Planting missing detection and complementary planting seedling planting system and method of vegetable transplanter

Abstract

The invention relates to the technical field of agricultural machinery, in particular to a miss-planting detection and complementary planting seedling system and method for a vegetable transplanter, and the system comprises a region sensing array, an edge calculation decision unit, a miss-planting positioning module, an intelligent complementary planting mechanism, and a farmland digital monitoring platform. Wherein the area sensing array is used for constructing a data acquisition network and acquiring seedling states and soil parameters of a transplanting area in real time; the edge calculation decision-making unit is used for processing transplanting area data, identifying a seedling planting missing state through a target detection algorithm and generating a complementary planting instruction; the missing planting positioning module is used for locking the missing planting position in the vegetable transplanting process; the intelligent complementary planting mechanism is used for driving a mechanical arm to perform complementary planting of seedlings according to the complementary planting instruction; the farmland digital monitoring platform is used for monitoring the farmland environment and the crop growth state in real time and displaying the missing planting distribution, the complementary planting progress and the equipment operation state. Therefore, the problems of poor detection signal, single detection algorithm, low seedling survival rate and the like in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery, and particularly to a missing-plant detection and seedling replanting system and method for a vegetable transplanter. Background Art

[0002] In the context of the accelerating global agricultural modernization process, the scale of large-scale vegetable planting continues to expand, and the demand for precise transplanting technology is becoming increasingly urgent. With the annual increase in labor costs and the increasing scarcity of arable land resources, during actual operation, the transplanter must not only ensure high planting efficiency but also maintain excellent planting quality to achieve the efficient use of land resources and maximize economic benefits. In this process, if the problem of missing plants cannot be solved in a timely manner, it will seriously affect the uniformity and integrity of vegetable planting, thereby resulting in yield losses. Therefore, building an intelligent system with precise missing-plant detection and efficient seedling replanting functions, by real-time monitoring the transplanting process, dynamically analyzing the seedling state, and accurately locating the missing-plant points, is of crucial significance for improving the standardization of vegetable planting and reducing the cost of manual replanting.

[0003] However, there are obvious shortcomings in the existing missing-plant detection and seedling replanting technologies for traditional vegetable transplanters. In the existing seedling replanting link, it mostly relies on mechanical contact sensors or single laser beam devices to trigger. Frequent operation of mechanical components is prone to wear, resulting in unstable detection signals. It is not only difficult to adapt to complex and changeable field terrains but also has a high false positive rate for missing-plant detection. In terms of replanting execution, the traditional system lacks the ability to integrate multi-dimensional information and cannot accurately match and dynamically coordinate key data such as the growth characteristics of seedlings, planting depth, and transplanting spacing with the movement of the robotic arm, resulting in problems such as seedling damage and inconsistent replanting depth during the replanting process. In addition, a single detection algorithm cannot accurately identify missing-plant situations in the face of complex scenarios such as overlapping seedlings, soil reflection, or drastic changes in light, resulting in frequent false negatives and missing the best replanting opportunity, ultimately affecting vegetable yield, seedling survival rate, and planting income, and it is difficult to meet the development requirements of modern agricultural precision and intelligence. Summary of the Invention

[0004] This application provides a missing-plant detection and seedling replanting system and method for a vegetable transplanter to solve problems such as poor detection signals, single detection algorithms, and low seedling survival rates in the prior art.

[0005] The first aspect of the embodiments of the present application provides a missing-plant detection and supplementary-planting seedling system for a vegetable transplanter, including: a regional perception array, an edge computing decision-making unit, a missing-plant positioning module, an intelligent supplementary-planting mechanism, and a farm digital monitoring platform. Among them, the regional perception array is used to construct a data acquisition network to obtain the seedling state and soil parameters of the transplanting area in real time and adapt to different lighting environments; the edge computing decision-making unit is used to process the data of the transplanting area, identify the missing-plant state of the seedlings through a target detection algorithm and generate a supplementary-planting instruction; the missing-plant positioning module is used to lock the missing-plant positions during the vegetable transplanting process; the intelligent supplementary-planting mechanism is used to drive the robotic arm to carry out supplementary-planting of seedlings according to the supplementary-planting instruction and adapt to different soil and seedling types; the farm digital monitoring platform is used to monitor the farm environment and the crop growth state in real time, and display the missing-plant distribution, supplementary-planting progress, and equipment operation status.

[0006] Preferably, the regional perception array includes a seedling state detection sensor group, a soil parameter sensor group, and an environment adaptation unit. Among them, the seedling state detection sensor group is used for seedling target detection and collects data on seedling morphology, spacing, and missing-plant areas; the soil parameter sensor group is used to collect soil humidity, temperature, and fertility parameters in real time; the environment adaptation unit is used to cope with environmental interferences such as different lighting and dust and stably obtain data.

[0007] Preferably, the edge computing decision-making unit includes a target detection and missing-plant identification module and a supplementary-planting strategy generation module. Among them, the target detection and missing-plant identification module is used for real-time detection and missing-plant classification of the seedlings; the supplementary-planting strategy generation module is used to generate the supplementary-planting path of the manipulator according to the distribution of the missing-plant positions.

[0008] Preferably, the missing-plant positioning module includes a satellite positioning unit and a coordinate correction module. Among them, the satellite positioning unit obtains the geographical coordinates of the transplanter in the farm in real time through the GPS satellite positioning system; the coordinate correction module is used to perform error correction according to the actual situation to determine the missing-plant positions and the coordinates of the replanting points.

[0009] Preferably, the intelligent supplementary-planting mechanism includes an electro-hydraulic proportional valve group, a soil resistance sensor, and a seedling supply system. Among them, the electro-hydraulic proportional valve group is used to dynamically adjust the supplementary-planting force to adapt to the soil compactness; the soil resistance sensor is used to feedback the plugging and unplugging resistance data in real time; the seedling supply system is used to store spare seedlings, perform automatic counting and warning of missing seedlings, and grab seedlings through the manipulator to carry out supplementary-planting of seedlings.

[0010] Preferably, the farm digital monitoring platform includes a visual real-time monitoring module and a human-computer interaction module. Among them, the visual real-time monitoring module is used to monitor the missing-plant distribution and replanting progress in the transplanting area in real time, and combine with the equipment operation status dashboard to display visual parameters; the human-computer interaction control module is used for operators to perform manual parameter setting, task assignment and process intervention through the touch operation interface.

[0011] The second aspect of the embodiments of the present application provides a method for detecting missing plants and replanting seedlings of a vegetable transplanter, including: obtaining seedling state data and soil parameters; processing the seedling state data and soil parameters, identifying the missing-plant area of the seedlings through the target detection algorithm, calibrating the coordinates of the missing-plant position and the replanting point in real time, and generating a replanting instruction; according to the replanting instruction, driving the robotic arm to grab and replant the seedlings, and based on the real-time feedback of the soil resistance data, combining with the physical characteristic parameters of the seedlings, adjusting the replanting force and the seedling depth through the force control algorithm. At the same time, using the path planning algorithm to generate an operation path map; based on the operation path map, the missing-plant distribution, the replanting progress and the equipment operation status data are displayed in real time through the human-computer control interface.

[0012] Preferably, the formula of the force control algorithm is:

[0013]

[0014] Wherein, is the target replanting force; is the real-time soil resistance feedback value; is the force error value; is the proportionality coefficient; is the integral coefficient; is the differential coefficient; is the replanting force applied by the robotic arm; is the force error is the rate of change of the force error with time t; is the cumulative effect of the force error with time; is the time differential element; is the target planting depth; is the depth correction amount;

[0015] The third aspect of the embodiments of the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the program to implement the method for detecting missing plants and replanting seedlings of a vegetable transplanter as described in the above embodiments.

[0016] The fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor to implement a method for detecting missed planting and replanting seedlings of a vegetable transplanter as described in the above embodiments.

[0017] Therefore, the present application has the following beneficial effects: In the embodiments of the present application, the area perception array stably collects seedling state data and soil parameters under all-weather lighting conditions, and the edge computing decision unit is equipped with an advanced object detection algorithm to accurately identify the missed planting state of seedlings, avoiding misjudgment and missed judgment caused by manual detection or traditional algorithms, and generating replanting instructions in a timely manner; the missed planting positioning module accurately locks the missed planting position, providing a clear target for the intelligent replanting mechanism; the intelligent replanting mechanism is adaptively controlled by the robotic arm based on the replanting instructions, flexibly adjusting the replanting force, depth and angle to adapt to different soil textures and seedling types, reducing seedling damage and improving the replanting survival rate. The farm digital monitoring platform integrates multi-source information such as farmland environment, crop growth, missed planting distribution, replanting progress and equipment operation, and presents it in a visual manner, enabling operators to grasp the overall operation in real time and providing data for planting decisions. It improves the automation and intelligence level of vegetable transplanting operations, reduces the cost of manual inspection and replanting, and reduces the yield loss caused by missed planting. Thus, the problems of poor detection signal, single detection algorithm and low seedling survival rate in the prior art are solved.

[0018] The additional aspects and advantages of the present application will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and / or additional aspects and advantages of the present application will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, where: Figure 1 FIG. is a schematic structural diagram of a system for detecting missed planting and replanting seedlings of a vegetable transplanter according to an embodiment of the present application; Figure 2 FIG. is a schematic diagram of an area perception array according to an embodiment of the present application; Figure 3 FIG. is a schematic diagram of monitoring a tomato planting base according to an embodiment of the present application; Figure 4 FIG. is a schematic diagram of an edge computing decision unit according to an embodiment of the present application; Figure 5 FIG. is a schematic diagram of a missed planting positioning module according to an embodiment of the present application; Figure 6 FIG. is a schematic diagram of an intelligent replanting mechanism according to an embodiment of the present application; Figure 7Schematic diagram of intelligent supplementary planting and seedling operation for vegetables provided according to an embodiment of the present application; Figure 8 Schematic diagram of a farm digital monitoring platform provided according to an embodiment of the present application; Figure 9 Schematic diagram of an operating system for supplementary planting and seedling in a large farm provided according to an embodiment of the present application; Figure 10 Schematic diagram of a missed planting detection and supplementary planting and seedling system for a vegetable transplanter provided according to an embodiment of the present application; Figure 11 Flowchart of a method for missed planting detection and supplementary planting and seedling of a vegetable transplanter provided according to an embodiment of the present application; Figure 12 Schematic diagram of an operating system for vegetable transplanting operation in a modern greenhouse provided according to an embodiment of the present application; Figure 13 Schematic diagram of a method for missed planting detection and supplementary planting and seedling of a vegetable transplanter provided according to an embodiment of the present application; Figure 14 Schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. Detailed implementation manners

[0020] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0021] The missing-plant detection and seedling replanting system of the vegetable transplanter according to the embodiments of the present application will be described below with reference to the accompanying drawings. In view of the problem of low survival rate of seedlings mentioned in the above background art, the present application provides a missing-plant detection and seedling replanting system for a vegetable transplanter. In this system, the area perception array stably collects seedling state data and soil parameters under all-weather lighting conditions. The edge computing decision-making unit is equipped with an advanced target detection algorithm to accurately identify the missing-plant state of seedlings, avoiding misjudgment and missed judgment caused by manual detection or traditional algorithms, and generating replanting instructions in a timely manner; the missing-plant positioning module accurately locks the missing-plant position, providing a clear target for the intelligent replanting mechanism; the intelligent replanting mechanism, based on the replanting instructions, performs adaptive control through the robotic arm, flexibly adjusting the replanting force, depth and angle to adapt to different soil textures and seedling types, reducing seedling damage and improving the survival rate of replanting. The farm digital monitoring platform integrates multi-source information such as the farm environment, crop growth, missing-plant distribution, replanting progress and equipment operation, and presents it in a visual manner, enabling operators to grasp the overall operation situation in real time and providing data for planting decisions. It improves the automation and intelligence level of vegetable transplanting operations, reduces the cost of manual inspection and replanting, and reduces the yield loss caused by missing plants. Thus, the problems of poor detection signal, single detection algorithm and low survival rate of seedlings in the prior art are solved.

[0022] Figure 1 It is a schematic structural diagram of a missing-plant detection and seedling replanting system for a vegetable transplanter provided by an embodiment of the present application.

[0023] An embodiment of the present application provides a missing-plant detection and seedling replanting system for a vegetable transplanter. The system 10 includes: An area perception array 100, an edge computing decision-making unit 200, a missing-plant positioning module 300, an intelligent replanting mechanism 400, and a farm digital monitoring platform 500.

[0024] Among them, the area perception array 100 is used to construct a data acquisition network, obtain the seedling state and soil parameters of the transplanting area in real time, and adapt to different lighting environments; the edge computing decision-making unit 200 is used to process the data of the transplanting area, identify the missing-plant state of seedlings through a target detection algorithm and generate replanting instructions; the missing-plant positioning module 300 is used to lock the missing-plant position during vegetable transplanting; the intelligent replanting mechanism 400 is used to drive the robotic arm to replant seedlings according to the replanting instructions, adapting to different soils and seedling types; the farm digital monitoring platform 500 is used to monitor the farm environment and crop growth status in real time, and display the missing-plant distribution, replanting progress and equipment operation status.

[0025] It can be understood that in the embodiments of the present application, the area perception array stably collects seedling state data and soil parameters under all-weather lighting conditions. The edge computing decision unit is equipped with an advanced object detection algorithm to accurately identify the state of missing seedlings, avoid misjudgment and missed judgment caused by manual detection or traditional algorithms, and generate replanting instructions in a timely manner; the missing seedling positioning module accurately locks the position of the missing seedlings to provide a clear target for the intelligent replanting mechanism; the intelligent replanting mechanism is adaptively controlled by the robotic arm based on the replanting instructions, flexibly adjusts the replanting force, depth and angle, adapts to different soil textures and seedling types, reduces seedling damage, and improves the survival rate of replanting. The farm digital monitoring platform integrates multi-source information such as farmland environment, crop growth, missing seedling distribution, replanting progress and equipment operation, and presents it in a visual manner, enabling operators to grasp the overall situation of the operation in real time and providing data for planting decisions. It improves the automation and intelligence level of vegetable transplanting operations, reduces the costs of manual inspection and replanting, and reduces the yield loss caused by missing seedlings. Thus, the problems of poor detection signal, single detection algorithm and low seedling survival rate in the prior art are solved.

[0026] In the embodiments of the present application, the area perception array 100 further includes: as Figure 2 shown, the seedling state detection sensor group, the soil parameter sensor group, and the environment adaptation unit.

[0027] Among them, the seedling state detection sensor group is used for seedling target detection and collects data on seedling morphology, spacing, and missing seedling areas; the soil parameter sensor group is used to collect soil humidity, temperature, and fertility parameters in real time; the environment adaptation unit is used to cope with environmental interference such as different lighting and dust and stably obtain data.

[0028] It can be understood that in the embodiments of the present application, the seedling state detection sensor group collects seedling morphology and spacing data in real time, accurately locates the missing seedling area, and monitors the seedling state; the soil parameter sensor group tracks key indicators such as soil humidity, temperature, and fertility in real time to provide detailed soil data for replanting seedlings; the environment adaptation unit resists environmental interference such as complex lighting and dust to ensure the stability of the data collection process. It enhances the adaptability in the changing farmland environment and improves the accuracy of missing seedling detection and the success rate of replanting seedlings.

[0029] For example, as Figure 3As shown in the figure, in the tomato planting base, a soil parameter sensor group is deployed to monitor soil conditions and accurately measure soil volume water content. The measurement range covers 0-100%, with an accuracy of ±2%. The monitoring of soil moisture in the unsaturated range is meticulous. When the sensor detects that the soil moisture in a certain area is lower than the lower limit of 40% of the suitable range for tomato growth, the system immediately links the irrigation equipment and automatically increases the irrigation volume in the area to restore the soil moisture to a reasonable range, effectively avoiding the growth of tomatoes due to lack of water. In the transition period between winter and spring, if the soil temperature is 15°C lower than the suitable growth temperature for tomatoes, the system can start the soil heating equipment in time to ensure the normal development of tomatoes. Through the application of the soil parameter sensor group, the tomato planting base has achieved a 30% increase in water resource utilization and an 18% increase in tomato yield, effectively improving the level of refined management during tomato transplanting and growth, and significantly enhancing the economic benefits and sustainability of tomato planting.

[0030] In the embodiment of the present application, the edge computing decision unit 200 includes: Figure 4 As shown, the target detection and missed planting identification module and the replanting strategy generation module.

[0031] Among them, the target detection and missed planting identification module is used to perform real-time detection and missed planting classification of seedlings; the replanting strategy generation module is used to generate the robotic arm replanting path according to the distribution of missed planting positions.

[0032] It can be understood that the embodiment of the present application captures the growth status of seedlings in real time through the target detection and missed planting identification module, quickly and accurately detects seedlings and classifies and identifies the missed planting status, reduces the missed reporting rate and the misjudgment rate, and ensures the accurate capture of missed planting events; the replanting strategy generation module dynamically generates the optimal replanting path based on the distribution of missed planting locations, combined with the farmland terrain and the motion parameters of the robotic arm, avoids repeated operations and path redundancy, improves replanting efficiency, shortens decision response time, reduces manual intervention, and improves the efficient operation capability of the vegetable transplanter in complex farmland environments.

[0033] It should be noted that the target detection and missed planting identification module is used to detect and classify seedlings in real time, and analyze the plant spacing. Calculate the actual Euclidean distance between the centroids of adjacent seedlings , compared with the preset spacing s, if , it is judged as continuous missing strains ( is the allowed standard deviation, is the plane coordinate of the centroid of adjacent seedlings), and combined with DIoU attribution judgment If the DIoU between the bounding box and the preset seedling hole center is less than 0.5, it is marked as missed planting ( is the overlap area ratio between the prediction box and the seedling hole center, is the Euclidean distance, is the center point coordinate of the prediction box, is the center point coordinate of the ground truth box, is the diagonal length of the minimum bounding rectangle, is the distance intersection over union).

[0034] In the embodiment of the present application, the missing planting positioning module 300 includes, as Figure 5 shown, a satellite positioning unit and a coordinate correction module.

[0035] Among them, the satellite positioning unit obtains the geographical coordinates of the transplanter in the farmland in real time through the GPS satellite positioning system; the coordinate correction module is used to perform error correction according to the actual situation to determine the missing planting position and the coordinates of the replanting point.

[0036] It can be understood that in the embodiment of the present application, the satellite positioning unit of the present application accurately obtains the geographical coordinates of the transplanter in the farmland in real time according to the GPS satellite positioning system, providing basic data for the determination of the missing planting position; the coordinate correction module corrects the positioning data according to the actual scenario, further accurately locks the missing planting position and determines the coordinates of the replanting point, dynamically tracks the operation trajectory of the transplanter, eliminates the positioning deviation caused by factors such as geographical environment and signal interference through error correction, ensures the high precision of the missing planting position identification and replanting point planning, avoids the problem of missing planting and missing replanting, and improves the uniformity and operation efficiency of farmland planting.

[0037] Using the feature point deflection calculation method ,( is the change amount of feature point displacement, is the feature point deflection angle, is the distance from the rotation axis to the image center, is the arctangent function) combined with the actual situation to locate the missing planting point, based on the two-dimensional four-parameter model ,( , is the new coordinate value after transformation, , is the original coordinate value, is the translation amount along the X-axis, is the translation amount along the Y-axis, is the scale factor, is the rotation angle, , is the trigonometric function value), to obtain the replanting coordinates and verify the accuracy.

[0038] In the embodiment of the present application, the intelligent replanting mechanism 400 includes: as Figure 6 shown, an electro-hydraulic proportional valve group, a soil resistance sensor, and a seedling supply system.

[0039] Among them, the electro-hydraulic proportional valve group is used to dynamically adjust the replanting force to adapt to the soil compactness; the soil resistance sensor is used to real-time feedback the plugging resistance data; the seedling supply system is used to store spare seedlings, perform automatic counting and warning of missing seedlings, and grab seedlings through a manipulator for replanting seedlings.

[0040] It can be understood that in the embodiment of the present application, the electro-hydraulic proportional valve group dynamically adjusts the replanting force according to the plugging resistance data real-time feedback by the soil resistance sensor, improves the replanting success rate in different soil environments, and effectively avoids problems such as seedling damage or insufficient planting depth caused by soil compactness differences; the seedling supply system, with the functions of automatic counting and warning of missing seedlings, improves the replanting continuity, combines the planned path and the manipulator, improves the replanting seedling efficiency, controls the transplanting positioning error, reduces the manual intervention cost, and improves the survival rate and operation efficiency of replanting seedlings.

[0041] For example, as Figure 7 shown, in the intelligent replanting operation of vegetables, the electro-hydraulic proportional valve group accurately adjusts the hydraulic system according to the data and the characteristics of vegetable varieties. Because the tomato root system is relatively developed and has a deep root, the valve group accurately regulates the pressure of the hydraulic system to 15-20 MPa according to the soil texture and the characteristics of tomato seedlings. In the hard clay area, the pressure value approaches the upper limit to ensure that the special hole-digging device can efficiently cut into the soil layer and dig out a standard planting hole with a depth of 25-30 cm and a diameter of about 15 cm, while avoiding excessive disturbance to the surrounding roots. In the loose sandy loam area, the pressure is appropriately reduced to about 15 MPa to prevent the cave wall from collapsing. In the transplanting link, the electro-hydraulic proportional valve group precisely controls the hydraulic flow rate, so that the transplanting robotic arm runs at a stable and uniform speed of 0.5 m / s, and the repeated positioning accuracy of each transplanting action can reach ±2 mm, accurately implanting the cultivated tomato seedlings into the hole, increasing the tomato replanting efficiency from the original 500 plants per hour to 1200 plants, an increase of 140%; the survival rate of seedlings has been steadily increased from 85% in the past to 96%, effectively reducing the yield loss caused by missing seedlings, and ensuring the stable supply and quality of tomatoes. In addition, due to the reduction of manual input and resource waste, the production cost can be saved by about 400 yuan per mu per year, effectively promoting the intelligent and efficient development of the vegetable planting industry.

[0042] In the embodiment of the present application, the farm digital monitoring platform 500 includes, as Figure 8 shown, a visual real-time monitoring module and a human-computer interaction module.

[0043] Among them, the visual real-time monitoring module is used to real-time monitor the distribution of missed planting and the replanting progress in the transplanting area, and display the visual parameters in combination with the equipment operation status dashboard; the human-computer interaction control module is used for the operator to perform manual parameter setting, task assignment and process intervention through the touch operation interface.

[0044] It is understandable that the visualization real-time monitoring module in the embodiments of the present application presents the missing-planting distribution, replanting progress, and equipment operation status in the transplanting area with intuitive visual parameters, making the operation situation clear and transparent, and facilitating timely grasping of the overall situation; the human-computer interaction module enables operators to flexibly set parameters manually, issue tasks, and intervene in the process through the touch interface for precise control. It improves the monitoring efficiency and intuitiveness, endows manual flexible intervention capabilities, and makes the control of farmland transplanting operations more intelligent and efficient.

[0045] For example, as Figure 9 shown, in a large farm, when the soil resistance sensor monitors that the resistance in the clay area exceeds the threshold, it feeds back the resistance data to the robotic arm. At the same time, it triggers an alarm and gives a double reminder with a red flashing icon and a buzzer on the 10.1-inch anti-glare touch screen. The operator switches to the "deep tillage mode" accordingly; by presetting the seedling parameters on the touch screen (such as tomato seedling height of 10 - 15 cm and stem diameter ≥ 3 mm), the parameter module automatically matches the grasping force (5 - 8 N adaptive adjustment) and planting depth (8 - 12 cm, accuracy ± 2 mm) of the replanting robotic arm, and the interface dynamically displays the pressure curve of the robotic arm to guide the adjustment of the electro-hydraulic proportional valve group parameters; the replanting path can also be planned by gesture sliding, and combined with the AI algorithm to avoid obstacles. Information such as the replanting progress can also be viewed in real time on the farmer's mobile phone through the remote APP. Among them, the parameter matching improves the efficiency by 40% compared with traditional manual adjustment, the survival rate of seedlings in compacted soil increases from 75% to 92%, the time-consuming for path planning is shortened by 60% compared with manual drawing, and the counting error of planted seedlings is < 0.5%.

[0046] It should be noted that the specific formula of the AI algorithm is:

[0047] Among them, is the total path cost of node n; is the actual movement cost from the starting point to node n; is the heuristic estimated cost from node n to the end point; is the heuristic weight coefficient; is the distance field function between node n and the obstacle; is the obstacle avoidance weight coefficient; is the path smoothing penalty coefficient; is the path curvature change rate.

[0048] A missing-plant detection and replanting seedling system for a vegetable transplanter proposed in an embodiment of the present application stably collects seedling state data and soil parameters under all-weather light environments through a regional perception array. The edge computing decision unit is equipped with an advanced object detection algorithm to accurately identify the missing-plant state of seedlings, avoiding misjudgment and missed judgment caused by manual detection or traditional algorithms, and generating replanting instructions in a timely manner. The missing-plant positioning module accurately locks the missing-plant position, providing a clear target for the intelligent replanting mechanism. Based on the replanting instructions, the intelligent replanting mechanism performs adaptive control through a robotic arm, flexibly adjusting the replanting force, depth, and angle to adapt to different soil textures and seedling types, reducing seedling damage and improving the replanting survival rate. The farm digital monitoring platform integrates multi-source information such as the farm environment, crop growth, missing-plant distribution, replanting progress, and equipment operation, presenting it in a visual manner, enabling operators to grasp the overall operation situation in real time and providing data for planting decisions. It improves the automation and intelligence level of vegetable transplanting operations, reduces the cost of manual inspection and replanting, and reduces yield losses caused by missing plants. Thus, it solves problems such as poor detection signals, single detection algorithms, and low seedling survival rates in the prior art.

[0049] The following will elaborate on a missing-plant detection and replanting seedling system for a vegetable transplanter through a specific embodiment, as Figure 10 shown, including: In a modern greenhouse, 10 Omron EE-SX670 contact micro sensors are installed at 20-cm intervals below the transplanting head of the transplanter. The output electrical signal triggered by the sensor being pressed is used to determine whether the seedlings are transplanted normally. If not triggered, it is determined as a missing plant. A Keyence IL-1000 laser distance sensor is additionally installed to assist in missing-plant judgment by monitoring the transplanting spacing in real time with an accuracy of ±0.1 mm. A Decagon EC-5 soil moisture sensor (measurement range 0 - 100%, accuracy ±3%), a 5TM soil temperature sensor (-30°C - 80°C, accuracy ±0.2°C), and a FLEX-4 fertility sensor are fixed at the front end of the transplanter. Soil moisture, temperature, and fertility parameters are collected every 5 seconds. A protective housing is customized for the sensor group, with a built-in temperature and humidity adjustment module to cope with environmental temperature and humidity changes, and an electromagnetic shielding layer is added to resist electromagnetic interference to ensure the stable operation of the sensors.

[0050] An NVIDIA Jetson AGX Xavier development board is used as the core computing device. During actual operation, in the target detection and missing-plant identification stage, the trigger signals of the micro sensors and the spacing data of the distance sensors are analyzed. According to the preset transplanting rules (spacing 20 cm, missing plants are determined when two consecutive sensors are not triggered and the spacing exceeds 25 cm), real-time detection and classification of missing plants are achieved. The path planning algorithm is used to calculate the shortest replanting path of the manipulator with the missing-plant position as the node and the farm path as the edge.

[0051] Equipped with the u-blox NEO-M8T module, it uses GPS, GLONASS, and Galileo multi-satellite systems to achieve centimeter-level positioning, and transmits the geographical coordinates of the transplanter in the farmland in real time through serial communication. Trimble R10 reference stations are installed at the four corners of the farmland, and real-time kinematic positioning is used to correct the coordinates for errors, controlling the accuracy within ±5 cm to accurately determine the coordinates of missed planting positions and replanting points.

[0052] The intelligent replanting mechanism is installed in the middle of the transplanter, and the operating range of the robotic arm covers an area with a width of 2 meters and a depth of 0.5 meters. When the soil resistance sensor (HBMU10M, range 0 - 5000N, resolution 0.1N) detects that the resistance in the clay area reaches 2800N, the electro-hydraulic proportional valve group (Rexroth DBETR-1X / 315G24K31A1M) can increase the insertion pressure of the replanting robotic arm from 1200N to 2000N within 80 ms, with a pressure control accuracy of ±20N. The double-layer chain seedling box of the seedling supply system can accommodate 500 seedlings. The built-in Omron E3Z-LS63 photoelectric sensor monitors in real time with a counting accuracy of ±1 plant, and triggers an alarm when the number of seedlings is less than 30. The robotic arm selects the UR5 collaborative robot with an arm span of 850 mm and a load of 5 kg, equipped with a customized pneumatic gripper, and the grasping success rate for seedlings with a plant height of 10 - 20 cm reaches 99.2%, with a grasping positioning error of ±0.8 mm. In actual operation, the average single-plant replanting time is 7.5 seconds, and the replanting depth is controlled within ±1.5 cm (for example, the standard depth of cucumber seedlings is 12 cm), and 350 - 420 seedlings can be replanted per hour.

[0053] Using a Dell Precision 7865 workstation as the server, a monitoring software platform is built based on Vue.js + ECharts. The missed planting points are marked with red icons on the electronic map, and the replanting path is shown with green lines. The replanting progress is displayed in a line chart, and the operating status parameters of the equipment such as the working duration of the robotic arm and the fluctuation of sensor data are presented in a bar chart. Equipped with a 24-inch touch screen, operators can manually set parameters such as the transplanter speed and replanting force, and issue task commands such as pause and restart to intervene in the replanting process.

[0054] When the vegetable transplanter travels at an operating speed of 5 km / h, the area perception array collects 30 groups of status data and 1 group of soil data per second, and the edge computing decision unit processes them synchronously. After detecting a missed planting, an instruction and a path are generated within 25 seconds. After receiving the information, the missed planting positioning module completes coordinate correction within 10 seconds and transmits it to the intelligent replanting mechanism. According to the instruction, the intelligent replanting mechanism has the soil resistance sensor to feedback data in real time, and the electro-hydraulic proportional valve group dynamically adjusts the force. The seedling supply system cooperates with the robotic arm to complete the grasping, with an average time-consuming of 6 seconds. The robotic arm moves along the planned path and accurately replants the seedlings within 3 seconds, and the soil covering and compaction operation takes 2 seconds. The farm digital monitoring platform records the whole replanting process in real time, generates an operation report every time 10 plants are replanted, and the report generation time < 15 seconds.

[0055] In summary, in the detection level of the embodiment of the present application, through 10 Omron microswitches arranged at an interval of 20 cm and a laser distance sensor with an accuracy of ±0.1 mm, the accurate identification of missed planting of seedlings is realized; the soil parameter sensor group collects data every 5 seconds, providing an environmental basis for replanting, and the protective shell ensures the stable and reliable data collection. In terms of calculation and decision-making, the NVIDIA Jetson AGX Xavier development board quickly processes data, completes the missed planting judgment and path planning within 25 seconds, and ensures the timeliness of response. In the positioning and replanting links, the multi-satellite system combined with the RTK technology realizes high-precision positioning of ±5 cm. The intelligent replanting mechanism can dynamically adjust the insertion and extraction force within 80 ms in the clay area. The UR5 robotic arm has a grasping success rate of 99.2% for seedlings with a height of 10 - 20 cm, and the average time for single-plant replanting is 7.5 seconds. It can replant 350 - 420 plants per hour, combining efficiency and precision. In terms of monitoring and management, the monitoring platform installed on the Dell workstation displays the operation status in real time, generates an operation report every 10 plants, and the operator can intervene in real time. The whole set of systems realizes the full-process automation and precision from detection, decision-making to replanting and monitoring, greatly improving the transplanting efficiency and quality, and reducing the labor cost and missed planting rate.

[0056] Next, a method for detecting missed planting and replanting seedlings of a vegetable transplanter according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0057] As Figure 11 shown, the method for detecting missed planting and replanting seedlings of the vegetable transplanter includes the following steps: In step S101, obtain the seedling status data and soil parameters.

[0058] It can be understood that in the embodiment of the present application, by monitoring the seedling height, leaf morphology growth indicators, pests, diseases, and growth problems can be detected in a timely manner; obtaining soil parameters, such as data on soil acidity, nutrient content, water content, etc., helps to judge the soil fertility and health status, provides data for the seedling replanting plan, improves the crop yield and quality, and reduces the planting cost.

[0059] In step S102, the seedling state data and soil parameters are processed. The target detection algorithm is used to identify the areas where seedlings are missed, and the coordinates of the missed planting positions and replanting points are calibrated in real time to generate a replanting instruction.

[0060] Among them, the target detection algorithm is an algorithm that can automatically identify the category of the target object and locate its position.

[0061] It should be noted that the specific formula of the target detection algorithm is:

[0062] Among them, is the comprehensive confidence level; is the target existence probability; is the intersection over union of the predicted box and the ground truth box.

[0063] It can be understood that in the embodiment of the present application, the target detection algorithm is used to process the seedling state data and soil parameters, which can quickly and accurately identify the areas where seedlings are missed, automatically locate the missed planting positions and generate accurate coordinates of the replanting points, and output the replanting instruction in real time, greatly improving the detection efficiency and accuracy, avoiding the omissions and inefficiencies of manual inspections, providing a scientific basis for timely replanting, ensuring the integrity and uniformity of seedling planting, and helping to improve the overall yield and planting benefits of crops.

[0064] For example, as Figure 12 shown, in the modern greenhouse vegetable transplanting operation, the system uses the target detection algorithm to achieve accurate missed planting identification. The target detection algorithm analyzes the collected seedling data. When it detects that the distance between adjacent tomato seedlings exceeds 40 cm (standard distance ±10%), it is automatically marked as a missed planting area. After actual measurement, the accuracy rate of missed planting detection for common vegetable seedlings such as tomatoes and cucumbers reaches 96.7%. The marked missed planting information is synchronously transmitted to the replanting strategy generation module. Combining the accurate coordinates provided by the missed planting positioning module (error controlled within ±3 cm), the mechanical arm replanting path is quickly planned, and the intelligent replanting mechanism is driven to complete the replanting of a single seedling within 7.5 seconds, realizing an efficient closed-loop from detection to replanting, and greatly improving the automation and accuracy of vegetable transplanting.

[0065] In step S103, according to the replanting instruction, the mechanical arm is driven to grab and replant the seedlings. Based on the real-time feedback of the soil resistance data and combined with the physical characteristic parameters of the seedlings, the replanting force and the depth of the seedlings are adjusted through the force control algorithm. At the same time, the path planning algorithm is used to generate an operation path map.

[0066] Among them, the force control algorithm is an algorithm that can dynamically adjust the control output by real-time sensing of the external force situation, so that the mechanical system can maintain precise force control and compliant interaction when in contact with the environment.

[0067] Specifically, the force control algorithm formula is as follows:

[0068]

[0069] Wherein, is the target replanting force; is the real-time soil resistance feedback value; is the force error value; is the proportional coefficient; is the integral coefficient; is the differential coefficient; is the replanting force applied by the robotic arm; is the force error The rate of change with time t; is the force error ; is the time differential element; is the target planting depth; is the depth correction amount; is the seedling insertion depth.

[0070] It can be understood that based on the soil pressure data real-time transmitted back by the soil resistance sensor and combined with physical characteristic parameters such as the stem diameter of the seedling and the root toughness, the present application embodiment dynamically adjusts the output of the electro-hydraulic proportional valve group, so that the replanting force of the robotic arm accurately adapts to different soil environments, stabilizes the pressure control accuracy within ±20N, and avoids damaging the root system of the seedling due to excessive force or insufficient force resulting in insecure planting. At the same time, the algorithm collaboratively controls the seedling depth to ensure that different crops accurately enter the soil according to the standard depth (±1.5 cm error), shortens the single-plant replanting time, improves the survival rate of the replanted seedlings, improves the accuracy, stability and operation efficiency of the replanting operation, and reduces the secondary loss caused by improper replanting.

[0071] It should be noted that based on the real-time feedback of the soil resistance data and combined with the physical characteristic parameters of the seedlings, the replanting force and the seedling depth are adjusted through the force control algorithm. If it is detected that the soil resistance is large, the algorithm will control the electro-hydraulic proportional valve group to gradually increase the insertion pressure of the robotic arm from the initial 1200N and maintain the pressure fluctuation within ±20N through feedback regulation; in the control of the seedling depth, the algorithm sets the standard depth according to the crop type (such as 12 cm for cucumber seedlings), and combines the data of the displacement sensor of the robotic arm to adjust the descending stroke in real time to ensure that the error is controlled within the range of ±1.5 cm, so as to achieve the precise dynamic adjustment of the replanting force and the seedling depth.

[0072] It should be noted that the path planning algorithm is specifically the sparrow search algorithm, and the formula is as follows:

[0073] Wherein, is the variable value of the \(i\)-th individual in the \(j\)-th dimension at the \(t\)-th time step; is the warning value; is the safety threshold; is the attenuation factor; is a random number following the standard normal distribution; is a matrix of all 1s (dimension \(1\times d\)), where \(d\) is the path dimension; is the maximum number of iterations; is the exponential function; is the variable value of the \(i\)-th individual in the \(j\)-th dimension at the \((t + 1)\)-th time step.

[0074] For example, in the operation of replanting seedlings by a vegetable transplanting machine, the force control algorithm plays a key role. When the robotic arm grabs a seedling for replanting, the soil resistance sensor detects the current soil resistance as 2200 N in real time. Combining with the physical characteristic parameters of the target seedling (cucumber seedling) with a stem diameter of 4 mm and a root length of 18 cm, the force control algorithm immediately activates the control strategy, sends an instruction to the electro-hydraulic proportional valve group, dynamically increases the insertion pressure of the robotic arm from the initially set 1500 N to 1900 N, and keeps the pressure fluctuation within the range of ±15 N. At the same time, according to the parameter of the standard planting depth of cucumber seedlings of 12 cm, the displacement sensor provides real-time feedback on the descending distance of the robotic arm, and the algorithm continuously fine-tunes the descending stroke of the robotic arm to ensure that the seedling depth error is controlled within ±1 cm. During the entire replanting process, the force control algorithm continuously makes dynamic adjustments according to the changes in soil resistance and seedling characteristics, enabling the robotic arm to ensure the smooth planting of seedlings in a complex soil environment and avoid damaging the seedlings due to improper force, ultimately achieving precise and efficient replanting operations, with the success rate of single-plant replanting reaching over 98%.

[0075] In step S104, based on the operation path map, the missing-plant distribution, the progress of replanting seedlings, and the equipment operation status data are real-time displayed through the man-machine control interface.

[0076] Among them, the operation path map is a visual graph generated by means of a path planning algorithm, integrating information such as the coordinates of missing-plant positions, the terrain of the farmland, the distribution of obstacles, and the motion constraints of the robotic arm, and is used to intuitively display the optimal movement route of the robotic arm from the current position to each missing-plant point.

[0077] It can be understood that in the embodiments of the present application, by presenting the heat map of missing-plant distribution in real time, the operator can intuitively grasp the missing-plant situation in the farmland and quickly locate the problem area; the progress bar chart of replanting seedlings is updated at a minute-level frequency, clearly showing the completion degree of the operation and facilitating the evaluation of work efficiency; the visual presentation of the equipment operation status data (such as the pressure of the robotic arm and sensor parameters) enables the operator to detect equipment abnormalities in a timely manner and avoid potential failure risks in advance. This significantly reduces the cost of manual inspection. Moreover, the replanting parameters can be adjusted with one key, and specific tasks can be paused or skipped through the touch interface, realizing precise intervention. It improves the operation efficiency, shortens the equipment fault response time, and enhances the controllability and intelligent level of the vegetable transplanting and replanting operations.

[0078] According to the method for detecting missing plants and replanting seedlings of a vegetable transplanter proposed in the embodiments of the present application, the area perception array stably collects the seedling status data and soil parameters under all-weather lighting conditions. The edge computing decision unit is equipped with an advanced object detection algorithm to accurately identify the missing-plant status of seedlings, avoiding misjudgment and missed judgment caused by manual detection or traditional algorithms, and generating replanting instructions in a timely manner; the missing-plant positioning module accurately locks the missing-plant position, providing a clear target for the intelligent replanting mechanism; based on the replanting instructions, the intelligent replanting mechanism performs adaptive control through the robotic arm, flexibly adjusting the replanting force, depth, and angle to adapt to different soil textures and seedling types, reducing seedling damage and improving the replanting survival rate. The farm digital monitoring platform integrates multi-source information such as the farmland environment, crop growth, missing-plant distribution, replanting progress, and equipment operation, and presents it in a visual manner, enabling the operator to grasp the overall situation of the operation in real time and providing data for planting decisions. It improves the automation and intelligent level of vegetable transplanting operations, reduces the cost of manual inspection and replanting, and reduces the yield loss caused by missing plants. Thus, the problems in the prior art such as poor detection signal, single detection algorithm, and low seedling survival rate are solved.

[0079] Next, a specific embodiment will be used to elaborate on a method for detecting missing plants and replanting seedlings of a vegetable transplanter, as Figure 13 shown, including: Obtain seedling status data and soil parameters: Ten Omron EE-SX670 contact micro sensors are installed at 20-cm intervals below the transplanting head of the vegetable transplanter. When the seedlings are transplanted normally, the sensors are triggered by pressure and output electrical signals. If not triggered, it is initially determined that there is a missed transplant at the corresponding position. At the same time, a Keyence IL-1000 laser distance sensor is added to monitor the transplanting spacing in real time with an accuracy of ±0.1 mm, providing auxiliary data for missed transplant judgment, and thus obtaining seedling status data. At the same time, the Decagon 5TE soil sensor and the Lumex APS Oil sensor collect soil humidity, temperature, and fertility data in real time, and complete a sampling every 60 seconds. The automatic dimming lens of the environmental adaptive unit automatically adjusts within the light range of 0 - 100,000 lux, and the dust-proof cover can resist dust and sand blown by an 8-level wind, ensuring the accuracy and stability of data collection.

[0080] Process data, identify the missed transplant area and generate replanting instructions: Using the NVIDIA Jetson AGX Xavier development board as the core, a dedicated object detection algorithm is developed. The trigger signals of the micro sensors and the spacing data of the laser distance sensor are input into it. Combining preset transplanting rules such as a spacing of 20 cm, the missed transplant area of the seedlings is accurately identified; at the same time, using the u-blox NEO-M8T module on board, centimeter-level positioning is achieved with the help of multi-satellite systems such as GPS, GLONASS, and Galileo. The geographical coordinates of the transplanter are obtained through serial communication, and then the coordinate accuracy is corrected to within ±5 cm through RTK of the Trimble R10 reference station at the four corners of the farmland, and the missed transplant position and the coordinates of the replanting points are calibrated in real time; finally, based on the identified missed transplant area and the calibrated coordinates, a replanting instruction containing information such as the missed transplant position, the coordinates of the replanting points, and the depth of the target seedlings is generated.

[0081] Execute the task of replanting seedlings: After receiving the replanting instruction, the intelligent replanting mechanism starts to grab the seedlings with the robotic arm (UR5 collaborative robot). The soil resistance sensor (HBM U10M, range 0 - 5000 N, resolution 0.1 N) real-time feedbacks soil resistance data. When the resistance in the clay area is detected to reach 2800 N, the force control algorithm, through the PID control strategy, controls the electro-hydraulic proportional valve group to increase the insertion pressure of the robotic arm from 1200 N to 2000 N within 80 ms, and the pressure control accuracy is ±20 N. At the same time, according to the physical characteristic parameters of the seedlings (such as the stem diameter of tomato seedlings is 3 mm and the root length is 15 cm), the force control algorithm adjusts the seedling depth to ensure that the replanting depth error is controlled within ±1.5 cm. In addition, the replanting strategy generation module uses the sparrow algorithm, combined with the distribution of obstacles such as columns and irrigation pipes in the greenhouse, to generate an operation path map within 180 ms, guiding the robotic arm to move along the optimal path. The average single-plant replanting time is 7.5 seconds, and 350 - 420 seedlings can be replanted per hour.

[0082] Real-time Monitoring and Data Display: The operation path map and related data are transmitted to the farm digital monitoring platform deployed on the Alibaba Cloud ECS server (configured with 8 cores and 16GB). Through the 5G network, the data is transmitted to the human-machine control interface with a latency of less than 50ms. The visual real-time monitoring module displays the missing-planting distribution in the form of a heat map in real time, and shows the progress of replanting seedlings in the form of a bar chart, which is updated every 10 seconds. At the same time, the device operation status data (such as robotic arm pressure, sensor parameters, etc.) is presented on the dashboard in real time. The 15.6-inch industrial-grade touch screen supports 10-point touch. Operators can complete parameter adjustments such as replanting depth and grasping force within 5 seconds, and the command response time is <2 seconds, which is convenient for operators to monitor and intervene in the replanting operation in real time.

[0083] In summary, in the data acquisition link of the embodiments of the present application, 10 Omron micro switches with an interval of 20 cm cooperate with a laser distance sensor with an accuracy of ±0.1mm, which can quickly capture the seedling transplanting state. The Decagon 5TE and Lumex APS sensors collect soil data once every 60 seconds, providing a scientific basis for replanting. The environment adaptive unit still ensures stable data in extreme light and dust environments; in terms of data processing, the NVIDIA Jetson AGX Xavier development board is equipped with a dedicated algorithm, combined with satellite positioning and RTK, and the missing-planting area can be identified and coordinate calibrated within 25 seconds to generate accurate replanting instructions; when performing the replanting task, the UR5 robotic arm is combined with the HBMU10M soil resistance sensor, which can accurately adjust the insertion pressure within 80ms in the clay area, and combines the sparrow algorithm to avoid greenhouse obstacles to generate the optimal path. It only takes 7.5 seconds to replant a single plant, and 350-420 seedlings can be processed per hour; in real-time monitoring, the Alibaba Cloud server and the 5G network achieve low-latency data transmission, the heat map and bar chart are updated in real time, and the 15.6-inch touch screen supports quick parameter adjustment, and the command response is <2 seconds. The entire system significantly reduces the missing-planting rate, improves the transplanting efficiency and quality, reduces manual intervention, and reduces the planting cost, providing reliable technical support for modern intelligent agriculture.

[0084] Figure 14 It is a schematic structural diagram of the electronic device provided by the embodiments of the present application. The electronic device may include: A memory 1401, a processor 1402, and a computer program stored on the memory 1401 and executable on the processor 1402.

[0085] When the processor 1402 executes the program, it implements the method for detecting missing plants and replanting seedlings of a vegetable transplanter provided in the above embodiments.

[0086] Furthermore, the electronic device further includes: A communication interface 1403 for communication between the memory 1401 and the processor 1402.

[0087] A memory 1401 for storing a computer program that can run on a processor 1402.

[0088] The memory 1401 may include a high-speed RAM (Random Access Memory) memory and may also include a non-volatile memory, such as at least one disk memory.

[0089] If the memory 1401, the processor 1402, and the communication interface 1403 are implemented independently, the communication interface 1403, the memory 1401, and the processor 1402 can be interconnected via a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 14 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0090] Optionally, in a specific implementation, if the memory 1401, the processor 1402, and the communication interface 1403 are integrated on a single chip, the memory 1401, the processor 1402, and the communication interface 1403 can communicate with each other via an internal interface.

[0091] The processor 1402 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0092] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements a method for detecting missed planting and replanting seedlings of a vegetable transplanter as described above.

[0093] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic expressions of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0094] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" can explicitly or implicitly include at least one of these features. In the description of this application, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0095] Any process or method description shown in a flowchart or described in other ways herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logic function or process. And the scope of the preferred embodiments of this application includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of this application belong.

[0096] It should be understood that each part of this application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following well-known technologies in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0097] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0098] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A missing seedling detection and replanting system for a vegetable transplanter, characterized in that, Including: A regional perception array, an edge computing decision-making unit, a missed-planting positioning module, an intelligent replanting mechanism, and a farm digital monitoring platform. Among them, The regional perception array is used to build a data acquisition network, obtain the seedling status and soil parameters of the transplanting area in real time, and adapt to different lighting environments; The edge computing decision-making unit is used to process the data of the transplanting area, identify the missed-planting status of seedlings through target detection algorithms, and generate replanting instructions; The missed-planting positioning module is used to lock the missed-planting positions during the vegetable transplanting process; The intelligent replanting mechanism is used to drive the robotic arm to replant seedlings according to the replanting instructions, and adapt to different soil and seedling types; The farm digital monitoring platform is used to monitor the farm environment and the growth status of crops in real time, and display the missed-planting distribution, replanting progress, and equipment operation status.

2. The seedling omission detection and replanting system of a vegetable transplanter according to claim 1, characterized in that, The regional perception array includes a seedling status detection sensor group, a soil parameter sensor group, and an environment adaptation unit. Among them, the seedling status detection sensor group is used for seedling target detection, and collects data on seedling morphology, spacing, and missed-planting areas; the soil parameter sensor group is used to collect soil humidity, temperature, and fertility parameters in real time; the environment adaptation unit is used to cope with environmental interference such as different lighting and dust, and stably obtain data.

3. The seedling omission detection and replanting system for a vegetable transplanter according to claim 1, characterized in that, The edge computing decision-making unit includes a target detection and missed-planting identification module and a replanting strategy generation module. Among them, the target detection and missed-planting identification module is used for real-time detection and missed-planting classification of seedlings; the replanting strategy generation module is used to generate the replanting path of the manipulator according to the distribution of missed-planting positions.

4. The missing plant detection and replanting seedling system of a vegetable transplanter according to claim 1, characterized in that The missed-planting positioning module includes a satellite positioning unit and a coordinate correction module. Among them, the satellite positioning unit obtains the geographical coordinates of the transplanter in the farm in real time through the GPS satellite positioning system; the coordinate correction module is used to correct errors according to the actual situation to determine the missed-planting positions and the coordinates of the replanting points.

5. The seedling omission detection and replanting system for a vegetable transplanter according to claim 1, characterized in that, The intelligent replanting mechanism includes an electro-hydraulic proportional valve group, a soil resistance sensor, and a seedling supply system. Among them, the electro-hydraulic proportional valve group is used to dynamically adjust the replanting force to adapt to the soil compactness; the soil resistance sensor is used to feedback the plugging and unplugging resistance data in real time; the seedling supply system is used to store spare seedlings, perform automatic counting and warning of missing seedlings, and grab seedlings through the manipulator for replanting.

6. The missing seedling detection and replanting seedling system of a vegetable transplanter according to claim 1, characterized in that, The farm digital monitoring platform includes a visual real-time monitoring module and a human-computer interaction module. Among them, the visual real-time monitoring module is used to monitor the missed-planting distribution and replanting progress of the transplanting area in real time, and combine with the equipment operation status dashboard to display visual parameters; the human-computer interaction control module is used for operators to perform manual parameter settings, task issuance, and process intervention through the touch operation interface.

7. A method for a missing-plant detection and replanting seedling system applied to the vegetable transplanter according to any one of claims 1-6, characterized in that, Including: Obtain seedling status data and soil parameters; Process the seedling status data and soil parameters, identify the missed-planting areas of seedlings through target detection algorithms, calibrate the missed-planting positions and the coordinates of the replanting points in real time, and generate replanting instructions; According to the replanting instruction, drive the robotic arm to grab and replant the seedlings. Based on the real-time feedback of soil resistance data and combined with the physical characteristic parameters of the seedlings, adjust the replanting force and the depth of the seedlings through a force control algorithm. At the same time, use a path planning algorithm to generate an operation path map; Based on the operation path map, the missing planting distribution, the progress of replanting seedlings, and the equipment operation status data are displayed in real time through a man-machine control interface.

8. A method for detecting missed transplanting and replanting seedlings of a vegetable transplanter according to claim 7, characterized in that, The formula of the force control algorithm is: ； ； Among them, is the target replanting force; is the real-time soil resistance feedback value; is the force error value; is the proportionality coefficient; is the integral coefficient; is the differential coefficient; is the replanting force applied by the robotic arm; is the force error the rate of change with time t; is the force error ; is the time differential element; is the target planting depth; is the depth correction amount; is the seedling insertion depth.

Images