Laser weeding method of weeding robot and laser weeding robot

Abstract

The invention relates to the technical field of agricultural mechanical equipment, and provides a laser weeding method of a weeding robot and the laser weeding robot, the weeding method comprises the following steps: obtaining weed information, and selecting a weeding mode based on the weed information; laser parameters are adjusted based on the selected weeding mode, and path planning is carried out; controlling the weeding robot to advance according to the planned path, controlling the laser beam to align with weeds to perform weeding operation, and continuously feeding back the real-time position of the laser beam; and comparing the real-time position with a predetermined target position, and maintaining the current position or adjusting the current position based on a comparison result. According to the method, the laser parameters can be adjusted based on the weeding mode, the path can be planned, and the mode can be suitable for various complex weeding environments and can adapt to the weeding environments better, so that weeds are accurately and rapidly removed.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery and equipment, and in particular to a laser weeding method of a weeding robot and a laser weeding robot. Background Art

[0002] With the continuous development of agricultural automation technology, laser weeding has attracted more and more attention as a precise and environmentally friendly weeding method. Laser weeding robots can efficiently identify and accurately kill weeds, reducing the impact on the environment and improving agricultural production efficiency. However, the current laser weeding technology still faces many challenges in practical applications.

[0003] In the related art, weeding operations are usually performed with a fixed laser beam, which makes it difficult to achieve high-speed, high-precision weeding operations. Summary of the invention

[0004] The present invention provides a laser weeding method of a weeding robot and a laser weeding robot, which are used to solve the defects of low accuracy and low efficiency of laser weeding in the prior art.

[0005] The present invention provides a laser weeding method for a weeding robot, comprising the following steps: obtaining weed information, and selecting a weed control mode based on the weed information; Adjusting laser parameters based on the selected weeding mode and performing path planning; Control the weeding robot to move along the planned path, control the laser beam to aim at the weeds for weeding, and continuously provide feedback on the real-time position of the laser beam; The real-time position is compared with a predetermined target position, and the current position is maintained or adjusted based on the comparison result.

[0006] According to the laser weeding method of the weeding robot provided by the present invention, weed information is obtained, a weeding mode is selected based on the weed information, and laser parameters are adjusted, which specifically includes the following steps: Obtain the morphology, distribution direction, area size and meristem position of the weeds in the area to be weeded, and select a weeding mode that is suitable for it; Adjust the size and focus of the laser beam based on the weeding mode selected.

[0007] According to the laser weeding method of the weeding robot provided by the present invention, the weeding modes include a sawtooth trajectory mode, a triangle trajectory mode, a horizontal sweep trajectory mode and a vertical sweep mode.

[0008] According to the laser weeding method of the weeding robot provided by the present invention, the real-time position is compared with the predetermined target position, and the current position is maintained or adjusted based on the comparison result, including the following steps: The real-time position point of the laser beam is obtained, and the real-time position point is compared with the preset target position to obtain the positioning error. If the positioning error exceeds the error threshold, automatic adjustment is performed to ensure that the positioning error is within the error threshold.

[0009] According to the laser weeding method of the weeding robot provided by the present invention, the specific steps of automatic adjustment include: S401, pre-set multiple target positions, and obtain the real-time positions of the calibration points; S402, according to the deviation between the real-time position and the target position, the position error is obtained by the following calculation and the position of the laser generator is corrected: ; ; = + , ; in,( , ) is the coordinate position of the initial recognition point, ( ) is the real-time coordinate position of the identification point at the Kth iteration, α is the adjustment factor, and the above steps are continuously iterated until the following stop condition is met: ; in, is the error threshold; S403, selecting multiple target points on the plane to repeat the above steps and perform iterative calibration, specifically including: converting the actual measured positions of the multiple target points ( ) and the preset target position ( , ) is fitted by polynomial regression, using the following regression model for global calibration of all points on the plane: in, , is the calibrated coordinate value of each target point, , , , It represents the adjustment factor of each target point coordinate, and n and m represent the number of summations; The optimal coefficients of the regression model are calculated by the polynomial regression method to minimize the calibration error of all target points: ; in,( , ) is the calibrated coordinate value of each target point, ( , ) is the preset target position; S404, calibrating and adjusting the Z coordinate, specifically including the following: On the specified screen, get the position where the red dot light spot is the smallest, and use this position as the target position of the Z coordinate value.

[0010] A laser weeding robot based on the laser weeding method of the weeding robot described in any one of the above, comprising: a mobile platform, an intelligent laser system, a focusing module and an industrial computer; the intelligent laser system is arranged on the mobile platform, the intelligent laser system comprises a recognition module and a laser module, the recognition module is electrically connected to the laser module, the recognition module is used to obtain visual data of weeds, the laser module comprises a laser generator and a galvanometer mechanism, the laser generator is used to generate laser, and the galvanometer mechanism is used to control the direction of the laser beam; the focusing module is arranged on the mobile platform, the laser system comprises a recognition module and a laser module, the recognition module is electrically connected to the laser module, the recognition module is used to obtain visual data of weeds, the laser module comprises a laser generator and a galvanometer mechanism, the laser generator is used to generate laser, and the galvanometer mechanism is used to control the direction of the laser beam; the focusing module is arranged on the mobile platform, the laser system comprises a recognition module and a laser module, the recognition module comprises a recognition module and a galvanometer mechanism, the laser generator is used to generate laser, and the galvanometer mechanism is used to control the direction of the laser beam; the focusing module is arranged on the mobile platform, the laser system comprises a recognition module and a laser module, the recognition module comprises a recognition module and a laser module, the recognition module comprises a recognition module and a galvanometer mechanism, the laser generator ... The focusing module comprises a focusing frame, a focusing shell, a focusing inner cylinder and a focusing driving component, the focusing frame is connected to the moving platform, the focusing shell is fixedly connected to the focusing frame, the focusing inner cylinder is arranged in the focusing shell, and the focusing driving component is transmission-connected to the focusing inner cylinder so as to drive the focusing inner cylinder to move in a vertical direction through the focusing driving component to realize focusing; the focusing driving component, the intelligent laser system and the focusing module are all electrically connected to the industrial computer, and the industrial computer is used to control the action of the laser weeding robot.

[0011] According to the laser weeding robot provided by the present invention, the galvanometer mechanism includes two groups of galvanometer motors and a reflector, the reflector is arranged at the output end of the galvanometer motor, and the galvanometer motor is used to drive the reflector to rotate.

[0012] According to the laser weeding robot provided by the present invention, a spiral groove is provided on the outer wall surface of the focusing inner cylinder; the focusing module comprises a focusing motor, a focusing ring, a fixing ring and a focusing gear, the fixing ring is connected to the focusing shell and is located at the bottom of the focusing shell, the focusing ring is rotatably connected to the fixing ring, the focusing gear is fixedly connected to the focusing ring, a driving gear is provided on the output end of the focusing motor, the driving gear is meshed with the focusing gear for transmission, a matching portion is provided on the inner wall surface of the focusing ring, and the matching portion matches with the spiral groove of the focusing inner cylinder; wherein, a plurality of guide shafts arranged in a vertical direction are provided at intervals around a circle of the fixing ring, a plurality of sliding holes are provided on one end of the focusing inner cylinder, the guide shaft is penetrated in the sliding hole, so that the focusing inner cylinder moves along the guide shaft under the drive of the focusing motor.

[0013] According to the laser weeding robot provided by the present invention, the mobile platform includes: a fixed base plate, a chassis support and a rotating wheel; a fixed frame is provided on the fixed base plate, the focusing module, the intelligent laser system and the industrial computer are all located in the fixed frame, and mounting seats are provided at the four corner positions of the fixed base plate, each of the mounting seats is provided with a steering motor, the output end of each steering motor is connected to one end of the chassis support, and the other end of the chassis support is connected to the rotating wheel; wherein a shock absorber is provided between the rotating wheel and the chassis support, and the shock absorber is used to absorb vibration during movement.

[0014] According to the laser weeding robot provided by the present invention, a connecting support frame is provided at the bottom of the fixed base plate, and the focusing frame is connected to the connecting support frame.

[0015] The laser weeding method of the weeding robot provided by the present invention can adjust laser parameters based on the weeding mode and plan the path. This method can be applicable to a variety of complex weeding environments and can be adapted to more weeding environments, thereby achieving accurate and rapid removal of weeds. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 The figure is a schematic flow chart of the laser weeding method provided by the present invention.

[0018] Figure 2 It is a schematic diagram of the flow chart of various modes in the laser weeding method provided by the present invention.

[0019] Figure 3 It is a schematic diagram of the automatic adjustment process in the weeding method provided by the present invention.

[0020] Figure 4 It is a logic flow chart of automatic adjustment in the weeding method provided by the present invention.

[0021] Figure 5 This is one of the overall structural schematic diagrams of the laser weeding robot provided by the present invention.

[0022] Figure 6 This is the second schematic diagram of the overall structure of the laser weeding robot provided by the present invention.

[0023] Figure 7The figure is a schematic diagram of the main viewing direction structure of the laser weeding robot provided by the present invention.

[0024] Figure 8 The figure is a schematic diagram of the structure of the laser weeding robot provided by the present invention in a side view direction.

[0025] Fig. 9 The figure is a schematic diagram of the structure of the laser weeding robot provided by the present invention when viewed from an upward direction.

[0026] Fig.10 The figure is a schematic diagram of the structure of the focusing module in the laser weeding robot provided by the present invention.

[0027] Reference numerals: 100, mobile platform; 101, fixed base plate; 102, chassis support; 103, steering motor; 104, rotating wheel; 105, fixed frame; 106, connecting support frame; 200, intelligent laser system; 201, laser module; 202, recognition module; 203, water-cooled chassis; 204, LED light board; 205, laser head; 206, galvanometer connector; 207, laser protection mirror; 300, Power supply; 400, focusing module; 401, focusing frame; 402, focusing shell; 403, focusing inner cylinder; 4031, spiral groove; 4032, movable optical lens; 404, focusing drive assembly; 4041, focusing motor; 4042, guide shaft; 4043, fixing ring; 4044, focusing ring; 4045, focusing gear; 500, galvanometer mechanism; 501, galvanometer motor; 502, reflector. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] In the description of the embodiments of the present invention, it should be noted that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of explaining the embodiments of the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limitations on the embodiments of the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0030] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0031] In the embodiments of the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0032] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present invention. In this specification, the schematic representations of the above terms are not necessarily for the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0033] In order to achieve harmless treatment of weeds, the related technology uses lasers to achieve weeding. Whether the laser weeding equipment in the prior art is based on a robotic arm, a galvanometer deflection design, or a multi-degree-of-freedom hydraulic system, the existing laser weeding robots have the following problems. First, the efficiency is low, the moving speed and range are limited, it is difficult to quickly cover a large area of ​​farmland, it is often difficult to achieve precise weeding in terms of weeding accuracy, and it is difficult to maintain efficient operation in complex terrain or environments with uneven weed distribution. Overall, it is difficult for laser weeding equipment in the related technology to strike a balance between efficiency, accuracy, and environmental adaptability, which seriously limits the promotion and application of laser weeding technology.

[0034] Regarding the problems in related technologies, such as Figure 1 As shown, this embodiment provides a laser weeding method for a weeding robot, comprising the following steps: Step S100, obtaining weed information, and selecting a weeding mode based on the weed information. In conventional laser weeding technical solutions, fixed laser beams are mostly used. Fixed laser beams are difficult to achieve adaptive adjustment during weeding, which results in weak versatility. Fixed laser beams are difficult to achieve precise weeding for weed forms in different situations, resulting in difficulty in improving weeding efficiency and weeding quality. In this embodiment, a pre-set weeding model enables different weeding modes to be achieved in different environments, so as to achieve precise and rapid weeding.

[0035] The specific implementation method of step 100 includes the following steps: Obtain the morphology, distribution direction, area size and meristem position of the weeds in the weeding area, and select a weeding mode that matches it. That is, before starting the laser for weeding, it is necessary to obtain the information of the weeds in the weeding area (i.e., weed morphology, distribution direction, area size and meristem position) so that the appropriate weeding mode can be matched based on the acquired information, so as to achieve targeted weeding and improve the accuracy and efficiency of weeding.

[0036] Specifically, to obtain weed information, image acquisition can be achieved through an acquisition module, such as a camera, and then the acquired image can be analyzed using the analysis algorithm of the industrial computer, so that different weeding modes can be matched based on the analysis results, and different laser parameters can be adjusted based on different modes, and the adjusted laser can be adapted to the weed information, so as to achieve accurate and efficient weeding operations.

[0037] In specific implementation, the pre-trained weed control model can be configured in the processing system. The processing system can call the relevant model according to the weed information obtained in the field (i.e., weed morphology, distribution direction, area size, and meristem position), thereby realizing different weed control modes and achieving efficient treatment of weeds in different situations.

[0038] Step S200, adjust laser parameters based on the selected weeding mode and perform path planning. After obtaining the weeding information, the relevant laser parameter values ​​can be obtained, and the industrial computer can control it to reach each parameter in the mode, and can also control the robot's motion trajectory, so as to achieve different weeding modes for different weed states.

[0039] Specifically, adjusting the laser parameters includes adjusting the size and focal length of the laser beam based on the selected weeding mode. Different spot sizes and focal lengths are required for different weed forms to achieve precise weeding. In this embodiment, after selecting the weeding mode, the parameters that need to be adjusted can be obtained, and then adjusted based on the parameters so that it can match the weeding mode to achieve the effect of precise weeding.

[0040] Different weeds have different shapes and require different laser beam sizes. A fixed laser beam is difficult to remove when a single weed is too large, and when a single weed is too small, it will waste laser energy and affect the crops around the weeds. In the embodiment, the selected weeding mode can adjust the laser parameters so that the laser beam size can be adaptively adjusted based on the actual weed shape, thereby achieving accurate and efficient weed removal.

[0041] Step S300, control the weeding robot to move forward along the planned path, control the laser beam to aim at the weeds for weeding, and continuously feedback the real-time position of the laser beam. After determining the path and weeding method, the industrial computer can control the weeding robot to start, move forward along the preset path, and remove the weeds in the process of moving forward.

[0042] Step S400, compare the real-time position with the predetermined target position, and maintain or adjust the current position based on the comparison result. The position of the laser beam irradiated by the weeding robot during operation will directly affect the accuracy of weeding and the efficiency of the operation, which requires ensuring the accuracy of the real-time position during movement. This embodiment collects the irradiation position of the laser beam and compares it with the target position, and can adjust it based on the comparison result, thereby realizing automatic calibration of the laser beam position and improving the accuracy of laser beam irradiation.

[0043] Specifically, the real-time position point of the laser beam is obtained, and the real-time position point is compared with the preset target position to obtain the positioning error. If the positioning error exceeds the error threshold, automatic adjustment is performed to ensure that the positioning error is within the error threshold. An error threshold is configured in the industrial computer, and the positioning error of the implementation position at the target position is compared with the error threshold to determine whether to start automatic calibration.

[0044] Furthermore, when the positioning error is within the error threshold range, it indicates that the real-time position of the laser beam is within the normal range, and no adjustment is required. When the positioning error exceeds the error threshold, it indicates that the real-time position of the laser beam has deviated from the normal range, and adjustment is required to ensure that the laser beam is always within the normal range. In this way, real-time calibration can be achieved, which improves the quality and accuracy of weeding.

[0045] In specific implementation, it can be realized through a red dot transmitter. Since laser is invisible light, laser calibration can be achieved through a red dot light source, thereby achieving real-time calibration based on the calibrated laser beam.

[0046] In a specific embodiment, the weeding modes include a sawtooth track mode, a triangle track mode, a horizontal sweep track mode, and a vertical sweep mode.

[0047] For example, when the weeds are small and distributed in long and thin strips, the weeding robot will prefer the zigzag trajectory mode. The growth direction of such weeds is relatively uniform but the area is not large. Through the zigzag scanning path, the robot can partially cover the same area multiple times in a short working stroke, thereby accurately capturing and removing the long and thin weeds.

[0048] Movement characteristics: The trajectory is a continuous "Z"-shaped or sawtooth-shaped lateral movement. During the movement, the industrial computer continuously adjusts the operating range of the weeding robot, allowing the robot to complete multiple scans in a small lateral offset. Through the sawtooth trajectory mode, while maintaining a high operating efficiency, it can effectively improve the accuracy of capturing and removing slender weeds, achieving accurate and fast weeding operations.

[0049] When the weed area is small and scattered, and the image features show that the outline of the weed meristem is obvious, the triangular trajectory mode is suitable. This mode is suitable for precise treatment of weed groups that are distributed in multiple points and are small but have clear structures.

[0050] Movement characteristics: The trajectory is a continuous triangular sweep of multiple segments, and the weeding robot turns at each sharp corner to achieve multi-angle coverage of the target weed area. This trajectory ensures that the same weed area is covered multiple times in different directions, increasing the thoroughness of the removal. The triangular trajectory mode can accurately locate and remove scattered weeds point by point, while making full use of multi-angle coverage to improve weeding accuracy.

[0051] When the detected weed area is large, the weed growth direction is mainly horizontal, and the meristem is not obvious, and a large area needs to be covered quickly, the sweeping trajectory mode is preferred.

[0052] Movement characteristics: The weeding robot continuously sweeps the target area horizontally with a large horizontal stride. By moving over a wide horizontal bandwidth, it reduces repeated coverage and excessive fine adjustments, and improves the efficiency of large-area processing. The sweeping trajectory mode focuses on operating efficiency, quickly covers and cleans weeds distributed horizontally over a large area, ensuring that weeding operations are completed in a shorter time and improving overall production capacity.

[0053] When the weed area is large and shows a clear vertical growth trend, the meristem is not prominent, and continuous attack from top to bottom or bottom to top is required, the vertical sweeping mode is used.

[0054] Movement characteristics: The weeding robot performs continuous up and down sweeping in the longitudinal direction. Compared with the horizontal sweeping mode, this mode is more suitable for vertically arranged weeds. By repeatedly sweeping vertically, it can achieve more uniform coverage and cleaning from the root to the top. The vertical sweeping mode improves the weeding effect of a single sweep by conforming to the characteristics of vertically distributed weeds, which helps to effectively control tall or vertically distributed weeds within a shorter longitudinal path.

[0055] In the above modes, the control unit automatically selects the corresponding trajectory mode based on the visual recognition results (weed density, morphological characteristics, distribution direction and area size) and the set error threshold. When the weed situation changes dynamically (such as from multiple small weed areas to large concentrated areas, or from horizontal expansion to scattered distribution), the system can switch the trajectory mode online, thereby achieving an efficient and intelligent weeding process.

[0056] In combination with the above embodiment, the positioning error needs to be automatically adjusted when it exceeds the error threshold. This embodiment provides a solution for automatic adjustment. The specific steps of the automatic adjustment include: S401, pre-set multiple target positions, and obtain the real-time position of the calibration point. Specifically, the industrial computer pre-sets multiple target positions and sends these target positions to the red dot transmitter, where the red dot transmitter is used to calibrate the position of the laser. The red dot transmitter emits a red dot to the camera's field of view in different planes at the received target position. After the camera receives the red dot, it uses image processing technology to locate the red dot and obtain the real-time position coordinates of the red dot. The actual position coordinates are the real-time position coordinates of the laser beam.

[0057] It should be understood that the red dot emitter is integrated into the laser generator. The laser generated by the laser generator is invisible light, while the red dot emitter is visible light. The red light emitted by the red dot emitter can facilitate the acquisition of the position and realize the calibration of the laser position.

[0058] S402, the camera feeds back the acquired real-time position to the industrial computer, and the industrial computer obtains the position error and corrects the position of the laser generator through the following calculation based on the deviation between the real-time position and the target position: ; ; = + , ; in,( , ) is the coordinate position of the initial recognition point, ( ) is the real-time coordinate position of the identification point at the Kth iteration, α is the adjustment factor, and the iteration is continued until the following stop condition is met: ; in, is the error threshold; S403, selecting multiple target points on the plane to repeat the above steps and perform iterative calibration, specifically including: converting the actual measured positions of the multiple target points ( ) and the preset target position ( , ) is fitted by polynomial regression, using the following regression model for global calibration of all points on the plane: in, , is the calibrated coordinate value of each target point, , , , It represents the adjustment factor of each target point coordinate, and n and m represent the number of summations; The optimal coefficients of the regression model are calculated by the polynomial regression method to minimize the calibration error of all target points: ; in,( , ) is the calibrated coordinate value of each target point, ( , ) is the preset target position; S404, calibrating and adjusting the Z coordinate, specifically including the following: On the specified screen, obtain the position where the red dot spot is the smallest, and use this position as the target position of the Z coordinate value. Specifically, the red light beam emitted by the red dot emitter can realize position calibration, and the depth camera can obtain the depth value of the corresponding point through the pixel coordinates in the field of view, and obtain the three-dimensional coordinate value of the camera coordinate system of the corresponding point through the internal reference coordinate conversion ( ) to obtain the Z coordinate of the laser beam target.

[0059] Specifically, when the red spot is the smallest, the red spot size can be compared by color segmentation. Specifically, the size of the spot is observed by adjusting the movable optical lens up and down. When it is the smallest, it is assumed to be focused on this plane at this time, and the Z coordinate value at this time is used as its target position. In specific implementation, by observing the spot size, when the spot size is less than 1mm, it can be regarded as focused, and this is used as the target position.

[0060] It is understandable that during daily use, the industrial computer continuously monitors the positioning accuracy of the laser beam. When it is detected that the positioning error of the laser beam exceeds the preset threshold value ϵ, the self-repair mechanism is automatically started. Specifically, the industrial computer monitors the red dot position obtained by the camera in real time and compares it with the preset target position to detect whether there is a positioning error exceeding ϵ. When the error exceeds the threshold, the industrial computer automatically resends the red dot signal, and identifies the new red dot position through the camera, calculates the new positioning error, and automatically adjusts the position of the red dot transmitter and the galvanometer according to the new error value, and re-corrects the position to ensure that the laser beam is accurately aligned with the target weeds. After the adjustment is completed, the system performs error detection again to ensure that the error has been reduced to within ϵ, ensuring the continuous high-precision operation of the laser weeding robot. Through the above process, the industrial computer continuously corrects the launch position of the red dot transmitter, and finally realizes high-precision, fully automatic positioning calibration and error repair. This automatic calibration and self-repair method has the advantages of high efficiency, accuracy and no need for human intervention, and can be widely used in scenarios requiring high-precision positioning and calibration. Through the above technical solution, the present invention can effectively reduce the errors in the traditional manual calibration process, improve the automation level and accuracy of the system, and has high industrial application value.

[0061] The following is a laser weeding robot provided by the present invention. The following weeding robot can execute the laser weeding method provided by the above embodiment. The structure of the laser weeding robot is described below.

[0062] like Figure 5-10As shown, a laser weeding robot based on the laser weeding method of the weeding robot provided by any of the above embodiments includes a mobile platform 100, an intelligent laser system 200, a focusing module 400 and an industrial computer, wherein the intelligent laser system 200 is arranged on the mobile platform 100, and the intelligent laser system 200 includes a recognition module 202 and a laser module 201, wherein the recognition module 202 is electrically connected to the laser module 201, and the recognition module 202 is used to obtain visual data of weeds, and the laser module 201 includes a laser generator and a galvanometer mechanism 500, wherein the laser generator is used to generate laser light, and the galvanometer mechanism 500 is used to control the direction of the laser beam; the focusing module 400 Arranged on the mobile platform 100, the focusing module 400 includes a focusing frame 401, a focusing shell 402, a focusing inner cylinder 403 and a focusing drive component 404. The focusing frame 401 is connected to the mobile platform 100, the focusing shell 402 is fixedly connected to the focusing frame 401, the focusing inner cylinder 403 is arranged in the focusing shell 402, and the focusing drive component 404 is transmission-connected with the focusing inner cylinder 403 so as to drive the focusing inner cylinder 403 to move in a vertical direction through the focusing drive component 404 to achieve focusing; the focusing drive component 404, the intelligent laser system 200 and the focusing module 400 are all electrically connected to the industrial control computer, and the industrial control computer is used to control the movement of the laser weeding robot. When performing weeding operations by laser, it is necessary to control the movement of the weeding robot and drive the movement of the intelligent laser system 200 through the movement of the weeding robot to achieve weed removal. In this embodiment, the overall movement is achieved through the mobile platform 100, and the laser emission is achieved through the intelligent laser system 200. Then, the laser beam is adjusted in size and direction through the focusing module 400, thereby achieving precise weeding operations. Through the setting of the focusing module 400, the spot size and automatic focus of the laser beam can be adjusted, thereby improving the accuracy and efficiency of weeding.

[0063] Specifically, the mobile platform 100 is a laser robot carrying platform, which can realize the carrying connection of different component equipment, and drive the overall movement to perform weeding operations through the movement of the mobile platform 100 itself. The recognition module 202 includes a camera, which is used to capture real-time images in the farmland environment and provide visual data of weeds (i.e., weed morphology, distribution direction, area size, and meristem position). The industrial computer is a high-performance computer that can process data from the camera and other sensors and execute push recognition and path planning algorithms.

[0064] According to the laser weeding robot of the weeding robot provided by the present invention, the galvanometer mechanism 500 includes two groups of galvanometer motors 501 and a reflector 502 , the reflector 502 is arranged at the output end of the galvanometer motor 501 , and the galvanometer motor 501 is used to drive the reflector 502 to rotate.

[0065] In the specific setting, a power supply 300 is provided on the mobile platform 100, and the power supply 300 supplies power for the movement of the entire laser weeding robot. A high-precision inertial navigation sensor is also provided on the mobile platform 100, and the inertial navigation sensor is used to monitor the posture changes of the robot, including speed, acceleration, rotation angle and overturning angle, so as to be able to provide real-time feedback on the current state of the robot. An LED light board 204 is also provided at the bottom of the mobile platform 100, which can provide necessary lighting to ensure that the color camera can clearly capture images under different lighting conditions. In the specific connection, a connecting support frame 106 is fixedly provided at the bottom of the mobile platform 100, and the camera, LED light board 204, and inertial navigation sensor are fixedly provided on the connecting support frame 106. The LED light board 204 and the camera are connected to the power supply 300 and the industrial computer through the Type-C data cable through the threading hole to ensure the stability and efficiency of data transmission.

[0066] When working, the combination of LED light board 204 and camera is used to realize stable extraction of image data in complex lighting environment. The industrial computer is used to distinguish weeds and crops in real time and issue weeding instructions. At the same time, the robot motion posture is obtained by combining high-precision inertial navigation sensors to realize stable tracking of weeds.

[0067] The laser module 201 also includes a water-cooled chassis 203 , which effectively dissipates heat through a liquid cooling system, thereby preventing the laser generator from overheating and improving the continuous working ability of the system.

[0068] The galvanometer mechanism 500 includes two groups of galvanometer motors 501 and a reflector 502. The reflector 502 is arranged at the output end of the galvanometer motor 501. The galvanometer motor 501 is used to drive the reflector 502 to rotate. The galvanometer motor 501 drives the reflector 502 to move, thereby realizing the deflection and scanning of the laser beam.

[0069] In the specific setting, the galvanometer mechanism 500 also includes a galvanometer connecting seat 206 and a laser protection mirror 207. The galvanometer connecting seat 206 is connected to the focusing frame 401, and the laser protection mirror 207 is arranged on the galvanometer connecting seat 206. The setting of the galvanometer connecting seat 206 and the laser protection mirror 207 can ensure the seamless connection between the galvanometer mechanism 500 and the laser generator to prevent the laser beam from leaking.

[0070] During specific operation, the laser generated by the laser generator is transmitted to the laser head 205 through the optical fiber. The laser head 205 controls the beam diameter to ensure that the spot area is smaller than the mirror surface of the reflector 502. Finally, the reflector 502 is deflected by the galvanometer motor 501 to achieve rapid and accurate control of the laser beam.

[0071] In combination with the above-mentioned embodiment, a spiral groove 4031 is provided on the outer wall of the focusing inner cylinder 403; the focusing module 400 includes a focusing motor 4041, a focusing ring 4044, a fixing ring 4043 and a focusing gear 4045, the fixing ring 4043 is connected to the focusing housing 402 and is located at the bottom of the focusing housing 402, the focusing ring 4044 is rotatably connected to the fixing ring 4043, the focusing gear 4045 is fixedly connected to the focusing ring 4044, and a focusing motor 4041 is provided at the output end thereof. The driving gear meshes with the focusing gear 4045 for transmission. A matching portion is provided on the inner wall surface of the focusing ring 4044, and the matching portion matches with the spiral groove 4031 of the focusing inner cylinder 403. Among them, a plurality of guide shafts 4042 arranged in the vertical direction are provided at intervals around the fixed ring 4043. A plurality of sliding holes are provided on one end of the focusing inner cylinder 403. The guide shafts 4042 are penetrated in the sliding holes so that the focusing inner cylinder 403 moves along the guide shafts 4042 under the drive of the focusing motor 4041. The spot size of the laser is different in different weeding modes, and the focusing operation needs to be realized through the focusing module 400. In this embodiment, the driving of the focusing ring 4044 is realized by adjusting the meshing of the gears on the motor. The rotation of the focusing ring 4044 can drive the focusing inner cylinder 403 to move along the guide shaft 4042, thereby realizing the focusing operation.

[0072] Specifically, a movable optical lens 4032 is provided at the bottom of the focusing inner cylinder 403, and a fixed optical lens is provided above the focusing inner cylinder 403. By adjusting the focusing inner cylinder 403 to move up and down, the movable optical lens 4032 can be moved away from or close to the fixed optical lens, thereby realizing laser focusing.

[0073] It can be understood that the guide shaft 4042 can improve the stability of vertical movement, thereby improving the stability of focusing and achieving stable focusing operation. The gear drive method improves the stability of transmission and can achieve precise control of the focusing distance, thereby improving its accuracy.

[0074] During operation, the laser beam enters the focusing module 400 through the reflector 502, and the focusing motor 4041 drives the focusing gear 4045 to rotate, thereby rotating the focusing ring 4044. The focusing ring 4044 and the focusing inner cylinder 403 are threadedly matched (i.e., the matching part matches the spiral groove 4031), so that the focusing inner cylinder 403 can move up and down along the guide shaft 4042 to achieve focusing.

[0075] like Fig.10As shown, one end of the focusing inner cylinder 403 is provided with an edge portion of a circular ring structure, the edge portion protrudes from the outer wall of the focusing inner cylinder 403, and a plurality of sliding holes are opened on the edge portion, and the sliding holes are slidably matched with the guide shaft 4042, so that the focusing inner cylinder 403 moves up and down along the guide shaft 4042 under the drive of the focusing motor 4041. Among them, the guide shaft 4042 is fixedly connected, one end of the guide shaft 4042 is fixedly connected, and the other end of the guide shaft 4042 is connected to the focusing housing 402, and the focusing ring 4044 is rotatably connected to the fixing ring 4043, so that the focusing ring 4044 rotates relative to the fixing ring 4043 under the drive of the focusing motor 4041.

[0076] According to the laser weeding robot provided by the present invention, the mobile platform 100 includes a fixed base plate 101, a chassis support 102 and a rotating wheel 104; a fixed frame 105 is provided on the fixed base plate 101, and the focusing module 400, the intelligent laser system 200 and the industrial computer are all located in the fixed frame 105, and mounting seats are provided at the four corners of the fixed base plate 101, and a steering motor 103 is provided in each mounting seat, and the output end of each steering motor 103 is connected to one end of the chassis support 102, and the other end of the chassis support 102 is connected to the rotating wheel 104; wherein, a shock absorber is provided between the rotating wheel 104 and the chassis support 102, and the shock absorber is used to absorb vibration during movement. The mobile platform 100 is an integral support and mobile base, and the overall structural strength is improved by the setting of the fixed frame 105.

[0077] Specifically, the rotating wheel 104 is a self-driving wheel, and the rotating wheel 104 is electrically connected to the industrial computer so as to control the start or shut down of the rotating wheel 104 through the industrial computer. The steering motor 103 is used to control the selected direction of the chassis support 102 and the rotating wheel 104 to achieve precise navigation, and the setting of the shock absorber improves the overall stability.

[0078] When working, the chassis support 102 is connected to the wheel hub fixing seat by bolts, and the two ends of the spring of the shock absorber are fixed on the chassis support 102 to provide an adjustable shock absorption effect. The steering motor 103 realizes the precise navigation and motion control of the robot by controlling the rotation of the chassis support 102 and the wheels, ensuring stable movement in complex terrain. The fixed frame 105 serves as the skeleton of the entire weeding robot, firmly connecting and protecting each system module, ensuring that the robot can operate stably and efficiently in a complex farmland environment. Through reasonable structural design, the orderly integration and efficient collaborative work of each system module are achieved.

[0079] In a specific implementation, a connecting support frame 106 is provided at the bottom of the fixed base plate 101, and the focusing frame 401 is connected to the connecting support frame 106. The connection between the focusing frame 401 and the connecting support frame 106 makes the overall structural stability of the focusing module 400 higher, thereby improving the overall stability.

[0080] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment achieves a comprehensive improvement in efficiency, accuracy and environmental adaptability of laser weeding operations through the flexible selection of adaptive zoom modules, recognition modules 202 and multiple weeding modes, and effectively reduces energy consumption and maintenance costs. Furthermore, various modes can automatically adjust the laser focus according to the height and distance of the weeds, so that the laser energy is accurately concentrated on the roots of the weeds. This method helps to completely remove weeds and prevent their secondary growth and recurrence. The visual recognition system can quickly and accurately distinguish weed targets and accurately transmit their coordinates to the laser control module, thereby improving weeding accuracy and stability. Furthermore, the present invention realizes fully automatic and high-precision laser beam positioning calibration through the coordination of the red dot transmitter and the camera, combined with multiple rounds of iterations and error correction. Calibration: When the initial installation or environmental changes, the initial calibration can be completed quickly, saving manual debugging time and maintenance costs. Self-repair: During daily use, if the laser beam deviation is detected to exceed the threshold, the system automatically triggers the repair process, resends the red dot and calculates the positioning error, dynamically adjusts the position of the red dot emitter and the galvanometer, and ensures that the laser beam is always accurately aimed at the target. The calibration and self-repair functions can be flexibly switched to ensure that the equipment maintains high-precision and stable weeding operations for a long time, while minimizing manual intervention and improving system reliability and adaptability. The above calibration and self-repair functions can be flexibly switched under different working conditions to ensure that the equipment is in a high-precision and stable weeding state for a long time, while minimizing the need for manual intervention and maintenance costs, and improving the reliability and adaptability of the overall system.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A laser weeding method for a weeding robot, characterized in that: The following steps are involved: obtaining weed information, and selecting a weed control mode based on the weed information; Adjusting laser parameters based on the selected weeding mode and performing path planning; Control the weeding robot to move along the planned path, control the laser beam to aim at the weeds for weeding, and continuously provide feedback on the real-time position of the laser beam; The real-time position is compared with a predetermined target position, and the current position is maintained or adjusted based on the comparison result.

2. The laser weeding method of the weeding robot according to claim 1, characterized in that: Acquiring weed information, selecting a weeding mode based on the weed information, and adjusting laser parameters specifically include the following steps: Obtain the morphology, distribution direction, area size and meristem position of the weeds in the area to be weeded, and select a weeding mode that is suitable for it; Adjust the size and focus of the laser beam based on the weeding mode selected.

3. The laser weeding method of the weeding robot according to claim 2, characterized in that: The weeding modes include a sawtooth track mode, a triangle track mode, a horizontal sweep track mode and a vertical sweep mode.

4. The laser weeding method of the weeding robot according to claim 1, characterized in that: Comparing the real-time position with a predetermined target position, and maintaining or adjusting the current position based on the comparison result, comprises the following steps: The real-time position point of the laser beam is obtained, and the real-time position point is compared with the preset target position to obtain the positioning error. If the positioning error exceeds the error threshold, automatic adjustment is performed to ensure that the positioning error is within the error threshold.

5. The laser weeding method of the weeding robot according to claim 4, characterized in that: The specific steps for automatic adjustment include: S401, pre-set multiple target positions, and obtain the real-time positions of the calibration points; S402, according to the deviation between the real-time position and the target position, the position error is obtained by the following calculation and the position of the laser generator is corrected: ； ； = + , ； in,( , ) is the coordinate position of the initial recognition point, ( ) is the real-time coordinate position of the identification point at the Kth iteration, α is the adjustment factor, and the above steps are continuously iterated until the following stop condition is met: ； in, is the error threshold; S403, selecting multiple target points on the plane to repeat the above steps and perform iterative calibration, specifically including: converting the actual measured positions of the multiple target points ( ) and the preset target position ( , ) is fitted by polynomial regression, using the following regression model for global calibration of all points on the plane: in, , is the calibrated coordinate value of each target point, , , , It represents the adjustment factor of each target point coordinate, and n and m represent the number of summations; The optimal coefficients of the regression model are calculated by the polynomial regression method to minimize the calibration error of all target points: ； in,( , ) is the calibrated coordinate value of each target point, ( , ) is the preset target position; S404, calibrating and adjusting the Z coordinate, specifically including the following: On the specified screen, get the position where the red dot light spot is the smallest, and use this position as the target position of the Z coordinate value.

6. A laser weeding robot based on the laser weeding method of the weeding robot according to any one of claims 1 to 5, characterized in that: include: Mobile platforms; An intelligent laser system, the intelligent laser system is arranged on the mobile platform, the intelligent laser system comprises an identification module and a laser module, the identification module is electrically connected to the laser module, the identification module is used to obtain visual data of weeds, the laser module comprises a laser generator and a galvanometer mechanism, the laser generator is used to generate laser, and the galvanometer mechanism is used to control the direction of the laser beam; A focusing module, the focusing module is arranged on the mobile platform, the focusing module comprises a focusing frame, a focusing outer shell, a focusing inner cylinder and a focusing drive assembly, the focusing frame is connected to the mobile platform, the focusing outer shell is fixedly connected to the focusing frame, the focusing inner cylinder is arranged in the focusing outer shell, and the focusing drive assembly is drivingly connected to the focusing inner cylinder, so that the focusing drive assembly drives the focusing inner cylinder to move in a vertical direction to achieve focusing; An industrial computer, the focus drive component, the intelligent laser system and the focus module are all electrically connected to the industrial computer, and the industrial computer is used to control the movement of the laser weeding robot.

7. The laser weeding robot according to claim 6, characterized in that: The galvanometer mechanism includes two groups of galvanometer motors and a reflector. The reflector is arranged at the output end of the galvanometer motor. The galvanometer motor is used to drive the reflector to rotate.

8. The laser weeding robot according to claim 6, characterized in that: A spiral groove is provided on the outer wall surface of the focusing inner cylinder; the focusing module comprises a focusing motor, a focusing ring, a fixing ring and a focusing gear, the fixing ring is connected to the focusing outer shell and is located at the bottom of the focusing outer shell, the focusing ring is rotatably connected to the fixing ring, the focusing gear is fixedly connected to the focusing ring, a driving gear is provided on the output end of the focusing motor, the driving gear is meshed with the focusing gear for transmission, a matching portion is provided on the inner wall surface of the focusing ring, and the matching portion matches with the spiral groove of the focusing inner cylinder; Among them, a plurality of guide shafts arranged in the vertical direction are arranged at intervals around a circle of the fixing ring, a plurality of sliding holes are opened on one end of the focusing inner cylinder, and the guide shafts are passed through the sliding holes so that the focusing inner cylinder moves along the guide shafts under the drive of the focusing motor.

9. The laser weeding robot according to claim 6, characterized in that: The mobile platform comprises: a fixed bottom plate, a chassis support and rotating wheels; The fixed base plate is provided with a fixed frame, the focusing module, the intelligent laser system and the industrial computer are all located in the fixed frame, the four corners of the fixed base plate are provided with mounting seats, each of the mounting seats is provided with a steering motor, the output end of each steering motor is connected to one end of the chassis support, and the other end of the chassis support is connected to the rotating wheel; Wherein, a shock absorber is provided between the rotating wheel and the chassis support, and the shock absorber is used to absorb vibration during the movement.

10. The laser weeding robot according to claim 9, characterized in that: A connecting support frame is provided at the bottom of the fixed base plate, and the focusing frame is connected to the connecting support frame.

Images