, the Cartographer algorithm is a simultaneous localization and mapping (SLAM) construction algorithm based on graph optimization. While constructing the point cloud map of the inter-row environment of the plant factory, the position of the cherry tomato patrol and insect repellent robot is located. When constructing the map, it does not rely on sensor data, that is, the sensor accuracy is the accuracy of the algorithm map construction, but is generated by splicing n (n>1) groups of sub-maps, and each sub-map is iterated by the lidar data from time t to time t+n. After all the sub-maps are generated, the map is corrected by continuously detecting the error between the sub-maps at time t-1 and time t, so as to perform a back-end loop test on the sub-maps of the inter-row environment of the plant factory. By performing a back-end loop test on the sub-maps of the inter-row environment of the plant factory, the cumulative error of the algorithm can be significantly reduced. When constructing the point cloud map of the inter-row environment of the plant factory, it does not rely too much on external equipment, and a relatively accurate map can be run through a lidar with lower accuracy, which greatly reduces the cost of the robot.

[0084] In some embodiments, see Figure 4 , after the steps of determining the point cloud map of the inter-row environment of the plant factory, including:

[0085] Step S2001, post-processing the point cloud map of the inter-row environment of the plant factory using a preset straight-through filtering algorithm, filtering and optimizing the ground point cloud noise points and invalid point clouds above the motion range of the cherry tomato garden patrol insect repellent robot based on the straight-through filtering algorithm according to a preset height threshold;

[0086] Step S2003, determine the filter radius and the number of filtered neighboring point clouds, based on a preset radius filtering algorithm according to the filter radius and the number of filtered neighboring point clouds, and set the Euclidean distance radius range for each point as the basis. If the number of point clouds within the range is less than the number of filtered neighboring point clouds, the point is filtered out as an outlier to determine the point cloud map of the inter-row environment of the plant factory.

[0087] Specifically, a through-filtering algorithm and a radius filtering algorithm are used to post-process the point cloud map of the inter-row environment of a plant factory. First, a preset through-filtering algorithm is used to post-process the point cloud map of the inter-row environment of the plant factory. Based on the through-filtering algorithm, according to a preset height threshold, the ground point cloud noise points and invalid point clouds above the motion range of the cherry tomato garden patrol and insect repellent robot are filtered and optimized; then, the radius filtering algorithm is used to set the filtering radius and the number of filtered neighboring point clouds, and the Euclidean distance radius range is set for each point as the basis. If the number of point clouds within the range is less than the set number of filtered neighboring point clouds, the point is regarded as an outlier and filtered out, and finally an optimized pcd point cloud map is obtained, and it is converted into a raster map for navigation of the cherry tomato garden patrol and insect repellent robot.

[0088] In some embodiments, the filtering principle constraint formula of the straight-through filtering algorithm is:

[0089] z i ∈[z min -Δ min , z max +Δ max ],

[0090] Among them, z i is the z-axis component of the point cloud, z min is the minimum point of the z-axis component point cloud, z max is the maximum value point of the z-axis component point cloud, Δ min is the minimum distance between each point cloud and the preset height threshold, Δ max It is the maximum distance between each point cloud and the preset height threshold.

[0091] Step S30, determining the initial posture and odometer data corresponding to the cherry tomato garden patrol and insect repellent robot, and using an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot;

[0092] After converting the point cloud map corresponding to the inter-row environment of the plant factory into a grid map for navigation of the cherry tomato patrol and insect repellent robot, the initial posture and odometer data corresponding to the cherry tomato patrol and insect repellent robot are determined, and an adaptive Monte Carlo positioning algorithm is used to track the posture of the cherry tomato patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato patrol and insect repellent robot; the adaptive Monte Carlo positioning algorithm can dynamically adjust the number and distribution of particles to adapt to different environmental conditions and accuracy requirements. For example: in complex environments such as inter-row environments of plant factories or when higher positioning accuracy is required, increasing the number of particles can improve the accuracy of positioning, and adjusting the resampling strategy and sensor model according to robot movement and environmental changes to improve the robustness and efficiency of positioning.

[0093] For further information, see Figure 5 The steps of using an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot include:

[0094] Step S301, in response to a self-positioning instruction of the cherry tomato garden patrol and insect repellent robot, a preset adaptive Monte Carlo positioning algorithm is used to generate particles obeying a Gaussian distribution near an initial position of the cherry tomato garden patrol and insect repellent robot, wherein the particles represent a possible position distribution of the cherry tomato garden patrol and insect repellent robot;

[0095] Step S303, updating the corresponding position and posture of each particle according to the odometer data of the cherry tomato garden patrol insect repellent robot to determine the updated particle position and posture, using the laser radar point cloud data and the likelihood domain model of the grid map to obtain the observation model of each particle, and assigning the particle a corresponding position and posture weight;

[0096] Step S305, correcting the posture weight of each particle according to the measurement probability of the laser radar point cloud data, and obtaining the particle posture and posture weight that are continuously iteratively updated during the movement of the cherry tomato garden insect repellent robot;

[0097] Step S307, estimating the current posture corresponding to the cherry tomato garden patrol and insect repellent robot according to the posture weight of the particle, and converting the laser radar point cloud data into a point cloud data set in the same coordinate system as the grid map based on the current posture corresponding to the cherry tomato garden patrol and insect repellent robot;

[0098] Step S309: input each frame of the point cloud data set into a preset iterative closest point algorithm to determine the rotation matrix and translation matrix of the point cloud data set of the current frame relative to the grid map, and calculate and determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot in the grid map based on the rotation matrix and the translation matrix.

[0099] Specifically, the iterative closest point algorithm is an ICP algorithm, which continuously optimizes the minimum distance between the lidar point cloud data and the point cloud map of the inter-row environment of the plant factory in an iterative manner until the best matching effect is achieved.

[0100] Based on the preset adaptive Monte Carlo positioning algorithm, the position of the cherry tomato garden patrol and insect repellent robot is preliminarily estimated, and a group of particles are created to represent the possible position distribution of the robot; according to the motion model of the cherry tomato garden patrol and insect repellent robot, the new position of each particle after the robot moves is predicted; the odometer data of the cherry tomato garden patrol and insect repellent robot is used to evaluate the possibility of the position of each particle, and the corresponding posture weight of the particle is updated accordingly; the posture weight of each particle is corrected according to the measurement probability of the laser radar point cloud data, and the particle weight that is continuously iteratively updated during the movement of the cherry tomato garden patrol and insect repellent robot is obtained. sub-pose and pose weight; according to the pose weight of the particle, estimate the current pose corresponding to the cherry tomato patrol garden pest control robot, and convert the lidar point cloud data into a point cloud data set in the same coordinate system as the grid map based on the current pose corresponding to the cherry tomato patrol garden pest control robot; input each frame of the point cloud data set into a preset iterative closest point algorithm to determine the rotation matrix and translation matrix of the point cloud data set of the current frame relative to the grid map, and calculate and determine the real-time positioning result of the cherry tomato patrol garden pest control robot in the grid map according to the rotation matrix and the translation matrix.

[0101] It can be seen from the above embodiments that the adaptive Monte Carlo positioning algorithm can achieve reliable positioning in the complex dynamic environment between rows of plant factories by performing real-time positioning of the cherry tomato garden patrol and insect repellent robot. By performing parallel calculation and updating of a large number of particles, the adaptive Monte Carlo positioning algorithm can provide higher positioning accuracy and adapt to changes and challenges in different environments.

[0102] Step S40, determining each target expelling operation point, and determining a global optimal expelling operation path of the cherry tomato garden expelling insect robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden expelling insect robot and each target expelling insect operation point;

[0103] After determining the real-time positioning result of the cherry tomato garden patrol and insect repellent robot on the grid map, each target insect repellent operation point is determined, and based on a preset global path planning algorithm, the global optimal insect repellent operation path of the cherry tomato garden patrol and insect repellent robot is determined according to the grid map, the real-time positioning result of the cherry tomato garden patrol and insect repellent robot and the target insect repellent operation points. The global path planning algorithm is an improved Di jkstra global path planning algorithm.

[0104] For further information, see Figure 6 , determining each target expelling operation point, and determining the global optimal expelling operation path of the cherry tomato garden expelling insect robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden expelling insect robot and each target expelling insect operation point, including:

[0105] Step S401, construct an open list and a closed list, which are used to store the path points to be determined and the determined path points respectively, and use the starting point as the origin to traverse and calculate the Euclidean distance and Manhattan distance with the nearby nodes. When the shortest path is found, the node is put into the corresponding path list, and the shortest path point is used as the center to traverse and calculate again until the shortest path to the target point is found. This path is the global shortest path;

[0106] Step S403, determining a preset cost threshold of the grid map, introducing potential cost values ​​into the open list to perform secondary traversal of the nodes, performing secondary cost value traversal of multiple directions adjacent to the current non-obstruction node, if the node cost values ​​of multiple directions of the current adjacent node exceed the preset cost threshold, move it to the overflow list and no longer perform subsequent traversal, if it does not exceed the preset cost threshold, put it into the retention list for subsequent path planning;

[0107] Step S405: traverse and expand the generated path to increase the density of path points, calculate the absolute distance between two adjacent points by traversing the path points, and if the absolute distance is greater than a preset length threshold, perform bilinear interpolation optimization on the original path;

[0108] Step S407: normalize the angle values ​​of the front and rear paths based on the normalization function to obtain processed angle values, so that the angle range is 0 to 2π, re-traverse the path to calculate the normalized angle between the current node and the adjacent node in the list, and if the calculated angle exceeds the preset angle threshold, optimize the adjacent node according to the median interpolation method;

[0109] Step S409, extracting the path list based on the preset JPS jumping point search algorithm, taking the initial node as the parent node, if the newly generated path connected to the second node after skipping the first node does not collide with the obstacle, and the potential cost values ​​of the interpolated child nodes of the path are all less than the preset cost threshold, then the first node is path cut to form a new path from the initial node to the second node, and the third node is traversed until the path collides or the potential cost value is greater than or equal to the preset cost threshold. The path list is closed to form a new sub-path list, and the above steps are repeated with the next node of the end node as the parent node, and the global planning path after straightening and smoothing is output to determine the global optimal insect repellent operation path of the cherry tomato garden insect repellent robot.

[0110] In some embodiments, the preset length threshold is 0.05 m, and the preset angle threshold is π / 4.

[0111] Specifically, the improved Di jkstra global path planning algorithm introduces a traversal path smoothing interpolation optimization method to smooth and optimize the curved path to reduce inflection points and turning angles, and introduces a path optimization method based on JPS jump point search to perform node pruning on straight sections to straighten the path, so that the global path planned by the cherry tomato garden patrol and insect repellent robot is more suitable for inter-row operations in plant factories.

[0112] The processing process of the improved Dijkstra global path planning algorithm includes:

[0113] Step S4001, establish an open list and a closed list, which are used to store the path points to be determined and the determined path points respectively. In order to ensure that the planned path is the optimal path, the starting point is used as the origin to traverse and calculate the Euclidean distance and Manhattan distance with the nearby nodes. When the shortest path is found, the node is put into the corresponding path list, and then the shortest path point is used as the center to traverse and calculate again until the shortest path to the target point is found. This path is the global shortest path;

[0114] Step S4002, design a traversal path smoothing interpolation optimization method, introduce potential cost values ​​into the open list to perform secondary traversal of nodes, the algorithm performs secondary cost value traversal of multiple directions adjacent to the current non-obstruction node by setting a preset cost threshold d of the grid map, if the node cost values ​​of multiple directions of the current adjacent node exceed the preset cost threshold d, it means that the adjacent node is very close to the expansion layer at this time, and its potential cost value is high, and it is moved to the overflow list and no subsequent traversal is performed, if it does not exceed the preset cost threshold d, it is put into the retention list for subsequent path planning;

[0115] In some embodiments, the constraints of the overflow list and the hold list in the open list are:

[0116]

[0117] Among them, P open Represented as an open list, P out Represented as an overflow list, P keep It represents the retention list, and d represents the preset cost threshold;

[0118] Step S4003: traverse and expand the generated path to increase the density of path points, calculate the absolute distance between two adjacent points by traversing the path points, and if the absolute distance is greater than a preset length threshold of 0.05m, perform bilinear interpolation optimization on the original path;

[0119] In some embodiments, the extended path points satisfy the following bilinear interpolation function formula:

[0120]

[0121] Among them, X p Indicates the x coordinate of the expansion path point, Y p Indicates the y coordinate of the expansion path point;

[0122] Step S4004, introduce an angle smoothing optimization function for constraint, first design a normalization function, and normalize the angle values ​​of the front and rear paths based on the normalization function to obtain the processed angle values, so that the angle range is 0 to 2π, and then the algorithm re-traverses the path to calculate the normalized angle between the current node and the adjacent node in the list, if the calculated angle exceeds the preset angle threshold π / 4, the adjacent nodes are optimized according to the median interpolation method when the preset cost constraint threshold is met;

[0123] In some embodiments, the normalization function formula is:

[0124]

[0125] Among them, θ NDH is the normalized angle value after conversion, and α is the angle value before conversion;

[0126] Step S4005: introduce a path optimization method based on JPS jump point search, extract the path list based on the JPS jump point search algorithm, take the initial node P0 as the parent node, if the newly generated path connected to the second node P2 after skipping the first node P1 does not collide with the obstacle, and the potential cost values ​​of the interpolated child nodes of the path are all less than the threshold, then cut the first node P1 to form a new path from the initial node P0 to the second node P2, and then continue to traverse the third node P3 until the path collides or the potential cost value is greater than or equal to the preset cost threshold, the path list is closed to form a new sub-path list, and the previous traversal process is repeated with the next node of the end node as the parent node, and the path satisfies the following constraints:

[0127]

[0128] Among them, d JPS is the Euclidean distance of the new path, d sum is the Euclidean distance of the path before clipping;

[0129] Step S4006: finally output a global planning path that is straightened and curved to smoothen, so as to determine a global optimal insect repellent operation path of the cherry tomato garden insect repellent robot.

[0130] It can be seen from the above embodiments that by adopting the above technical scheme, the global planning path constructed based on the improved Dijkstra global path planning algorithm has a better straightening effect between straight rows, and has the effect of reducing the number of corners and making the corners smoother when planning the path between curved rows, which is conducive to the cherry tomato garden patrol and insect repellent robot to perform navigation operations between rows in the plant factory.

[0131] Step S50: Use a preset local path planning algorithm to perform obstacle avoidance correction on the globally optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol insect repellent robot, and control the cherry tomato garden patrol insect repellent robot to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation, so as to complete the cherry tomato garden patrol insect repellent control.

[0132] After determining the global optimal insect repellent operation path of the cherry tomato garden patrol insect repellent robot, a preset local path planning algorithm is used to perform obstacle avoidance correction on the global optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol insect repellent robot, and the cherry tomato garden patrol insect repellent robot is controlled to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation, so as to complete the cherry tomato garden patrol insect repellent control, and the local path planning algorithm is the TEB local path planning algorithm.

[0133] Specifically, before using the preset TEB local path planning algorithm to perform obstacle avoidance correction on the globally optimal insect repellent operation path, an Ackerman structure cherry tomato garden insect repellent robot model is established, which has a constraint condition of a minimum turning radius R. The expression of the minimum turning radius R is as follows:

[0134]

[0135] The arc length O A O B is the component velocity V X The integral of t, the posture rotation angle θ is the component velocity V z The integral with respect to t;

[0136] Under the constraint condition of satisfying the minimum turning radius of the cherry tomato garden patrol insect repellent robot, the TEB local path planning algorithm is used to optimize the weighted minimum turning radius constraint objective function to finally form a local planning path.

[0137] The preset TEB local path planning algorithm is used to perform obstacle avoidance correction on the global optimal insect repellent operation path. The TEB (Timed Elastic Band) local path planning algorithm is particularly suitable for dynamic environments and high-speed mobile robots. It combines the time window strategy and elastic band optimization to achieve safe and efficient path planning; the TEB local path planning algorithm performs path planning within a time window within a certain period of time in the future, taking into account the dynamic characteristics of the robot and changes in the environment. This method can avoid the problem of static path planning failing in a dynamic environment; the TEB local path planning algorithm can describe the path through a series of discrete points, and these points can be dynamically adjusted according to the movement of the robot and environmental changes; the TEB local path planning algorithm optimizes the position and speed of these discrete points to ensure the smoothness and safety of the path; the TEB local path planning algorithm will take into account the local obstacle information around the cherry tomato garden insect repellent robot and avoid collisions and conflicts by adjusting the path.

[0138] In some embodiments, see Figure 8 , Figure 8 It is a kinematic analysis schematic diagram of a cherry tomato garden patrol insect repellent robot using an Ackerman structure in an embodiment of the present application, which satisfies the constraint of the minimum turning radius R. By constraining the minimum turning radius, a TEB local path planning algorithm is designed, which has better local path planning capabilities, and the output local path is more in line with the robot kinematics.

[0139] In some embodiments, see Figure 7The step of controlling the cherry tomato garden patrol and insect repellent robot to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation comprises:

[0140] Step S501, responding to the cherry tomato deworming operation instruction, determining a safe deworming operation path of the cherry tomato garden deworming robot, and reaching each deworming operation target point based on the safe deworming operation path;

[0141] Step S503, acquiring an image of a cherry tomato plant in a binocular camera at each of the target points of the insect repellent operation, and performing target detection on the image of the cherry tomato plant based on a preset tomato fruit detection model to determine a target detection frame corresponding to the cherry tomato;

[0142] Step S505: driving the cherry tomato patrol and insect repellent robot to perform ultrasonic insect repellent on the target tomato fruit according to the target detection frame according to a preset frequency threshold, so as to complete the patrol and insect repellent control of cherry tomatoes.

[0143] Specifically, the tomato fruit detection model is a model that takes the cherry tomato plant image as input and the ripe tomato label and its bounding box indicating the position of the ripe tomato image as output. The tomato fruit detection model is implemented as a preferred neural network model. For example, in an embodiment of the present application, the tomato fruit detection model is a Mask-RCNN that has been trained to convergence. Alternatively, the tomato fruit detection model can select a variety of relatively excellent target detection models in the prior art, including but not limited to: YOLO series models, other R-CNN series models, SSD models, DETR, etc., all of which are mature target detection models.

[0144] After the cherry tomato garden patrol and insect repellent robot reaches the insect repellent operation target point according to the safe insect repellent operation path determined by the above scheme, the cherry tomato garden patrol and insect repellent robot obtains the cherry tomato plant image based on the binocular camera in the cherry tomato garden patrol and insect repellent robot, obtains the cherry tomato plant image in the binocular camera at each insect repellent operation target point, and performs target detection on the cherry tomato plant image based on the tomato fruit detection model that has been trained to a convergent state to determine the target detection frame corresponding to the cherry tomato; the cherry tomato garden patrol and insect repellent robot is driven to perform ultrasonic insect repellent on the target tomato fruit according to the target detection frame according to a preset frequency threshold to complete the garden patrol and insect repellent control of cherry tomatoes.

[0145] It can be seen from the above embodiments that, compared with the prior art, the present application aims at the problems that the environment map constructed by the cherry tomato garden patrol and insect repellent robot in the prior art has many point cloud noise points, the planned path has poor applicability between the rows of the plant factory where the cherry tomato garden patrol and insect repellent robot operates, and the efficiency and safety of the robot operation are seriously reduced. The present application includes but is not limited to the following beneficial effects:

[0146] First, the method for controlling pest control in a cherry tomato plant factory between rows is described in the present application. A 3D laser radar is combined with a point cloud filtering algorithm to finally construct a grid map required for navigation. The method has the advantages of stronger environmental applicability and richer perception information. The adaptive Monte Carlo positioning scheme that integrates robot posture and mileage information for multi-sensor fusion has the advantage of stronger robustness. The traversal path smoothing interpolation optimization method and the path optimization method based on JPS jump point search are introduced to improve the Dijkstra global path planning algorithm. The global path planned by the algorithm has the characteristics of being smoother and straighter. The local path planning is performed in combination with the TEB local path planning algorithm that meets the minimum turning radius constraint condition of the cherry tomato patrol pest control robot, thereby optimizing the effect of the planned path.

[0147] Secondly, the present application is suitable for driving navigation of a cherry tomato garden patrol and insect repellent robot in the inter-row operation scenario of a plant factory. The overall navigation method is designed using a 3D laser radar, and the algorithm is improved in environmental adaptability in map construction and path planning. The traditional navigation method has poor mapping and path planning effects in the inter-row environment of a plant factory, and finally provides a set of efficient and safe navigation methods suitable for cherry tomato garden patrol and insect repellent robot inter-row operations in plant factories;

[0148] Third, the Cartographer algorithm of the present application locates the position of the cherry tomato patrol and insect repellent robot while constructing the point cloud map of the inter-row environment of the plant factory. By performing a back-end loop test on the sub-map of the inter-row environment of the plant factory, the cumulative error of the algorithm can be significantly reduced. When constructing the point cloud map of the inter-row environment of the plant factory, it does not rely too much on external equipment, and a relatively accurate map can be run through a laser radar with lower precision, which greatly reduces the cost of the robot;

[0149] Fourthly, the adaptive Monte Carlo positioning algorithm of the present application can dynamically adjust the number and distribution of particles to adapt to different environmental conditions and accuracy requirements. In complex environments such as between rows in plant factories or when higher positioning accuracy is required, increasing the number of particles can improve the accuracy of positioning, and adjusting the resampling strategy and sensor model according to robot movement and environmental changes to improve the robustness and efficiency of positioning;

[0150] Fifth, the improved Di jkstra global path planning algorithm of the present application introduces a traversal path smoothing interpolation optimization method to smooth and optimize the curved path to reduce inflection points and turning angles, and introduces a path optimization method based on JPS jump point search to perform node pruning on straight sections to straighten the path, so that the global path planned by the cherry tomato garden patrol and insect repellent robot is more suitable for working between rows in plant factories.

[0151] See also Fig. 9 , a cherry tomato plant factory inter-row patrol and insect repellent control device provided to meet one of the purposes of the present application, including. Among them, the data acquisition module 1100 is configured to receive the cherry tomato patrol and insect repellent task, and obtain the laser radar point cloud data from the laser radar; the grid map determination module 1200 is configured to use a preset point cloud map construction algorithm to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, perform a back-end loop test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and convert the point cloud map into a grid map for navigation of the cherry tomato patrol and insect repellent robot; the positioning result determination module 1300 is configured to determine the initial posture and odometer data corresponding to the cherry tomato patrol and insect repellent robot, and use an adaptive Monte Carlo positioning algorithm to track the cherry tomato patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map. The position and posture of the cherry tomato garden patrol and insect repellent robot are used to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot; the global path determination module 1400 is configured to determine each target insect repellent operation point, and based on the preset global path planning algorithm, according to the grid map, the real-time positioning result of the cherry tomato garden patrol and insect repellent robot and the target insect repellent operation points, determine the global optimal insect repellent operation path of the cherry tomato garden patrol and insect repellent robot; the insect repellent operation control module 1500 is configured to use a preset local path planning algorithm to perform obstacle avoidance correction on the global optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol and insect repellent robot, and control the cherry tomato garden patrol and insect repellent robot to reach the various insect repellent operation target points according to the safe insect repellent operation path to perform insect repellent operations, so as to complete the garden patrol and insect repellent control of cherry tomatoes.

[0152] Based on any embodiment of this application, please refer to Fig.10 Another embodiment of the present application further provides an electronic device, which can be implemented by a computer device, such as Fig.10As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement a method for controlling insects in the inter-row patrol of a cherry tomato plant factory. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the method for controlling insects in the inter-row patrol of a cherry tomato plant factory of the present application. The network interface of the computer device is used to connect and communicate with a terminal. Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0153] In this embodiment, the processor is used to execute Fig. 9 The memory stores the program code and various data required to execute the above modules or submodules. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program code and data required to execute all modules / submodules in the inter-row patrol and insect repellent control device of the cherry tomato plant factory of this application, and the server can call the program code and data of the server to execute the functions of all submodules.

[0154] The present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the inter-row patrol and insect control method for a cherry tomato plant factory described in any embodiment of the present application.

[0155] The present application also provides a computer program product, including a computer program / instruction, which, when executed by one or more processors, implements the steps of the inter-row patrol and insect control method for a cherry tomato plant factory described in any embodiment of the present application.

[0156] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the present application can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a computer-readable storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0157] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

[0158] In summary, the 3D laser radar combined with the point cloud filtering algorithm is used to finally construct the grid map required for navigation, which has the advantages of stronger environmental applicability and richer perception information. The adaptive Monte Carlo positioning scheme that integrates the robot posture and mileage information for multi-sensor fusion has the advantage of stronger robustness. The traversal path smoothing interpolation optimization method and the path optimization method based on JPS jump point search are introduced to improve the Di jkstra global path planning algorithm. The planned global path has the characteristics of smoother and straighter path. The local path planning is carried out in combination with the TEB local path planning algorithm that meets the minimum turning radius constraint of the cherry tomato patrol insect repellent robot, which optimizes the effect of the planned path.

Claims

1. A method for controlling insects in cherry tomato plant rows, characterized in that: include: Receive the task of patrolling the cherry tomato garden and removing insects, and obtain the lidar point cloud data from the lidar; A preset point cloud map construction algorithm is used to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, a back-end loop test is performed on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and the point cloud map is converted into a grid map for navigation of the cherry tomato garden patrol insect repellent robot; Determine the initial posture and odometer data corresponding to the cherry tomato garden patrol and insect repellent robot, and use an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot; Determine each target insect repellent operation point, and determine a global optimal insect repellent operation path of the cherry tomato garden insect repellent robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden insect repellent robot, and each target insect repellent operation point; A preset local path planning algorithm is used to perform obstacle avoidance correction on the globally optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol insect repellent robot, and the cherry tomato garden patrol insect repellent robot is controlled to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation, so as to complete the garden patrol insect repellent control of cherry tomatoes.

2. The method for controlling insects in the cherry tomato plant factory rows according to claim 1, characterized in that: The steps of using a preset point cloud map construction algorithm to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data, and performing a back-end loopback test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory include: In response to the point cloud map construction instruction, the laser radar point cloud data is input into the Cartographer algorithm, and the local mapping module in the Cartographer algorithm generates a sub-map of the inter-row environment of the plant factory, and the point cloud map construction algorithm is the Cartographer algorithm; The global mapping module in the Cartographer algorithm performs a back-end loop test on the sub-map of the inter-row environment of the plant factory, and corrects the map by detecting the error between the sub-maps at adjacent moments to determine the sub-map after the loop test; The sub-map after the loop test is subjected to posture optimization to generate a point cloud map of the inter-row environment of the plant factory.

3. The method for controlling insects in the cherry tomato plant rows according to claim 1, characterized in that: The steps to determine the point cloud map of the plant factory inter-row environment include: A preset straight-through filtering algorithm is used to post-process the point cloud map of the inter-row environment of the plant factory, and based on the straight-through filtering algorithm and a preset height threshold, the ground point cloud noise points and the invalid point clouds above the motion range of the cherry tomato garden patrol insect repellent robot are filtered and optimized; Determine the filter radius and the number of filtered neighboring point clouds, and set the Euclidean distance radius range for each point based on the preset radius filtering algorithm according to the filter radius and the number of filtered neighboring point clouds. If the number of point clouds within the range is less than the number of filtered neighboring point clouds, the point is filtered out as an outlier to determine the point cloud map of the inter-row environment of the plant factory.

4. The method for controlling insects in the cherry tomato plant factory rows according to claim 1, characterized in that: The steps of using an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine the real-time positioning result of the cherry tomato garden patrol and insect repellent robot include: In response to a self-positioning instruction of the cherry tomato garden patrol and insect repellent robot, a preset adaptive Monte Carlo positioning algorithm is used to generate particles obeying a Gaussian distribution near an initial position of the cherry tomato garden patrol and insect repellent robot, wherein the particles represent a possible position distribution of the cherry tomato garden patrol and insect repellent robot; The corresponding position and posture of each particle is updated according to the odometer data of the cherry tomato garden patrol and insect repellent robot to determine the updated particle position and posture, the observation model of each particle is obtained by using the laser radar point cloud data and the likelihood domain model of the grid map, and the particle is assigned a corresponding position and posture weight; The pose weight of each particle is corrected according to the measurement probability of the laser radar point cloud data, so as to obtain the particle pose and pose weight that are continuously iteratively updated during the movement of the cherry tomato garden patrol insect repellent robot; According to the pose weights of the particles, the current pose corresponding to the cherry tomato garden patrol and insect repellent robot is estimated, and based on the current pose corresponding to the cherry tomato garden patrol and insect repellent robot, the laser radar point cloud data is converted into a point cloud data set in the same coordinate system as the grid map; Each frame of the point cloud data set is input into a preset iterative closest point algorithm to determine the rotation matrix and translation matrix of the point cloud data set of the current frame relative to the grid map, and the real-time positioning result of the cherry tomato garden patrol and insect repellent robot in the grid map is determined based on the rotation matrix and the translation matrix.

5. The method for controlling insects in the cherry tomato plant rows according to any one of claim 1, characterized in that: The global path planning algorithm is an improved Dijkstra global path planning algorithm; the local path planning algorithm is a TEB local path planning algorithm.

6. The method for controlling insects in the cherry tomato plant rows according to claim 5, characterized in that: The step of determining each target insect repellent operation point, and determining a global optimal insect repellent operation path of the cherry tomato garden insect repellent robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden insect repellent robot and each target insect repellent operation point comprises: Construct an open list and a closed list to store the path points to be determined and the determined path points respectively. Take the starting point as the origin to traverse and calculate the Euclidean distance and Manhattan distance to the nearby nodes. When the shortest path is found, put the node into the corresponding path list, and traverse and calculate again with the shortest path point as the center until the shortest path to the target point is found. This path is the global shortest path. Determine a preset cost threshold of the grid map, introduce potential cost values ​​into the open list to perform secondary traversal of nodes, perform secondary cost value traversal of multiple directions adjacent to the current non-obstruction node, if the node cost values ​​of multiple directions of the current adjacent node exceed the preset cost threshold, move it to the overflow list and no longer perform subsequent traversal, if it does not exceed the preset cost threshold, put it into the retention list for subsequent path planning; The generated path is traversed and expanded to increase the density of path points. The absolute distance between two adjacent points is calculated by traversing the path points. If the absolute distance is greater than a preset length threshold, bilinear interpolation optimization is performed on the original path. Based on the normalization function, the angle values ​​of the front and rear paths are normalized to obtain the processed angle values, so that the angle range is 0 to 2π, and the path is re-traversed to calculate the normalized angle between the current node and the adjacent node in the list. If the calculated angle exceeds the preset angle threshold, the adjacent node is optimized according to the median interpolation method; Extract the path list based on the preset JPS jump point search algorithm, take the initial node as the parent node, if the newly generated path connected to the second node after skipping the first node does not collide with the obstacle, and the potential cost value of the interpolated child nodes of the path is less than the preset cost threshold, then cut the first node to form a new path from the initial node to the second node, continue to traverse the third node until the path collides or the potential cost value is greater than or equal to the preset cost threshold, the path list is closed to form a new sub-path list, and the above steps are repeated with the next node of the end node as the parent node; The global planning path after straightening and smoothing of curves is output to determine the global optimal insect repellent operation path of the cherry tomato garden insect repellent robot.

7. The method for controlling insects in the cherry tomato plant factory rows according to claims 1 to 6, characterized in that: The step of controlling the cherry tomato garden patrol and insect repellent robot to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operation comprises: In response to the cherry tomato pest control operation instruction, determine a safe pest control operation path for the cherry tomato garden pest control robot, and reach each pest control operation target point based on the safe pest control operation path; Acquire an image of a cherry tomato plant in a binocular camera at each of the target points of the insect repellent operation, and perform target detection on the image of the cherry tomato plant based on a preset tomato fruit detection model to determine a target detection frame corresponding to the cherry tomato; The cherry tomato garden patrol and insect repellent robot is driven to perform ultrasonic insect repellent on the target tomato fruit according to the target detection frame according to a preset frequency threshold, so as to complete the garden patrol and insect repellent control of the cherry tomatoes.

8. A device for controlling insect repellent in rows of cherry tomato plant factories, characterized in that: include: A data acquisition module is configured to receive a cherry tomato garden patrol and insect repellent task and obtain laser radar point cloud data from the laser radar; A grid map determination module is configured to generate a sub-map of the inter-row environment of the plant factory according to the laser radar point cloud data using a preset point cloud map construction algorithm, perform a back-end loop test on the sub-map to determine the point cloud map of the inter-row environment of the plant factory, and convert the point cloud map into a grid map for navigation of the cherry tomato garden patrol insect repellent robot; A positioning result determination module is configured to determine an initial posture and odometer data corresponding to the cherry tomato garden patrol and insect repellent robot, and use an adaptive Monte Carlo positioning algorithm to track the posture of the cherry tomato garden patrol and insect repellent robot in real time according to the initial posture, the odometer data and the grid map to determine a real-time positioning result of the cherry tomato garden patrol and insect repellent robot; A global path determination module is configured to determine each target insect repellent operation point, and determine a global optimal insect repellent operation path of the cherry tomato garden insect repellent robot based on a preset global path planning algorithm according to the grid map, the real-time positioning result of the cherry tomato garden insect repellent robot and each target insect repellent operation point; The insect repellent operation control module is configured to use a preset local path planning algorithm to perform obstacle avoidance correction on the globally optimal insect repellent operation path to determine the safe insect repellent operation path of the cherry tomato garden patrol insect repellent robot, and control the cherry tomato garden patrol insect repellent robot to reach each insect repellent operation target point according to the safe insect repellent operation path to perform insect repellent operations, so as to complete the cherry tomato garden patrol insect repellent control.

9. An electronic device, comprising a central processing unit and a memory, characterized in that: The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 7 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

Citation Information

Patent Citations

  • Service robot navigation method and system

    CN108375976A

  • JPS-improved unmanned vehicle path planning method suitable for complex environment

    CN117075607A

  • Indoor article transportation robot and control method thereof

    CN117234208A

  • Robot inspection path planning method and system based on cloud platform

    CN117782097A

  • Laser forklift positioning method

    CN118089755A

Cited By

  • Automatic row control algorithm for agricultural machinery based on laser radar perception

    CN120252712A