Self-adaptive parameter adjusting system of soil leaf vegetable harvester

Abstract

The invention discloses a self-adaptive parameter adjusting system of an over-soil leafy vegetable harvester, and relates to the technical field of parameter adjustment. Performing dynamic morphology reconnaissance on the soil leaf vegetables through a reconnaissance mechanism to obtain multi-dimensional morphology feature information; calling a flexible flapping strategy to analyze the multi-dimensional morphology feature information to obtain a flexible flapping control scheme; on the basis of a flexible flapping control scheme, flapping and carding the on-soil leafy vegetables to obtain pre-carded on-soil leafy vegetables; activating a supporting harvesting mechanism of the harvester for the leaf vegetables on the soil, and supporting and harvesting the leaf vegetables on the soil through the supporting harvesting mechanism. The technical problems that in the prior art, when leaf vegetables on soil are harvested, flexible adjustment cannot be achieved according to actual crop conditions, crop damage is caused during harvesting, harvesting is incomplete, and efficiency is low are solved, and by conducting flexible carding before harvesting, the technical effect that leaf vegetables on soil are harvested efficiently, completely and losslessly is achieved.

Description

Technical Field

[0001] This invention relates to the field of parameter adjustment technology, and more specifically to an adaptive parameter adjustment system for a leafy vegetable harvester. Background Technology

[0002] With the rapid development of agricultural automation technology, mechanized harvesting of leafy vegetables is gradually being applied in modern agricultural production. However, due to the complex growing environments of different leafy vegetable crops, and the varying heights, densities, and growth patterns of leafy vegetables, traditional harvesting equipment cannot be flexibly adjusted according to the actual crop conditions, easily leading to problems such as crop damage, incomplete harvesting, and low efficiency. Traditional leafy vegetable harvesting equipment usually relies on fixed parameters, which cannot adapt to diverse field environments and crop conditions, making it difficult to meet the demands of modern agriculture for efficient and precise operations. Summary of the Invention

[0003] This application provides an adaptive parameter adjustment system for a leafy vegetable harvester, which is used to address the technical problems of existing technologies that cannot flexibly adjust parameters according to the actual crop conditions when harvesting leafy vegetables, resulting in crop damage, incomplete harvesting, and low efficiency.

[0004] In view of the above problems, this application provides an adaptive parameter adjustment system for a leafy vegetable harvester.

[0005] This application provides an adaptive parameter adjustment system for a leafy vegetable harvester, the system comprising: The system comprises the following components: an initial machine position setting module, which sets the initial harvesting position of the leafy vegetable harvester based on multi-dimensional planting characteristics of the leafy vegetables; a morphology reconnaissance module, which uses the reconnaissance mechanism of the harvester to dynamically reconnoiter the leafy vegetables and obtain multi-dimensional morphological feature information; a control scheme determination module, which uses a flexible beating strategy to analyze the multi-dimensional morphological feature information and obtain a flexible beating control scheme; a beating and combing module, which uses the front combing mechanism of the harvester to beate and comb the leafy vegetables based on the flexible beating control scheme to obtain pre-combed leafy vegetables; and a support harvesting module, which activates the support harvesting mechanism of the harvester and uses the support harvesting mechanism to harvest the pre-combed leafy vegetables.

[0006] The technical solution provided in this application has at least the following technical effects or advantages: This application sets the initial harvester position of the leafy vegetable harvester based on the multi-dimensional planting characteristics of the leafy vegetables; the reconnaissance mechanism of the leafy vegetable harvester performs dynamic morphological reconnaissance of the leafy vegetables to obtain multi-dimensional morphological characteristic information; a flexible beating strategy is called to analyze the multi-dimensional morphological characteristic information to obtain a flexible beating control scheme; the front combing mechanism of the leafy vegetable harvester performs beating and combing of the leafy vegetables based on the flexible beating control scheme to obtain pre-combed leafy vegetables; the support harvesting mechanism of the leafy vegetable harvester is activated, and the pre-combed leafy vegetables are harvested with support through the support harvesting mechanism. This invention addresses the technical problems of existing technologies for harvesting leafy vegetables on the ground, which cannot flexibly adjust to the actual crop conditions, resulting in crop damage, incomplete harvesting, and low efficiency. By acquiring multi-dimensional planting characteristic information of the leafy vegetables, the initial position of the harvester is dynamically set, and the morphological characteristics of the leafy vegetables are acquired in real time using a reconnaissance mechanism. Subsequently, a flexible beating strategy is invoked to generate a control scheme, in which a pre-positioned combing mechanism gently beats and combs the leafy vegetables. Finally, the supporting harvesting mechanism is activated to ensure that the leafy vegetables are stable and harvested accurately, achieving the technical effect of efficient, complete, and damage-free harvesting of leafy vegetables on the ground. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A schematic diagram of the adaptive parameter adjustment system for a leafy vegetable harvester provided in an embodiment of this application; Figure 2 This is a schematic flowchart of the adaptive parameter adjustment method for a leafy vegetable harvester provided in an embodiment of this application.

[0009] Explanation of reference numerals in the attached diagram: Initial camera position setting module 11, topography reconnaissance module 12, control scheme determination module 13, patting and combing module 14, and supporting harvesting module 15. Detailed Implementation

[0010] This application provides an adaptive parameter adjustment system for a leafy vegetable harvester, addressing the technical problems of existing technologies that fail to flexibly adjust to actual crop conditions during harvesting of leafy vegetables, resulting in crop damage, incomplete harvesting, and low efficiency. By acquiring multi-dimensional planting characteristic information of the leafy vegetables, the system dynamically sets the harvester's initial position and uses a reconnaissance mechanism to acquire real-time leafy vegetable morphological characteristics. Subsequently, a flexible beating strategy is invoked to generate a control scheme, with a pre-positioning mechanism gently beating and combing the leafy vegetables. Finally, the supporting harvesting mechanism is activated to ensure the leafy vegetables are securely and accurately harvested, achieving efficient, complete, and damage-free harvesting of leafy vegetables.

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0012] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0013] like Figure 1 As shown, this application provides an adaptive parameter adjustment system for a leafy vegetable harvester, used to perform actions such as... Figure 2 The adaptive parameter adjustment method for a topsoil leafy vegetable harvester is shown. The adaptive parameter adjustment system of the topsoil leafy vegetable harvester is communicatively connected to the harvester. The system includes: The initial machine position setting module 11 sets the initial harvesting machine position of the leafy vegetable harvester according to the multi-dimensional planting characteristic information of the leafy vegetables.

[0014] In this embodiment, multi-dimensional planting feature information is first obtained, including multiple sets of planting data for multi-hole leafy vegetables. Among this data, the planting data of the first hole leafy vegetable is matched, and its ridge width and hole width are extracted. Next, the planting location of the leafy vegetable is calculated based on these two parameters. Then, the planting location is analyzed to form an initial harvesting path, which includes the coordinates and orientation of the harvesting start point. Finally, the initial harvesting path is used as the initial harvester position.

[0015] Furthermore, in the system provided in the application embodiment, the initial machine position setting module 11 is also used for: The multidimensional planting feature information includes multiple sets of planting data for multi-hole leafy vegetables; the first planting data of a first-hole leafy vegetable is matched among the multiple sets of planting data, where the first-hole leafy vegetable is any hole of the multi-hole leafy vegetable, and the first planting data includes a first planting ridge width and a first planting hole width; the first planting point of the first-hole leafy vegetable is obtained based on the first planting ridge width and the first planting hole width; the first planting point is analyzed to form an initial harvesting path, where the initial harvesting path includes a harvesting start point; the coordinates and orientation of the harvesting start point are used as the initial harvester position.

[0016] In this embodiment of the application, the multidimensional planting feature information includes multiple sets of planting data for leafy vegetables in multiple holes, which are acquired by preset sensors such as lidar and cameras. The multidimensional planting feature information includes the growth height of the leafy vegetables, their distribution, the width of the ridges and the width of the holes in the planting area, etc.

[0017] After acquiring multiple sets of planting data, a random algorithm was used to randomly select the first leafy vegetable in each planting hole. This leafy vegetable was one of many planted leafy vegetables. Then, the planting data for this leafy vegetable was traversed to extract its first planting ridge width and first planting hole width. The first planting ridge width refers to the width of the planting row, while the first planting hole width is the spacing between planting holes. Based on the matched first planting ridge width and first planting hole width, a geometric algorithm was used to calculate the planting point of the first leafy vegetable, representing its precise coordinate position. During the calculation, the spacing between planting rows and the spacing between holes were used to deduce the horizontal and vertical position of the hole within the entire ridge. Using a two-dimensional coordinate geometric algorithm, the position of the leafy vegetable within the entire planting area was analyzed and converted into specific X and Y coordinates.

[0018] After determining the planting location for the first flush of leafy greens, path planning begins. Specifically, based on the calculated planting point, a path planning algorithm, such as A* or Dijkstra's algorithm, is used to generate an initial harvesting path. Typically, the harvesting path adopts a zigzag pattern to ensure that the harvester can efficiently cover all planting areas.

[0019] Finally, based on the generated initial harvesting path, the harvesting start point of the path is extracted. This start point is the first position where the harvester begins its operation. The coordinates and orientation of this point are recorded, and the coordinates and orientation of the harvesting start point are used as the initial harvester position of the harvester.

[0020] The shape detection module 12 performs dynamic shape detection on the leafy vegetables on the ground through the detection mechanism of the leafy vegetable harvester, and obtains multi-dimensional shape feature information.

[0021] In this embodiment, a reconnaissance mechanism of a topsoil leafy vegetable harvester is used to dynamically reconnoiter the topsoil leafy vegetables. This reconnaissance mechanism consists of a lidar and an RGB-D camera. During reconnaissance, the lidar generates a three-dimensional point cloud map by emitting and receiving laser beams, while the RGB-D camera generates a two-dimensional image with depth perception capabilities by combining depth information with visible light images. The collected data is then processed to extract parameters such as the spatial height, width, and density of the crops, generating multi-dimensional morphological feature information.

[0022] Furthermore, in the system provided in the application embodiment, the topography reconnaissance module 12 is also used for: The reconnaissance device acquires real-time images of the leafy vegetables on the ground; reads the visual saliency strategy, and analyzes the real-time images based on the visual saliency strategy to obtain the multidimensional morphological feature information; wherein, the multidimensional morphological feature information includes the real-time spatial height and real-time spatial width of the leafy vegetables on the ground.

[0023] In this embodiment, real-time images of the leafy vegetables on the ground are first acquired using a reconnaissance device. Specifically, a visible light image and depth information of the leafy vegetables are captured using an RGB-D camera, and a three-dimensional point cloud map of the leafy vegetables is generated by a lidar device that emits and receives laser beams. After acquiring the real-time images, a pre-stored visual saliency strategy is read, and salient features in the real-time images, such as the shape, color contrast, and depth changes of the leafy vegetables, are analyzed according to the visual saliency strategy to quickly identify the most representative parts and obtain multi-dimensional morphological feature information. The multi-dimensional morphological feature information includes the real-time spatial height and real-time spatial width of the leafy vegetables on the ground.

[0024] Furthermore, in the system provided in the application embodiment, the topography reconnaissance module 12 is also used for: The real-time image is divided into grids to obtain a grid set, and the first grid in the grid set is randomly extracted; a first saliency value of the first grid is determined based on the visual saliency strategy; when the first saliency value reaches the saliency threshold, the first grid is labeled with a first leafy vegetable; a real-time saliency map of the real-time image is generated based on the first leafy vegetable label; the real-time saliency map is analyzed to obtain the real-time spatial height and the real-time spatial width of the leafy vegetable on the ground.

[0025] In this embodiment, the real-time image is first divided into grids using an image segmentation algorithm, such as uniform grid partitioning, to obtain a grid set. Next, a random sampling algorithm is used to randomly select a grid from the grid set, referred to as the first grid. Subsequently, based on a visual saliency strategy, the selected first grid is analyzed, and its saliency value is calculated to obtain the first saliency value of the first grid. After calculating the saliency value of the first grid, it is compared with a preset saliency threshold. The saliency threshold is a pre-set threshold value; if the saliency value of the first grid exceeds this threshold, the grid is considered to contain key visual information, and then the grid is labeled as the first leafy vegetable identifier. By repeatedly detecting the saliency values ​​of multiple grids, all labeled grids are combined to generate a real-time saliency map.

[0026] After generating a real-time saliency map, morphological analysis is performed on the salient regions to extract the real-time spatial height and width of the leafy vegetables. During morphological analysis, the Canny edge detection algorithm is first used to identify the edges of the leafy vegetables based on gradient changes in the saliency map. Specifically, the Canny algorithm first smooths the image to reduce noise, then calculates the image's intensity gradient, and finally obtains clear leafy vegetable boundaries through non-maximum suppression and double thresholding. Next, the RANSAC plane fitting algorithm is used to extract the real-time spatial height of the leafy vegetables. The RANSAC algorithm removes noise by filtering points from the 3D point cloud image generated by LiDAR that support the fitting of a plane model, thus fitting the surface plane of the leafy vegetables. By fitting the points at the top and bottom of the leafy vegetables, the vertical distance between the highest and lowest points is obtained, which is the real-time spatial height.

[0027] While calculating the height, a depth map clustering algorithm is used to analyze the depth information in the saliency map, clustering the depth data of the leafy vegetable region to identify the left and right boundaries of the leafy vegetables. Based on these boundary points, the real-time spatial width of the leafy vegetables is calculated using the Euclidean distance formula.

[0028] Finally, the extracted height and width information are integrated to generate complete multidimensional morphological feature information, which includes the real-time spatial height and real-time spatial width of the leafy vegetable.

[0029] Furthermore, in the system provided in the application embodiment, the topography reconnaissance module 12 is also used for: Based on the visual saliency strategy, a first evaluation grid set is constructed, which includes M neighboring grids and N non-neighboring grids, where M and N are both positive integers. It is determined whether the first neighboring grid among the M neighboring grids and the first non-neighboring grid among the N non-neighboring grids meet the grid similarity constraint. If they do not meet the constraint, the first neighboring grid is added to the saliency grid set, and the total number of grids in the saliency grid set is counted and denoted as Q. The ratio of the total number of grids Q to the N non-neighboring grids is taken as the first saliency value.

[0030] In this embodiment, the real-time image is first processed based on a visual saliency strategy. The SLIC superpixel algorithm is used to divide the image into multiple raster regions, forming a first evaluation raster set. This evaluation raster set includes M neighboring raster regions and N non-neighboring raster regions, where M and N are both positive integers. Neighboring raster regions are physically close to each other, while non-neighboring raster regions are raster regions that are farther away.

[0031] Next, by comparing color, texture, and depth features, it is determined whether the first neighboring raster and the first non-neighboring raster meet the raster similarity constraints. Specifically, firstly, the color histograms of the two rasters are calculated, and the Bhattacharyya distance is used to measure color similarity. If the Bhattacharyya distance between the two rasters is greater than 0.2, their colors are considered significantly different and do not meet the similarity constraints. For example, if the Bhattacharyya distance between the color histogram of the first neighboring raster and the first non-neighboring raster is 0.35, which is much higher than the 0.2 threshold, then the color features of these two rasters are significantly different. Then, a Gabor filter is used to analyze the texture features of the two rasters. By analyzing the frequency and orientation of the texture, their similarity is calculated. If the difference in texture orientation between the two rasters is greater than 30 degrees, they are considered to have significant texture differences. For example, if the texture orientation of the first neighboring raster is 60 degrees, while the orientation of the first non-neighboring raster is 95 degrees, this difference is greater than 30 degrees, and therefore does not meet the texture similarity constraints.

[0032] If adjacent and non-adjacent rasters differ significantly in color and texture, they are considered to violate the similarity constraint, and the first neighboring raster is added to the salient raster set. The salient raster set contains all rasters that are visually significantly different. The number of rasters in the salient raster set is then counted, denoted as Q, where Q represents the total number of rasters in the current salient raster set. Finally, the ratio of the total number of rasters Q to the number of N non-adjacent rasters is calculated to obtain the first salient value.

[0033] The control scheme determination module 13 calls the flexible tapping strategy to analyze the multi-dimensional morphological feature information to obtain a flexible tapping control scheme.

[0034] In this embodiment of the application, key parameters of leafy vegetables, such as real-time spatial height, real-time spatial width and leaf density, are first extracted from multi-dimensional morphological feature information.

[0035] Next, a flexible tapping strategy is invoked. This strategy sets multiple feedback coefficients based on different morphological characteristics, including tapping height feedback coefficient, tapping force feedback coefficient, and tapping frequency feedback coefficient. After invoking these feedback coefficients, the specific morphological characteristics of the leafy vegetables are analyzed to determine the flexible tapping control scheme.

[0036] Furthermore, the system provided in the application embodiments also includes: The flexible tapping strategy includes a predetermined tapping height feedback coefficient, a predetermined tapping force feedback coefficient, and a predetermined tapping frequency feedback coefficient; the real-time tapping height is obtained based on the predetermined tapping height feedback coefficient and the real-time spatial height; the real-time tapping force is obtained based on the predetermined tapping force feedback coefficient and the real-time spatial height; the real-time tapping frequency is obtained based on the predetermined tapping force feedback coefficient and the real-time spatial height; the real-time tapping height, the real-time tapping force, and the real-time tapping frequency constitute the flexible tapping control scheme.

[0037] In this embodiment, the flexible tapping strategy includes a predetermined tapping height feedback coefficient, a predetermined tapping force feedback coefficient, and a predetermined tapping frequency feedback coefficient. These feedback coefficients are parameters preset by technical experts based on the multidimensional morphological characteristics of leafy vegetables, and are used to adjust the precision and force of the tapping action.

[0038] When determining the real-time tapping height, a linear proportional calculation is used based on the predetermined tapping height feedback coefficient and the real-time spatial height of the leafy greens. For example, if the real-time height of the leafy greens is 30 cm and the feedback coefficient is 0.8, then the tapping height is calculated to be 24 cm by multiplying the real-time height of the leafy greens by the feedback coefficient.

[0039] After determining the slapping height, the real-time slapping force is calculated based on a predetermined slapping force feedback coefficient and the real-time spatial height of the leafy greens. A linear mapping method is used here, directly adjusting the force according to the height of the leafy greens. For example, if the set force feedback coefficient is 0.5 and the leafy green height is 30 cm, then multiplying the force feedback coefficient by the leafy green height yields a slapping force of 15 Newtons. Similarly, the real-time slapping frequency is calculated using a predetermined slapping frequency feedback coefficient and the real-time spatial height of the leafy greens. A proportional relationship is used, and the slapping frequency is simply calculated based on the leafy green height and the feedback coefficient. For example, if the frequency feedback coefficient is 0.1 and the real-time height of the leafy greens is 30 cm, then multiplying the frequency feedback coefficient by the real-time height of the leafy greens yields a slapping frequency of 3 times per second.

[0040] After calculating the real-time tapping height, real-time tapping force, and real-time tapping frequency, these results are combined to generate a flexible tapping control scheme.

[0041] The beating and combing module 14, through the front combing mechanism of the soil leafy vegetable harvester based on the flexible beating control scheme, beats and combs the soil leafy vegetables to obtain pre-combed soil leafy vegetables.

[0042] In this embodiment, the leafy vegetables are combed and patted using a pre-combing mechanism of a soil-grown leafy vegetable harvester, based on a flexible patting control scheme. Specifically, the pre-combing mechanism, consisting of a robotic arm, a flexible patting device, and related sensors, is activated. The robotic arm precisely adjusts to the patting height and position according to the patting control scheme to ensure the accuracy of the patting action. The flexible patting device is made of flexible material to avoid damage to the leafy vegetables during the patting process. During the patting process, the patting force and frequency are adjusted according to the flexible patting control scheme. Through continuous patting actions, the leaves are gently combed, resulting in smoother and more even leaf arrangement, thus avoiding incomplete harvesting or leaf damage caused by disordered leaves during harvesting. After combing and patting, pre-combed soil-grown leafy vegetables are obtained.

[0043] The supporting harvesting module 15 activates the supporting harvesting mechanism of the leafy vegetable harvester and supports the harvesting of the leafy vegetables on the soil through the supporting harvesting mechanism.

[0044] In this embodiment, the supported harvesting mechanism of the soil-grown leafy vegetable harvester is activated to support and harvest the leafy vegetables that have already been combed. Specifically, the supported harvesting mechanism, consisting of a support component and a harvesting component, is activated first. The support component supports the stems or leaves of the leafy vegetables using a robotic arm or a flexible clamping device. During the support process, a force feedback sensor is used to monitor the applied force in real time, ensuring that the support force is just right, stabilizing the leafy vegetables without damaging the leaves or stems. The flexible clamping device is designed with soft materials to ensure that it provides sufficient support when in contact with the leafy vegetables without damaging the surface structure. Next, the harvesting component is activated to perform the cutting operation. The harvesting component consists of a high-speed rotating blade or a shearing device, responsible for cutting the stems of the leafy vegetables. At this time, since the leafy vegetables have been stably supported by the support component, the harvesting component can accurately align with the stems and cut at the most suitable angle, ensuring a neat harvest with minimal loss. Finally, with the harvest completed, the leafy vegetables are stably cut and collected, completing the supported harvesting of the previously combed soil-grown leafy vegetables.

[0045] Furthermore, the system provided in the application embodiments also includes: The supporting harvesting mechanism includes a supporting component and a harvesting component; the supporting component supports the leafy greens on the pre-combed soil, and the harvesting component performs the cooperative harvesting of the leafy greens on the pre-combed soil.

[0046] In this embodiment, the harvesting support mechanism includes a support component and a harvesting component. The support component supports the leafy vegetables, ensuring they remain stable and do not tilt during harvesting. The support component consists of a flexible gripping device or a robotic arm. During operation, the support component gently grips the stems or leaves of the leafy vegetables using the flexible gripping device, preventing excessive external force from affecting the vegetables during harvesting. The harvesting component is responsible for performing the actual cutting task and consists of a high-speed rotating blade or a shearing device. After the support component securely grips the leafy vegetables, the harvesting component activates the blades or shearing device to cut the stems of the leafy vegetables.

[0047] In summary, the support component gently holds and secures the leafy greens with a flexible clamp, while the harvesting component cuts the greens using high-speed rotating blades or a shearing device. These two components work together to achieve stable and efficient leafy green harvesting.

[0048] Furthermore, the system provided in the application embodiments also includes: The post-sorting mechanism is activated, and the harvested leafy vegetables on the ground are gently oscillated and sorted through the post-sorting mechanism.

[0049] In this embodiment, after the leafy vegetables on the soil are harvested, a post-grooming mechanism is activated. This mechanism includes an oscillating device and a flexible grooming component. The oscillating device generates vibrations, gently shaking off the soil adhering to the roots of the leafy vegetables. The flexible grooming component is made of a flexible material, such as rubber or silicone, ensuring that the parts in contact with the leafy vegetables during vibration are soft and do not damage the roots.

[0050] After the rear combing mechanism is activated, the vibration device begins its preset gentle vibration operation. This device gently shakes the roots of the leafy vegetables using a pre-set vibration frequency and intensity. Typically, the vibration frequency is preset to 2-3 times per second to ensure moderate vibration amplitude and prevent damage to the roots. During this process, the gentle combing component contacts the roots of the leafy vegetables, and the force generated by the vibration shakes off the soil adhering to the roots. This process completes the gentle vibration combing of the leafy vegetables on the soil, improving their neatness and facilitating subsequent bundling and packing.

[0051] In summary, the embodiments of this application have at least the following technical effects: This application sets the initial harvester position of the leafy vegetable harvester based on the multi-dimensional planting characteristics of the leafy vegetables; the reconnaissance mechanism of the leafy vegetable harvester performs dynamic morphological reconnaissance of the leafy vegetables to obtain multi-dimensional morphological characteristic information; a flexible beating strategy is called to analyze the multi-dimensional morphological characteristic information to obtain a flexible beating control scheme; the front combing mechanism of the leafy vegetable harvester performs beating and combing of the leafy vegetables based on the flexible beating control scheme to obtain pre-combed leafy vegetables; the support harvesting mechanism of the leafy vegetable harvester is activated, and the pre-combed leafy vegetables are harvested with support through the support harvesting mechanism. This invention addresses the technical problems of existing technologies for harvesting leafy vegetables on the ground, which cannot flexibly adjust to the actual crop conditions, resulting in crop damage, incomplete harvesting, and low efficiency. By acquiring multi-dimensional planting characteristic information of the leafy vegetables, the initial position of the harvester is dynamically set, and the morphological characteristics of the leafy vegetables are acquired in real time using a reconnaissance mechanism. Subsequently, a flexible beating strategy is invoked to generate a control scheme, in which a pre-positioned combing mechanism gently beats and combs the leafy vegetables. Finally, the supporting harvesting mechanism is activated to ensure that the leafy vegetables are stable and harvested accurately, achieving the technical effect of efficient, complete, and damage-free harvesting of leafy vegetables on the ground.

[0052] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0053] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0054] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An adaptive parameter adjustment system for a leafy vegetable harvester, characterized in that, The adaptive parameter adjustment system of the above-ground leafy vegetable harvester is communicatively connected to the harvester, and the adaptive parameter adjustment system of the above-ground leafy vegetable harvester includes: An initial machine position setting module sets the initial harvesting machine position of the leafy vegetable harvester based on the multi-dimensional planting characteristic information of the leafy vegetables. The morphology reconnaissance module uses the reconnaissance mechanism of the soil-grown leafy vegetable harvester to conduct dynamic morphology reconnaissance of the soil-grown leafy vegetables and obtain multi-dimensional morphology feature information. The control scheme determination module calls the flexible tapping strategy to analyze the multi-dimensional morphological feature information to obtain a flexible tapping control scheme. The beating and combing module, through the front combing mechanism of the soil leafy vegetable harvester, beats and combs the soil leafy vegetables based on the flexible beating control scheme to obtain pre-combed soil leafy vegetables; The supporting harvesting module activates the supporting harvesting mechanism of the leafy vegetable harvester and performs supporting harvesting of the leafy vegetables on the soil through the supporting harvesting mechanism.

2. The adaptive parameter adjustment system of the leafy vegetable harvester according to claim 1, characterized in that, include: The multidimensional planting feature information includes multiple sets of planting data for multi-hole leafy vegetables; Match the first planting data of the first hole leafy vegetable in the multiple sets of planting data, wherein the first hole leafy vegetable is any hole of the multiple hole leafy vegetables, and the first planting data includes the first planting ridge width and the first planting hole width; The first planting point of the first hole leaf vegetable is obtained based on the width of the first planting ridge and the width of the first planting hole. The analysis of the first planting point forms an initial harvesting path, which includes a harvesting starting point; The coordinates and orientation of the harvesting starting point are used as the initial harvester position.

3. The adaptive parameter adjustment system of the leafy vegetable harvester according to claim 1, characterized in that, include: The reconnaissance agency collects real-time images of the leafy vegetables on the ground; Read the visual saliency strategy and analyze the real-time image based on the visual saliency strategy to obtain the multidimensional morphological feature information; The multidimensional morphological feature information includes the real-time spatial height and real-time spatial width of the leafy vegetables on the soil.

4. The adaptive parameter adjustment system of the leafy vegetable harvester according to claim 3, characterized in that, include: The real-time image is divided into grids to obtain a grid set, and the first grid in the grid set is randomly extracted; The first saliency value of the first grid is determined based on the visual saliency strategy; When the first significant value reaches the significant value threshold, the first grid is identified as the first leafy vegetable. A real-time salience map of the real-time image is generated based on the first leafy vegetable identifier; By analyzing the real-time saliency map, the real-time spatial height and the real-time spatial width of the leafy vegetables on the soil are obtained sequentially.

5. The adaptive parameter adjustment system of the leafy vegetable harvester according to claim 4, characterized in that, include: A first evaluation grid set is constructed based on the visual saliency strategy. The first evaluation grid set includes M neighboring grids and N non-neighboring grids, where M and N are both positive integers. Determine whether the first neighboring grid among the M neighboring grids and the first non-neighboring grid among the N non-neighboring grids meet the grid similarity constraint; If it does not meet the requirements, add the first neighboring grid to the salient grid set, and count the total number of grids in the salient grid set, denoted as Q; The ratio of the total number of grid cells Q to the N non-neighboring grid cells is taken as the first significant value.

6. The adaptive parameter adjustment system of the leafy vegetable harvester according to claim 3, characterized in that, include: The flexible tapping strategy includes a predetermined tapping height feedback coefficient, a predetermined tapping force feedback coefficient, and a predetermined tapping frequency feedback coefficient. The real-time striking height is obtained based on the predetermined striking height feedback coefficient and the real-time spatial height. The real-time striking force is obtained based on the predetermined striking force feedback coefficient and the real-time spatial height; The real-time tapping frequency is obtained based on the predetermined tapping force feedback coefficient and the real-time spatial height. The real-time tapping height, the real-time tapping force, and the real-time tapping frequency constitute the flexible tapping control scheme.

7. The adaptive parameter adjustment system for the leafy vegetable harvester according to claim 1, characterized in that, include: The supporting harvesting organization includes support components and harvesting components; The supporting components support the leafy greens on the pre-combed soil, and the harvesting components harvest the leafy greens on the pre-combed soil.

8. The adaptive parameter adjustment system of the leafy vegetable harvester according to claim 7, characterized in that, The post-sorting mechanism is activated, and the harvested leafy vegetables on the ground are gently oscillated and sorted through the post-sorting mechanism.

Images