Multi-mode laser group intelligent weeding robot and weeding control method

Through the intelligent weeding robot of multi-mode laser group, the machine vision and Beidou positioning system are used to accurately identify and locate weeds, which solves the problem that laser weeding is difficult to accurately illuminate, improves the efficiency and safety of weeding, reduces the use of chemical herbicides, and realizes the realization of It has efficient and environmentally friendly weeding effect.

CN119073295APending Publication Date: 2024-12-06SHANDONG AGRICULTURAL UNIVERSITY

Patent Information

Application Number
CN202411273033.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-12
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Existing herbicide robots find it difficult to accurately illuminate the growth points of weeds during laser weeding, and chemical weeding methods pose a threat to crops and the environment, mechanical weeding efficiency is low, affecting crop yield and environmental safety.

Method used

A multi-mode laser group intelligent weeding robot is designed, equipped with a machine vision system and a swing steer. Through the depth camera and Beidou differential positioning system, weeds are identified and positioned in real time, and the laser output power and angle are adjusted to ensure that the laser accurately illuminates weeds. growth point.

Benefits of technology

It improves the success rate and efficiency of weeding, reduces the impact on crops and the environment, reduces the use of chemical herbicides, and achieves a more efficient and safer weeding effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119073295A_ABST
    Figure CN119073295A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-mode laser group intelligent weeding robot and a weeding control method, and relates to the technical field of intelligent agricultural equipment.The multi-mode laser group intelligent weeding robot comprises a multi-mode laser group weeding device and a machine vision system which are arranged on a mobile platform, and the machine vision system recognizes and classifies weeds; the multi-mode laser group weeding device is provided with a linear module and a swing steering engine, an output laser is fixedly arranged at the tail end of the swing steering engine, positioning of a determined growing point of the age of weeds is achieved through a machine vision system, and according to the age of target weeds needing to be cleared and the posture of the growing point of the weeds, the target weeds are cleared. The linear module in the multi-mode laser group weeding device controls the output laser device with the corresponding output power to move to the target weed area, the swing steering engine controls the output laser device to be aligned with the growing point of the target weed to conduct accurate fixed-point laser burning, and the weeding success rate is increased.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of intelligent agricultural equipment, and in particular to a multi-mode laser group intelligent weeding robot and a weeding control method. Background Art

[0002] In the early stages of crop growth, crops generally have lower adaptability to the environment than weeds, and often find it difficult to obtain the nutrients needed for growth and development in the competition with weeds, causing poor crop development or even death. At the same time, the presence of weeds provides an environment for pests and pathogens to survive and reproduce, increases the population density of pests and pathogens, and greatly increases the possibility of pests and diseases. In summary, weeds are very detrimental to the growth and development of crops, seriously affecting crop yields, causing huge economic losses, and becoming one of the important factors restricting agricultural production.

[0003] Limited by the low level of intelligent agricultural production and the increasing labor costs, the currently widely used weeding method is chemical weeding, that is, field weeding by manually or mechanically spreading chemical agents. This weeding method is difficult to avoid the adverse effects of herbicides on crops, and is prone to drug residues, posing a serious threat to food safety. Herbicides remaining in the soil and water sources will also affect the local ecosystem and cause environmental pollution, which is particularly unfavorable for farmland with crop rotation, seriously restricting the high-quality development of agriculture. At present, the weeding method of weeding robots mainly adopts mechanical methods, that is, various mechanical structures of different configurations are used to destroy weed tissues to remove them, which can be mainly divided into two categories: pulling and cutting. This type of weeding machine is generally complex in structure and has low weeding efficiency. Chinese patent CN113632779A provides a laser intelligent weeding robot and a weeding method, which uses laser weeding to avoid the impact on crops and the surrounding soil environment, and improves weeding efficiency.

[0004] Although the laser intelligent weeding robot in the above patent can achieve weeding by laser, the laser emitter uses a fixed point and emits laser downward perpendicular to the horizontal plane when weeding. However, the key to using laser weeding is to use laser to irradiate its growth point. There are large differences in the position and posture of the growth point of various weeds. The stems, leaves and growth points of some weeds will tilt toward the light side due to phototropism, making it difficult for the laser emitted vertically downward to accurately irradiate the growth point of the weed. In addition, due to the obstruction of the leaves of crops and other weeds, it is difficult to ensure the actual treatment effect by simply using a fixed point and fixed direction to emit laser. Summary of the invention

[0005] In order to solve the above technical problems, this application proposes the following technical solutions: In a first aspect, an embodiment of the present application provides a multi-mode laser group intelligent weeding robot, comprising: a mobile platform, a multi-mode laser group weeding device and a machine vision system arranged on the mobile platform, the machine vision system being used to identify and classify weeds, the multi-mode laser group weeding device being provided with a swinging servo, an output laser being fixedly arranged at the end of the swinging servo, the output lasers with different output powers being controlled to operate according to the variety and age of the weeds in the recognition results of the machine vision system, and the swinging servo being controlled so that the output laser is aimed at the growth point of the weeds.

[0006] In a possible implementation, the swing servos include two groups, the laser output powers of the output lasers arranged on different swing servos are different, the swing servos are slidably connected to the vertical linear module through a vertical connector, and the vertical linear module is slidably connected to the horizontal linear module through a horizontal connector.

[0007] In a possible implementation, the machine vision system includes a gimbal disposed at the bottom of the forward end of the mobile platform and a depth camera disposed on the gimbal, and the depth camera acquires weed image information to determine weed color and weed coverage area.

[0008] In a possible implementation, it also includes a field environment perception system, which includes an inter-row navigation camera set on the mobile end of the mobile platform and a Beidou differential positioning system satellite antenna set on the top of the mobile platform. The Beidou differential positioning system communicates with satellite signals to perform real-time positioning of the weeding robot and transmits coordinate information to the control system, which compares and analyzes the collected coordinate information with the initial reference map; the inter-row navigation camera collects image information in front of the travel direction, and identifies and detects it to ensure that the weeding robot does not deviate from the row; the control system finally makes a decision through data fusion and issues corresponding steering instructions to the mobile platform.

[0009] In a second aspect, an embodiment of the present application provides a weed control method, using the multi-mode laser group intelligent weeding robot described in any possible implementation of the first aspect, the method comprising: The operating position of the multi-mode laser group intelligent weeding robot is determined through the Beidou differential positioning system and the inter-row navigation camera; When the machine vision system’s depth camera identifies the weeds that need to be removed, it determines the weed’s species, age, and growth point; After the variety, age and growth point of the weeds are determined, the multi-mode laser group weeding device is controlled to perform weeding.

[0010] In a possible implementation, determining the operating position of the multi-mode laser group intelligent weeding robot by using the Beidou differential positioning system and the inter-row navigation camera includes: The Beidou differential positioning system locates the weeding robot in real time and transmits coordinate information to the control system, which compares and analyzes the collected coordinate information with the initial reference map; The row navigation camera collects image information in the direction of travel and identifies and detects it to ensure that the weeding robot does not deviate from the row; The processing results of the Beidou differential positioning system and the inter-line navigation camera are transmitted to the control system in real time. The control system performs data fusion on the received processing results and finally makes a decision and issues corresponding instructions to the mobile platform.

[0011] In a possible implementation, after the depth camera of the machine vision system determines the weeds to be removed, the variety, age and growth point of the weeds are determined, including: The depth camera acquires a depth image of the weeds in the working area; determining the variety, age and growth point of the weeds according to the texture features of the weeds in the RGB image; The growth point of the weeds is determined, and the growth point of the weeds is used as the weeding point of the output laser.

[0012] In a possible implementation, after determining the variety, age and growth point of the weeds, controlling the multi-mode laser group weeding device to weed, includes: Determining the laser to be operated according to the variety and age of the weeds, wherein the laser includes a low-power laser, a high-power laser, or a combination of the low-power laser and the high-power laser; Controlling the vertical linear module and the horizontal linear module corresponding to the output laser to move the output laser to the weed area to control the distance between the focusing lens and the weeds; The angle of the output laser is adjusted by swinging the steering gear, so that the output laser is aimed at the growth point of the weeds for laser irradiation.

[0013] In a possible implementation, the step of determining the output laser that needs to be operated according to the age of the weeds includes: If it is determined according to the variety and age of the weeds that the weeds have not yet matured, the tissues show a physiological state of light green color, and there are few leaves, then it is determined that the output laser is a low-power laser; or, If the weeds are determined to be mature according to their species and age, and their tissues show a physiological state of being firm, dark in color, and having many leaves, it is determined that the output laser is a high-power laser or a combination of a low-power laser and a high-power laser.

[0014] In the embodiment of the present application, the weed variety, age and growth point are determined by a machine vision system. According to the weed variety and age of the target weeds to be removed, the mobile module in the multi-mode laser group weeding device controls the output laser of the corresponding output power to move to the target weed area, and the swing servo controls the output laser to aim at the growth point of the target weeds for precise and fixed-point laser burning, thereby improving the success rate of weeding. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the structure of a multi-mode laser group intelligent weeding robot provided in an embodiment of the present application; Figure 2 A bottom schematic diagram of a multi-mode laser group intelligent weeding robot provided in an embodiment of the present application; Figure 3 A side view of a multi-mode laser group intelligent weeding robot provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of a multi-mode laser group weeding device provided in an embodiment of the present application; Figure 5 A schematic diagram of a weed control method provided in an embodiment of the present application; Figure 6 A schematic diagram of laser output control during weeding operations provided in an embodiment of the present application; Figure 7 A schematic diagram of the swing range of the output laser provided in an embodiment of the present application; Figure 1-Figure 7 In the symbol, it is represented as: 1-mobile platform, 2-multi-mode laser group weeding device, 3-machine vision system, 4-swing servo, 5-output laser, 6-Beidou differential positioning system satellite antenna, 7-inter-row navigation camera, 8-lidar, 9-vertical connector, 10-vertical linear module, 11-horizontal connector, 12-horizontal linear module, 13-end connector, 14-pan-tilt head, 15-depth camera, 16-human-computer interaction system. DETAILED DESCRIPTION

[0016] The present solution is described below in conjunction with the accompanying drawings and specific implementation methods.

[0017] See also Figure 1-Figure 4The multi-mode laser group intelligent weeding robot provided in the embodiment of the present application includes: a mobile platform 1, a multi-mode laser group weeding device 2 and a machine vision system 3 arranged on the mobile platform 1, the machine vision system 3 is used to identify and classify weeds, the multi-mode laser group weeding device 2 is provided with a swing servo 4, an output laser 5 is fixedly arranged at the end of the swing servo 4, the output laser 5 with different output powers is controlled to operate according to the variety and age of the weeds in the recognition result of the machine vision system 3, and the swing servo 4 is controlled so that the output laser 5 is aimed at the growth point of the weeds.

[0018] The mobile platform 1 drives the multi-mode laser group weeding device 2 and the machine vision system 3 to move along the extension direction of the row ridge. The two rear wheels of the mobile platform 1 are driven by a DC servo motor to provide driving force for the mobile platform 1; the two front wheels are not equipped with a DC servo motor and cannot provide driving force, but an Ackerman steering mechanism is used. When the direction of travel needs to be adjusted, the inner and outer wheels are controlled at different steering angles, so that the inner and outer wheels can maintain a small steering radius when turning, thereby realizing turning and changing rows.

[0019] The top of the chassis of the mobile platform 1 is equipped with a Beidou differential positioning system satellite antenna 6, which is conducive to better communication with satellite signals. The weeding robot is positioned in real time through the Beidou differential positioning system, and the coordinate information is transmitted to the control system, which compares and analyzes the collected coordinate information with the initial reference map. The inter-row navigation camera 7 installed on the front side of the weeding robot collects image information in the direction of travel, and identifies and detects it to ensure that the weeding robot does not deviate from the row ridge. The processing results of the two are transmitted to the control system in real time, and the control system fuses the processing results of the two and finally makes a decision, and issues a corresponding steering instruction to the mobile platform 1. At the same time, the laser radar 8 installed on the front side of the weeding robot detects the surrounding environment of the weeding robot in real time, and transmits the collected point cloud data to the control system. After analysis and processing, the control system makes corresponding obstacle avoidance decisions and controls the weeding robot to stop or turn to avoid collision.

[0020] When using the Beidou differential positioning system for the first time in a certain area, the robot is first manually remotely controlled to traverse its working route several times and take the average to obtain the accurate coordinates of each point on the route and establish an initial reference map. Subsequently, while the Beidou differential positioning system locates the weeding robot in real time and compares and analyzes it with the initial reference map, the inter-row navigation camera 7 mounted on the front side of the robot collects image information on the front side of the route in real time, and identifies and compares it with the information collected at the previous moment. The control system fuses the analysis results of the two through the sensor fusion algorithm and makes a final decision to control the movement of the weeding robot. In addition, a laser radar 8 is used to detect the surrounding environment of the weeding robot in real time, and the collected point cloud data is transmitted to the control system in real time. The control system analyzes the point cloud data to determine whether there are obstacles near its working area, avoid collisions of the weeding robot, and ensure production safety and equipment reliability.

[0021] The swing servo 4 includes two groups, and the laser output power of the output laser 5 set on different swing servos 4 is different. The swing servo 4 is slidably connected to the vertical linear module 10 through the vertical connector 9, and the vertical linear module 10 is slidably connected to the horizontal linear module 12 through the horizontal connector 11. Referring to Figure 3, the horizontal linear module 12 is fixed to the upper side of the profile frame of the mobile platform 1, and its slider can make linear motion along the horizontal linear module 12. The horizontal connector 11 is fixed to the slider of the horizontal linear module 12. The vertical linear module 10 is fixed to the horizontal connector 11, so that the horizontal linear module 12 can drive the vertical linear module 10 to move, and the slider of the vertical linear module 10 can move along the vertical linear module 10, and the vertical connector 9 is fixed to the slider of the vertical linear module 10. The swing servo 4 is fixed to the slider of the vertical linear module 10, so that the vertical linear module 10 can drive the swing servo 4 to move along the vertical linear module 10. The end connector 13 is fixedly connected to the output shaft of the swing servo 4, and can be used to fix the output laser 5, so that the swing servo 4 can drive the output laser 5 to swing. The vertical linear module 10 is used to control the distance between the laser output head and the weeds. The horizontal linear module 12 drives the vertical linear module 10 and its connecting structure to move toward the target weeds. The swing servo 4 then swings a certain angle toward the side of the target weeds so that the output laser 5 is aligned with the growth point of the weeds. Finally, the vertical linear module 10 controls the distance between the output laser 5 and the target weeds to control the spot size and ensure sufficient energy density.

[0022] The machine vision system 3 includes a pan-tilt 14 arranged at the bottom of the forward end of the mobile platform 1 and a depth camera 15 arranged on the pan-tilt 14. The depth camera 15 obtains the RGB image information of the weeds to determine the variety, age and growth point of the weeds. The pan-tilt 14 (camera attitude stabilizer) installed at the bottom of the mobile platform 1 can ensure that the depth camera 15 always collects images between rows at a fixed perspective, and the perspective of the depth camera 158 will not change due to the bumpy terrain, resulting in the image collected in the bumpy area and the image collected in the stable area. The scale is inconsistent, avoiding the machine vision system 3 from transforming the image. During the inter-row operation in the field environment, the robot will have a relatively strong bump, resulting in the image collected by the depth camera There are ghosting and distortion, affecting the identification and positioning of the weeds, so the pan-tilt is used to mount the depth camera to ensure that the depth camera can ensure that the perspective is stable and unchanged during the inter-row operation while improving the quality of the collected image.

[0023] The collected image information is identified and processed by the YOLOv8-Pose detection model and classified. The coordinate conversion algorithm is then used to convert the pixel coordinates of each weed in the image into its world coordinates in the real world, so as to achieve weed grading (variety, age) and precise positioning. When the mobile platform 1 moves forward, the multi-mode laser group intelligent weeding device moves as a whole. According to the positioning information of each weed, the slider of the horizontal linear module 12 drives the vertical linear module 10, the swing motor and the output laser 5 to move to the corresponding position, so that the output laser 5 is vertically downwardly aimed at the weeds to be treated, and the vertical linear module 10 moves vertically so that the focusing lens of the output laser 5 maintains a certain distance from the key growth point of the weed to ensure the spot size of the laser irradiated at the key growth point, and ensure that there is enough energy to be absorbed by the weed tissue, so that the cell tissue is dehydrated and denatured, thereby inhibiting its growth and development.

[0024] In this embodiment, a fiber laser is used as the light source of the laser. The fiber laser uses doped fiber as a gain medium, forms initial photons through excitation energy levels, and amplifies the stimulated radiation inside the fiber and the back-and-forth reflection of light, and finally forms a high-intensity laser output in the optical resonant cavity. The laser emitted from the laser generator is transmitted to the laser via the laser transmission fiber, which has high stability and consistency, but at this time the laser beam diameter is large and the energy is still relatively dispersed, so a focusing lens is needed to further focus the laser to increase its energy density.

[0025] The multi-mode laser group intelligent weeding robot in this embodiment also includes a human-computer interaction system 16. The human-computer interaction system 16 is electrically connected to the control system of the multi-mode laser group intelligent weeding robot, and can view the equipment status of the weeding robot, such as power information, laser power, moving speed of the linear module, etc. The human-computer interaction system 16 can also be used to simply debug the system and simply configure the control system to facilitate debugging and maintenance by operators. This weeding robot has built-in Bluetooth and WIFI communication functions. Relying on the Bluetooth communication protocol, it can be operated beside the robot. According to different scenes and tasks, its operating mode can be flexibly selected, which is convenient for operators to intervene and control it at any time; based on the built-in WIFI module, a cloud platform can be built in the later stage to remotely control the weeding robot online, further enhance its intelligence, and improve production efficiency.

[0026] Corresponding to the multi-mode laser group intelligent weeding robot provided in the above embodiment, the present application also provides a weeding control method using the multi-mode laser group intelligent weeding robot in the above embodiment.

[0027] See also Figure 5 The weed control method in this embodiment comprises: S101, determines the operating position of the multi-mode laser group intelligent weeding robot through the Beidou differential positioning system and the inter-row navigation camera.

[0028] The weeding robot is positioned in real time through the Beidou differential positioning system, and the coordinate information is transmitted to the control system, which compares and analyzes the collected coordinate information with the initial reference map; the inter-row navigation camera installed on the front side of the weeding robot collects image information in the direction of travel, and identifies and detects it to ensure that the weeding robot does not deviate from the row ridge; the processing results of the two are transmitted to the control system in real time, and the control system fuses the data of the two processing results and finally makes a decision, issuing corresponding steering instructions to the mobile platform.

[0029] S102, when the depth camera of the machine vision system determines the weeds that need to be removed, the weed age and weeding point of the weeds are determined.

[0030] The depth camera acquires the depth image of the weeds in the working area, and the collected depth image information is identified and processed by the YOLOv8-Pose detection model and classified. The variety and age of the weeds are determined according to the texture characteristics or morphological structure of the weeds in the depth image. For example, a certain weed is not yet mature, and the tissue shows a physiological state of light green color and few leaves. However, if the weeds tend to mature, the tissue shows a physiological state of tight texture and dark color, and many leaves. Finally, the growth point of the weed is determined, and the growth point of the weed is used as the weeding point of the output laser.

[0031] S103, after determining the variety, age and growth point of the weeds, controlling the multi-mode laser group weeding device to perform weeding.

[0032] The laser or laser combination that needs to participate in the operation is determined according to the variety and age of the weeds, and the output laser includes a low-power laser and a high-power laser. The laser combination is a combination of a low-power laser and a high-power laser. If it is determined according to the variety and age of the weeds that the weeds are not yet mature, the tissues show a physiological state of light green color, and there are few leaves, then the output laser is determined to be a high-power laser. The high-power laser has a small power and limited energy density. In this embodiment, the high-power laser has a wavelength of 450nm and a power of 60w.

[0033] If the weeds are determined to be mature according to their species and age, and their tissues show a physiological state of being firm and dark in color, with many leaves, then the output laser is determined to be a combination of a low-power laser and a high-power laser, and greater energy is used to destroy their cell structure to achieve a weeding effect. Low-power lasers have larger wavelengths and smaller powers. In this embodiment, the low-power laser has a wavelength of 1064nm and a power of 50w.

[0034] The machine vision system will identify the species and age of weeds based on their characteristics and select a weed control strategy: for older weeds, some species of weeds show more leaves and larger leaf sizes, while some species only show an increase in leaves; for plants with more leaves, higher energy is often required, so two lasers are used to treat them one after another. For younger weeds, only high-power lasers are used to treat them to achieve the weed control effect, while for older weeds, low-power lasers are used for secondary treatment after high-power laser treatment to increase the treatment dose of the weeds and ensure the weed control effect.

[0035] Different weeding strategies are adopted according to the weed species composition and actual physiological conditions in the actual scene. This not only ensures the efficiency of laser weeding, but also saves energy and improves the endurance of the equipment.

[0036] See also Figure 6 and Figure 7 After determining the laser to be used, the vertical linear module and the horizontal linear module corresponding to the output laser are controlled to move the output laser to the weed area to control the distance between the focusing lens and the weeds. The angle of the output laser is adjusted by swinging the servo so that the output laser is aimed at the growth point of the weeds for laser irradiation.

[0037] Specifically, the vertical linear module is used to control the distance between the laser output head and the weeds. The horizontal linear module drives the vertical linear module and its connecting structure to move toward the target weeds. The swing servo then swings toward the target weeds at a certain angle to align with the growth point of the weeds. Finally, the vertical linear module controls the distance between the output laser and the target weeds to control the spot size and ensure sufficient energy density.

[0038] In the embodiments of the present application, "at least one" refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and subsequent associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0039] The above is only a specific implementation of the present application. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. The protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multi-mode laser group intelligent weeding robot, It is characterized in that include: A mobile platform, a multi-mode laser group weeding device and a machine vision system arranged on the mobile platform, the machine vision system is used to identify and classify weeds, the multi-mode laser group weeding device is provided with a swinging servo, an output laser is fixedly arranged at the end of the swinging servo, the output laser with different output powers is controlled to operate according to the variety and age of the weeds in the recognition result of the machine vision system, and the swinging servo is controlled so that the output laser is aimed at the growth point of the weeds.

2. The multi-mode laser group intelligent weeding robot according to claim 1, It is characterized in that The swing servos include two groups, and the laser output powers of the output lasers arranged on different swing servos are different. The swing servos are slidably connected to the vertical linear module via a vertical connector, and the vertical linear module is slidably connected to the horizontal linear module via a horizontal connector.

3. The multi-mode laser group intelligent weeding robot according to claim 2, It is characterized in that The machine vision system includes a pan-tilt platform arranged at the bottom of the forward end of the mobile platform and a depth camera arranged on the pan-tilt platform. The depth camera acquires weed image information to determine the variety, age and growth point of the weeds.

4. The multi-mode laser group intelligent weeding robot according to claim 3, It is characterized in that It also includes a field environment perception system, which includes an inter-row navigation camera arranged at the mobile end of the mobile platform and a Beidou differential positioning system satellite antenna arranged on the top of the mobile platform. The Beidou differential positioning system communicates with satellite signals to locate the weeding robot in real time and transmits coordinate information to the control system, which compares and analyzes the collected coordinate information with the initial reference map; the inter-row navigation camera collects image information in front of the direction of travel, and identifies and detects it to ensure that the weeding robot does not deviate from the row ridge; the control system finally makes a decision through data fusion and issues corresponding steering instructions to the mobile platform.

5. A method for controlling weeds, It is characterized in that Using the multi-mode laser group intelligent weeding robot according to any one of claims 1 to 4, the method comprises: The operating position of the multi-mode laser group intelligent weeding robot is determined through the Beidou differential positioning system and the inter-row navigation camera; When the machine vision system’s depth camera identifies the weeds that need to be removed, it determines the weed’s species, age, and growth point; After the weed age and weed growth point of the weeds are determined, the multi-mode laser group weeding device is controlled to perform weeding.

6. The weed control method according to claim 5, It is characterized in that The method of determining the operating position of the multi-mode laser group intelligent weeding robot by using the Beidou differential positioning system and the inter-row navigation camera includes: The Beidou differential positioning system locates the weeding robot in real time and transmits coordinate information to the control system, which compares and analyzes the collected coordinate information with the initial reference map; The row navigation camera collects image information in the direction of travel and identifies and detects it to ensure that the weeding robot does not deviate from the row; The processing results of the Beidou differential positioning system and the inter-line navigation camera are transmitted to the control system in real time. The control system performs data fusion on the received processing results and finally makes a decision and issues corresponding instructions to the mobile platform.

7. The weed control method according to claim 6, It is characterized in that After the depth camera of the machine vision system determines the weeds that need to be removed, the variety, age and growth point of the weeds are determined, including: The depth camera acquires RGB images and depth images of weeds in the working area; determining the age of the weeds according to the texture features and the number of leaves of the weeds in the depth image; The growth point of the weeds is determined, and the growth point of the weeds is used as the weeding point of the output laser.

8. The weed control method according to claim 7, It is characterized in that After determining the species, age and growth point of the weeds, the multi-mode laser group weeding device is controlled to weed, including: Determining the output laser that needs to be operated according to the variety and age of the weeds, wherein the output laser includes a low-power laser, a high-power laser, or a combination of a low-power laser and a high-power laser; Controlling the vertical linear module and the horizontal linear module corresponding to the output laser to move the output laser to the weed area to control the distance between the focusing lens and the weeds; The angle of the first output laser is adjusted by swinging the steering gear, so that the output laser is aimed at the growth point of the weeds for laser irradiation.

9. The weed control method according to claim 8, It is characterized in that The output laser that needs to be operated is determined according to the age of the weeds, including: If it is determined according to the species and age of the weeds that the weeds are not yet mature and are at a relatively young age, and the tissues are in a light green physiological state and have few leaves, a high-power laser is used for treatment; or, If the weeds are determined to be mature and at a relatively high age according to the grass age, and the tissues show a physiological state of firm texture, dark color, and many leaves, high-power and low-power lasers are used. First, the high-power laser is used for irradiation treatment, and then the mobile platform moves forward a distance, which is the distance between the two output lasers, and then the low-power laser is used to perform a secondary treatment on the weeds.

Citation Information

Patent Citations

  • Intelligent laser weeding robot and weeding method

    CN113632779A

  • Movable weeding device

    CN116569907A

  • Laser weeding machine with shadowless lamp assisting in image recognition and weeding method

    CN118120730A

  • Laser weeding robot and operation method thereof

    CN118340148A

  • Laser weeding trolley

    CN219894406U

Cited By

  • Selective weeding robot with weed growth state detection function

    CN120549054A

  • Laser weeding machine

    CN122123355A