A crop intelligent maintenance method, server and medium

Abstract

A method, server, and medium for intelligent crop maintenance relate to the field of electronic digital data processing technology. The method includes: obtaining the real-time robot posture and real-time sensor position within the target planting area; determining a set of locations accessible to the soil data sensor based on the real-time robot posture, real-time sensor position, and the telescopic length range of the telescopic device; determining the predicted distance between each location in the position set and the corresponding crop root system; determining a collection location based on the predicted distance, the movement route between the soil data sensor and each location, and the soil density distribution in the target planting area; sending the collection location to a follower robot; obtaining soil data collected by the soil data sensor at the collection location; evaluating the growth of crops in the target planting area using a preset evaluation model; and controlling preset intelligent maintenance equipment to perform corresponding maintenance tasks. Implementing the above technical solution improves the accuracy of data collection by the soil data sensor.

Description

Technical Field

[0001] The present application relates to the field of electronic digital data processing, and in particular to a method, server and medium for intelligent crop maintenance. Background Art

[0002] With the acceleration of agricultural modernization, intelligent crop maintenance technology has become a key means of improving agricultural production efficiency and quality. By real-time monitoring of soil, weather, and other environmental data, and precisely controlling irrigation, fertilization, and other maintenance measures, it can effectively achieve cost reduction, efficiency improvement, and green and sustainable development.

[0003] Currently, smart agriculture systems often utilize IoT technology. By deploying pre-set sensors in farmland, they monitor key parameters such as soil temperature, humidity, pH, and nutrient content in real time. These systems then combine big data analysis with machine learning algorithms to optimize crop care decisions. Some systems also incorporate edge computing and cloud computing architectures, enabling local data processing and cloud-based collaborative analysis.

[0004] However, the metabolic activity of soil microorganisms and the burrowing and movement of organisms like earthworms can gradually shift the spatial position of pre-set sensors in the soil. This shift not only causes the data collected by the sensors to deviate from the target monitoring area, distorting data such as soil temperature, humidity, and nutrient concentration, but also interferes with the machine learning model's ability to accurately assess soil environmental trends. Consequently, maintenance decisions such as irrigation and fertilization plans are out of sync with actual needs, potentially leading to water waste, soil compaction, and other issues, and reducing the reliability of smart agriculture systems. Summary of the Invention

[0005] This application provides a crop intelligent maintenance method, server and medium, which can improve the accuracy of soil data sensor data collection.

[0006] In the first aspect, the present application provides a method for intelligent crop maintenance, which is applied to a server of a smart agricultural system. The smart agricultural system also includes a follower robot, a soil data sensor and a communication gateway. The follower robot is arranged in the soil below the crop root system, and the soil data sensor is arranged above the follower robot. The soil data sensor is used to collect soil data below the crop root system. The follower robot is used to control the up and down movement of the built-in telescopic device to adjust the position of the soil data sensor. The server is connected to the soil data sensor and the follower robot respectively through the communication gateway. The method includes: obtaining the real-time robot posture of the follower robot and the real-time sensor position of the soil data sensor in the target planting area, and the follower robot cannot autonomously adjust its own position and posture; based on the real-time robot posture, the real-time sensor position and the telescopic device, the follower robot can automatically adjust its own position and posture. The telescopic length range of the telescopic device is used to determine a set of positions that the soil data sensor can reach; based on the predicted root growth of the crops corresponding to each position in the position set, the predicted distance between each position and the corresponding crop root is determined; based on the predicted distance, the movement route between the soil data sensor and each position, and the soil density distribution in the target planting area, the collection position corresponding to the movement route that does not affect the growth of the crop is determined; the collection position is sent to the follower robot; after the follower robot moves the soil data sensor to the collection position, the soil data collected by the soil data sensor at the collection position is obtained; based on the soil data, the crop leaf image and the growth status video, the growth status of the crops in the target planting area is evaluated by a preset evaluation model; based on the growth status, the preset intelligent maintenance equipment is controlled to perform corresponding maintenance tasks.

[0007] Using this technical solution, the position of the soil data sensor can be dynamically adjusted by a follower robot, enabling the sensor's movement to better capture sensor data as plant roots change seasonally. By acquiring the follower robot's real-time position and posture, the set of accessible locations for the soil data sensor is determined. Collection locations are then selected based on crop root growth predictions, movement paths, and soil density distribution. Soil data is collected at the optimal sensor-accessible location, ensuring that the collected data more closely matches the data collected at the preset locations. This reduces the impact of follower robot position drift (i.e., sensor position drift) on data collection accuracy, improving soil data collection accuracy. Furthermore, by integrating soil data, crop leaf images, and growth status videos, the system assesses crop growth, generates maintenance tasks, and controls intelligent maintenance equipment to execute these tasks. This enables automated crop maintenance and improves the efficiency and quality of agricultural production.

[0008] In combination with some embodiments of the first aspect, in some embodiments, determining a collection position corresponding to a movement route that does not affect crop growth based on the predicted distance, the movement route between the soil data sensor and each location, and the soil density distribution in the target planting area specifically includes: obtaining a soil density distribution in the target planting area, the soil density distribution including a first soil density distribution in the crop root area, a second soil density distribution in the robot movement area, and a third soil density distribution in the biological activity area; determining a target soil density distribution when the soil data sensor arrives at each location based on the movement route between the soil data sensor and each location, the soil density distribution, and the impact of the soil data sensor on a preset density of soil within a preset area when moving; determining a target degree of impact of soil density change on crop growth based on the soil density distribution and the target soil density distribution; obtaining a set of target locations whose target degree of impact is within a preset range; and selecting, based on the predicted distance and the corresponding preset ideal distance of each target location in the target location set, the target location corresponding to the minimum distance difference as the collection position, where the minimum distance difference is the minimum among the distance differences of all target locations, and the distance difference is the difference between the predicted distance and the corresponding ideal distance.

[0009] Using this technical solution, the soil density distribution of the target planting area is obtained. Combined with the sensor's movement route, the target soil density distribution when the sensor arrives at each location is estimated, thereby quantifying the change in soil density before and after the sensor's movement. The impact of the sensor's movement on crop growth is then determined based on the change. A set of target locations with a reasonable degree of impact is selected, and the final collection location is determined based on the difference between the predicted distance and the ideal distance. This ensures that the sensor movement does not have a significant negative impact on the soil environment and crop growth, while also ensuring that the collection location maintains an optimal distance from the crop root system. This ensures that the collected data more accurately reflects the soil environment information required for crop root growth, further improving the accuracy and effectiveness of soil data collection, providing more reliable data support for subsequent maintenance decisions, and ensuring the stability and sustainability of the crop growth environment.

[0010] In combination with some embodiments of the first aspect, in some embodiments, obtaining the soil density distribution of the target planting area specifically includes: dividing the target planting area into a crop root area, a robot movement area, and a biological activity area based on the predicted root growth of each crop in the target planting area, the historical movement trajectories of the follower robot and the soil sensor, and the range of soil biological activity; determining a first soil density distribution in the crop root area based on the predicted root growth; determining the number of movements and the movement position offset of the follower robot within a preset time period based on the historical movement trajectory; predicting the biological activity in the biological activity area based on the number of movements and the movement position offset; determining a second soil density distribution in the robot movement area and a third soil density distribution in the biological activity area based on the biological activity, the historical movement trajectory, and the influence of the soil data sensor on the preset density of the soil within the preset area when it moves.

[0011] Using this technical solution, the target planting area was zoned by comprehensively considering crop root growth, robot movement trajectory, and the range of soil biological activity. This defined the factors influencing soil density distribution in different regions. By determining soil density in the crop root area based on predicted root growth, combining historical robot movement data with predicted biological activity, and considering the impact of sensor movement on soil density, the team was able to more accurately determine the soil density distribution in the robot movement area and biological activity area. This avoided soil density data deviations caused by ambiguous zoning and incomplete consideration of influencing factors, providing more reliable soil environmental data for subsequent determination of optimal collection locations.

[0012] In combination with some embodiments of the first aspect, in some embodiments, after the follower robot moves the soil data sensor to the collection position and obtains the soil data collected by the soil data sensor at the collection position, the method also includes: obtaining a first ideal collection position of the target planting area; when the soil data sensor fails, screening out a second position set from the first position set that the first soil data sensor can reach, whose distance from the first ideal collection position is within a preset distance range, so that the first soil data sensor can reach the target planting area; based on the first moving route of the first soil data sensor, the soil density distribution, and the fourth soil density distribution of the first planting area of ​​the first soil data sensor, determining the degree of influence of the soil density change corresponding to each first moving route on crop growth; when the degree of influence is within the preset influence degree range, based on the new collection position corresponding to the second moving route with the minimum influence degree, controlling the first soil data sensor to reach the new collection position to collect new soil data; based on the new collection position and the new soil data, updating the collection position and the soil data.

[0013] Using the above technical solution, if a soil data sensor fails, repairing or replacing it, because it is located below the crops, will affect the crops growing above it. In this case, the optimal replacement solution is selected by screening other planting areas surrounding the faulty sensor for a first soil data sensor that can reach the target planting area of ​​the faulty sensor and whose arrival position in the target planting area is within a preset range from the ideal collection position in the target planting area (i.e., data collected at this arrival position can reflect soil data within the target planting area). The impact of the first soil data sensor's movement route on the soil environment is evaluated, and the optimal replacement solution is selected. Soil data for the target planting area is collected according to the optimal replacement solution, avoiding damage to crops caused by repairing or replacing the sensor, ensuring the accuracy and continuity of soil data collection, and improving the fault response capabilities and overall operational efficiency of the smart agriculture system.

[0014] In combination with some embodiments of the first aspect, in some embodiments, obtaining the first ideal collection position of the target planting area specifically includes: determining the crop growth difference area and the boundary area of ​​different crop types based on the growth conditions and crop types of each crop in the target planting area; determining the root-dense area based on the predicted root growth conditions of each crop in the target planting area; screening out the target horizontal coordinate position in the growth difference area, the boundary area and the root-dense area; calculating the target vertical coordinate position based on the ideal distance between the crop root corresponding to the target horizontal coordinate position and the sensor and the crop root depth; and obtaining the first ideal collection position based on the target horizontal coordinate position and the target vertical coordinate position.

[0015] Using this technical solution, we identify growth differences, boundaries, and root-dense areas based on crop growth conditions, crop species, and root growth predictions to identify target horizontal coordinate locations. We then calculate the target vertical coordinates corresponding to each target horizontal coordinate location, combining the ideal distance and root depth, to determine the first ideal collection location. This allows the sensor to collect data at this ideal collection location that better reflects soil conditions in key crop growth areas, providing a basis for subsequent crop growth assessments and the development of appropriate maintenance strategies.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of screening out a second position set from the first position set that the first soil data sensor can reach and whose distance from the first ideal collection position is within a preset distance range when the soil data sensor fails, the method further includes: obtaining a third position set from the first position set within the first planting area of ​​the first soil data sensor; calculating the data confidence of each third position based on the second ideal collection position of the first planting area; screening out a fourth position set from the third position set whose data confidence is higher than a preset confidence threshold; and planning a first moving route of the first soil data sensor from the current position to the second position, and then from the second position to the fourth position.

[0017] Using this technical solution, after selecting a set of second locations with a suitable distance from the first ideal collection location, the team further identifies a set of third locations within the planting area where the first sensor resides. Data confidence is calculated based on the second ideal collection location, and a fourth set of locations with high confidence is selected to ensure reliable data collection. Simultaneously, a movement route from "current location - second location - fourth location" is planned, ensuring the emergency replacement of data from a faulty sensor while also meeting soil data collection needs in the area where the first sensor resides.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of updating the collection position and the soil data based on the new collection position and the new soil data, the method further includes: if the first ideal collection position and the collection position are not within a preset deviation range, determining a correction value of each soil parameter in the soil data based on the deviation between the first ideal position and the collection position and the soil density distribution; and updating the parameter value of each soil parameter in the soil data based on the correction value.

[0019] By adopting the above technical solution, by comparing the first ideal collection position with the actual collection position, when the deviation between the two exceeds the preset range, the soil parameter correction value is determined in combination with the soil density distribution, and the soil data parameters are updated, thereby compensating for the soil data error caused by the collection position deviation, and making the soil data more accurately reflect the actual soil conditions of the target planting area.

[0020] In combination with some embodiments of the first aspect, in some embodiments, the smart agriculture system also includes a drone, which is equipped with a camera, which is used to collect images and videos of crops. The drone is used to move the camera to the collection position and perform maintenance tasks. The server is connected to the drone through the communication gateway, and based on the soil data, crop leaf images and growth status videos, the server evaluates the growth of crops in the target planting area through a preset evaluation model, specifically including: obtaining the crop leaf images and growth status videos collected by the camera carried by the drone in the target planting area; inputting the crop leaf images, growth status videos and the soil data into the preset evaluation model to obtain the growth status of crops in the target planting area.

[0021] Using this technical solution, drone-mounted cameras capture crop leaf images and growth status videos, which are then fed into a pre-set assessment model along with precisely collected soil data, enabling in-depth fusion analysis of multi-source data. This approach transcends the limitations of a single data dimension, enabling a comprehensive and multi-dimensional assessment of crop growth from multiple perspectives, including soil environment and crop phenotype, avoiding biased assessments caused by incomplete information. This not only improves the accuracy and reliability of crop growth assessments, but also provides a more scientific and comprehensive basis for decision-making on maintenance tasks, promoting the intelligent and refined development of agricultural production.

[0022] In a second aspect, an embodiment of the present application provides a server comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code comprising computer instructions, the one or more processors calling the computer instructions to enable the server to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] In a third aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions. When the instructions are executed on a server, the server executes the method described in the first aspect and any possible implementation of the first aspect.

[0024] In a fourth aspect, the present application provides a computer program product, which, when running on a server, enables the server to execute the method described in the first aspect and any possible implementation of the first aspect.

[0025] It is understandable that the server provided in the second aspect, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided in this application. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods and will not be repeated here.

[0026] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0027] 1. This application obtains the real-time position and real-time posture of the follower robot, determines the set of positions that the soil data sensor can reach, and selects the collection position in combination with the crop root growth prediction, movement route and soil density distribution. The optimal collection position that the sensor can reach is selected to collect soil data, so that the collected data is more consistent with the data collected at the preset position, reducing the impact of the follower robot position offset (i.e., sensor position offset) on the accuracy of data collection, and improving the accuracy of soil data collection.

[0028] 2. This application selects the optimal alternative by screening other planting areas around the faulty sensor to set up a target planting area that can reach the faulty sensor, and by evaluating the impact of the movement route of the first soil data sensor on the soil environment. Collecting soil data from the target planting area according to the optimal replacement plan avoids damage to crops caused by repairing or replacing sensors, ensures the accuracy and continuity of soil data collection, and improves the fault response capability and overall operational efficiency of the smart agriculture system.

[0029] 3. This application compares the first ideal collection position with the actual collection position. When the deviation between the two exceeds the preset range, the soil parameter correction value is determined based on the soil density distribution, and the soil data parameters are updated to compensate for the soil data error caused by the collection position deviation, so that the soil data can more accurately reflect the actual soil conditions of the target planting area. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a structural diagram of a system architecture applicable to the intelligent crop maintenance method in the embodiment of the present application;

[0031] Figure 2 This is a flow chart of the smart crop maintenance method in the embodiment of the present application;

[0032] Figure 3 This is another flow chart of the smart crop maintenance method in the embodiment of the present application;

[0033] Figure 4 This is another structural diagram of a system architecture applicable to the intelligent crop maintenance method in the embodiment of the present application;

[0034] Figure 5 This is a schematic diagram of an exemplary hardware structure of a server in an embodiment of the present application. DETAILED DESCRIPTION

[0035] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of this application, the singular expressions "a," "an," "said," "above," "the," and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations comprising one or more of the listed items.

[0036] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0037] Figure 1 It is a structural diagram of a system architecture to which the intelligent crop maintenance method in the embodiment of the present application can be applied.

[0038] See also Figure 1 ,The smart agriculture system includes a follower robot, soil data sensor, ,communication gateway, drone, camera and server.

[0039] The server, as the core component of the system, analyzes and processes data collected by soil data sensors and cameras, and transmits control commands to the follower robot and drone via a communication gateway. The communication gateway converts the data collected by the soil data sensors and cameras into a protocol and transmits it to the server. It also converts the server's control commands into a format and transmits them to the follower robot and drone. The follower robot, positioned in the soil beneath the crop root system, controls the upward and downward movement of its built-in telescopic mechanism to adjust the position of the soil data sensor within the soil. The soil data sensor, located above the follower robot, collects soil data beneath the crop root system and transmits this data to the server via the communication gateway. The drone moves its onboard camera to a pre-set collection position to collect data and perform maintenance tasks. The camera, mounted on the drone, collects image and video data of the crops and transmits this data to the server via the communication gateway.

[0040] Through the above system architecture, the smart agriculture system can identify the growth conditions of crops based on the soil data collected by soil data sensors and the image data and video data collected by cameras, and perform corresponding maintenance tasks according to the growth conditions of crops to ensure the healthy growth of crops and realize intelligent and efficient agricultural production.

[0041] Smart agriculture systems often utilize the Internet of Things (IoT) to deploy pre-set sensors in farmland to monitor key parameters such as soil temperature, humidity, pH, and nutrient content in real time. These sensors then combine big data analysis with machine learning algorithms to optimize crop maintenance decisions. However, the metabolic activity of soil microorganisms and the burrowing and movement of organisms like earthworms can gradually shift the spatial position of pre-set sensors in the soil. This shift not only causes the data collected by the sensors to deviate from the target monitoring area, distorting data such as soil temperature, humidity, and nutrient concentration, but also interferes with the machine learning model's ability to accurately assess soil environmental trends. Consequently, maintenance decisions, such as irrigation and fertilization plans, can become out of sync with actual needs, potentially leading to water waste, soil compaction, and other issues, reducing the reliability of smart agriculture systems.

[0042] The intelligent crop maintenance method in the embodiment of the present application is adopted to obtain the real-time position and real-time posture of the follower robot, determine the set of positions that the soil data sensor can reach, and select the collection position in combination with the crop root growth prediction, movement route and soil density distribution. The optimal collection position that the sensor can reach is selected to collect soil data, so that the collected data is more consistent with the data collected at the preset position, reducing the impact of the follower robot position offset (i.e., sensor position offset) on the accuracy of data collection, and improving the accuracy of soil data collection.

[0043] The following combination Figure 2 To illustrate the method of the embodiment of the present application.

[0044] See also Figure 2 , which is a flow chart of the intelligent crop maintenance method in the embodiment of this application.

[0045] S201: Acquire the real-time robot posture of the follower robot and the real-time sensor position of the soil data sensor in the target planting area.

[0046] Among them, the follower robot cannot adjust its own position and posture autonomously.

[0047] Specifically, the server first controls multiple preset ultra-wideband (UWB) positioning base stations to send ultra-wideband wireless signals to the UWB tag module. The UWB tag module is installed on the follow-up robot, adjacent to the soil data sensor. After receiving the radio wave signals emitted by each UWB base station, the UWB tag module determines the distance information between each base station by measuring the round-trip propagation time of the radio waves. Subsequently, the UWB tag module will upload the measured distance information to the server in real time through the wireless communication module. Afterwards, the server will use a three-dimensional spatial positioning algorithm (such as a three-dimensional multilateral positioning algorithm or a least squares method) based on the received distance information between the UWB tag module and the multiple base stations, combined with the known coordinates of the base stations, to calculate the horizontal and vertical coordinates of the UWB tag module in the target planting area (that is, the robot's position on the two-dimensional plane and its depth in the soil), and obtain the current position of the UWB tag module.

[0048] Ultra-wideband (UWB) positioning base stations are pre-installed in the soil of the crop planting area. Their locations are precisely measured and recorded in a server database, forming a known, fixed spatial coordinate system. The UWB base stations are positioned to cover the entire target planting area and enable cross-coverage of spatial positioning signals, ensuring accurate measurement of the follower robot's position.

[0049] For real-time robot posture, obtain the robot's real-time posture angles (such as pitch, roll, and yaw) in space as measured by the robot's built-in inertial measurement unit (IMU). The IMU includes sensors such as a three-axis accelerometer, a three-axis gyroscope, and a magnetometer. The accelerometer measures the robot's acceleration along the three axes (X, Y, and Z). Velocity and displacement are obtained by integrating the acceleration. The gyroscope measures the robot's angular velocity around the three axes. Integrating the angular velocity yields the angular change, which in turn determines the robot's posture. The magnetometer measures the direction of the Earth's magnetic field, providing the robot with an absolute orientation reference and helping to calibrate posture data.

[0050] In some embodiments, the real-time robot posture can also be determined based on the preset position of the UWB tag module when the follower robot is placed vertically and the current position of the UWB tag module. Based on the preset position and the current position, the difference between the coordinate value of the current position in the X, Y, and Z axis directions and the coordinate value corresponding to the preset position is calculated to obtain the offset value of the UWB tag module in each direction. According to the calculated offset values ​​in the X, Y, and Z axis directions, combined with a certain posture calculation model or algorithm, the real-time posture of the follower robot is determined. For example, the robot's tilt angle, rotation angle and other posture information in space can be inferred by calculating the proportional relationship and angle relationship between the offset values. The specific calculation model or algorithm can be selected and optimized according to the actual robot design and application scenario, for example, the posture angle is determined based on mathematical methods such as vector operations and trigonometric functions, such as calculating parameters such as pitch angle, roll angle and yaw angle.

[0051] For the real-time sensor position, first obtain the preset relative position vector of the UWB tag module relative to the soil data sensor (that is, the fixed offset distance on the X, Y, and Z axes), then calculate the rotation matrix based on the pitch, roll, and yaw angles of the robot, rotate the relative position vector, and obtain the relative position vector after considering the posture in the global coordinate system. Finally, add the relative position vector to the current position coordinates of the UWB tag module to obtain the real-time sensor position of the soil data sensor.

[0052] S202: Determine a set of positions that the soil data sensor can reach based on the real-time robot posture, the real-time sensor position, and the telescopic length range of the telescopic device.

[0053] Specifically, a local coordinate system is first established based on the real-time sensor position and the real-time robot posture, with the center point of the follower robot as the reference. In this local coordinate system, the real-time sensor position is used as the origin, and the directions of the coordinate axes are determined based on the real-time robot posture. The Z axis represents the direction of extension and retraction of the telescopic mechanism.

[0054] Next, determine the range of coordinate positions that the soil data sensor can reach in the local coordinate system. Since the telescopic device can only move up and down, its movement primarily affects the coordinates in the Z-axis direction in the local coordinate system. Based on the maximum extended length and minimum retracted length within the telescopic device's telescopic length range, determine the range of coordinate positions that the soil data sensor can reach in the Z-axis direction (i.e., the range from the minimum retracted length minus the current telescopic length of the telescopic rod to the maximum retracted length minus the current telescopic length of the telescopic rod). For the horizontal directions (X and Y axes), since the follower robot itself cannot adjust its position autonomously, the reachable position of the soil data sensor in the horizontal direction is fixed and consistent with the horizontal coordinate position of the origin position in the horizontal direction. By integrating the reachable range in the Z-axis direction and the horizontal coordinate position in the horizontal direction, the reachable coordinate position range of the soil data sensor in the local coordinate system is obtained.

[0055] Next, based on the posture of the follower robot, the set of positions that the soil data sensor can reach is determined in the global coordinate system. Through coordinate system conversion, each coordinate position in the reachable coordinate position range in the local coordinate system is converted to the global coordinate system, and the multiple converted coordinate positions in the global coordinate system are integrated to obtain the set of positions that the soil data sensor can reach. The specific coordinate conversion process requires rotation and translation operations based on the real-time posture of the robot. For example, if the robot has a certain pitch angle and roll angle, the coordinates in the local coordinate system need to be rotated and transformed accordingly during the conversion process; at the same time, the converted coordinates are translated according to the current position of the robot to align them with the global coordinate system. Among them, the global coordinate system is a unified spatial coordinate system established in the entire farmland, which is consistent with the coordinate system of the robot's current position obtained in step S201. The farmland is divided into multiple planting areas.

[0056] S203: Determine a predicted distance between each position and the corresponding crop root system based on the predicted root growth of the crop corresponding to each position in the position set.

[0057] Specifically, a pre-built crop root growth prediction model is first obtained. This model is based on machine learning algorithms, such as neural networks and decision trees. It is trained using a large amount of crop growth data, allowing the model to learn the root growth patterns of different crop types under different growth environments. During training, the model continuously adjusts its parameters based on a large amount of historical data, such as the weights and biases in the neural network and the partitioning rules in the decision tree, to minimize the error between the predicted results and the actual root growth situation. Crop growth data includes crop variety information, planting time, soil characteristics, meteorological data, and historical root growth monitoring data.

[0058] Next, relevant information about each crop within the target planting area is input into the crop root growth prediction model. This information includes crop variety, planting time, current soil properties, meteorological data, and historical root growth data. Based on these inputs, the model uses learned patterns to predict the current and future root growth of the crop, including root topology data. This root topology data includes each root type (e.g., taproot, primary lateral root, secondary lateral root, etc.), its growth path in three-dimensional space (a series of connected node coordinates in the global coordinate system), and its connections to other root segments (parent-child relationships).

[0059] The target planting area is then gridded based on a three-dimensional coordinate system, divided into multiple small spatial grids according to horizontal position division rules. For each grid, the root distribution within that grid is determined. The root growth path of each root in the crop's root topology data within the preset grid range is traversed. The three-dimensional coordinates of the nodes in the growth path are compared with the coordinate range of the grid in the global coordinate system, filtering out root segments that are completely or partially within the grid. For each selected root segment, the server classifies and counts it based on its type field (taproot, first-level lateral root, etc.). If it is a "taproot" type, the taproot counter within the grid is incremented by 1. If it is a lateral root type, the lateral root counter at the corresponding level is incremented based on its hierarchical relationship (e.g., tracing back through the parent-child connection field). Simultaneously, the coordinates of all nodes within the root segment are extracted and sorted in ascending order by Z-axis coordinate value. The minimum value is determined as the starting depth, and the maximum value is determined as the ending depth. During this processing, the server performs special processing on overlapping or intersecting root segments. When multiple root segments share some nodes, they are distinguished by unique identifiers and connection fields to avoid double counting. If a root segment spans multiple grids, the server only counts the nodes within the current grid and generates continuous depth data based on the node connection order. After processing all root segments, the server stores the counted information, including the number of taproots and lateral roots, the starting and ending depths of each root segment, and the growth path within the grid, in the root distribution data for the corresponding grid as structured data.

[0060] Next, for each location in the set, the 3D coordinates of that location are compared with the coordinate ranges of each grid in the global coordinate system to determine the target grid in which that location resides. Next, the ending depths of each root segment within that grid are extracted from the root distribution data table for the target grid. The deepest of all ending depths is selected. The vertical coordinate value (Z-axis) of the location is subtracted from the deepest depth to obtain the predicted distance for that location.

[0061] S204: Determine a collection location corresponding to a movement route that does not affect crop growth based on the predicted distance, the movement route between the soil data sensor and each location, and the soil density distribution in the target planting area.

[0062] Specifically, first, based on the predicted root growth of each crop in the target planting area, the historical movement trajectory of the soil data sensor and the follow-up robot, and the range of soil biological activity, the target planting area is divided into crop root area, robot movement area, and biological activity area.

[0063] To delineate the crop root zone, we obtain all the ending depths from the root distribution data for each grid in the target planting area. For each grid, we filter out the deepest of these ending depths. Then, based on the global coordinate system, we identify the three-dimensional spatial regions above the deepest depth within each grid. These regions are then superimposed and merged to obtain the crop root zone.

[0064] To delineate the robot's movement area, the robot first acquires historical trajectory data (trajectory points defined by the robot's center point) and the soil sensor's second historical trajectory data (trajectory points defined by the sensor's center point) over a preset time period. Based on the mechanical parameters of the robot and soil sensor and a preset influence radius, a cylindrical or hemispherical influence buffer zone is generated, centered around each trajectory point in the trajectory data. Using the spatial overlay analysis capabilities of a geographic information system (GIS), all buffer zones are merged and overlapped to determine the first robot movement area. The portion of the first robot movement area that belongs to the crop root zone is removed to obtain the final robot movement area.

[0065] To delineate biological activity zones, we first extract habit data on various soil organisms, such as earthworms and microorganisms, from a soil bioinformation database. Then, combining real-time environmental data such as humidity, temperature, and pH with meteorological information collected by soil sensors, we use spatial interpolation algorithms like Kriging to generate a continuous spatial distribution layer of soil environmental parameters. The soil organism habit data is matched with the layer to identify potential activity zones. For overlapping activity zones, the boundaries of the activity zones are adjusted based on pre-defined biological symbiosis or competition relationships. GIS spatial analysis capabilities are then used to integrate the processed areas, removing duplicates and merging adjacent areas to determine the primary biological activity zone within the entire target planting area. The portions of the first biological activity zone that belong to the crop root system and the robot movement area are removed to determine the biological activity zone.

[0066] Next, based on the predicted root growth, the primary soil density distribution within the crop root zone is determined. Root distribution data is obtained for each grid in the target planting area. For each grid, the root type and the growth nodes within each subgrid of the grid along the three-dimensional growth path are extracted. A subgrid is a grid divided at a preset depth. Each growth node within each subgrid is classified and counted by root type. Each taproot is counted as a unit; lateral roots at the same level are grouped together and counted based on their hierarchical relationships. Then, based on the volume of the subgrid (determined by the grid's length, width, and the preset depth), the number of different root types per unit volume (i.e., root density) is calculated. Soil densities corresponding to the root density of different root types in each subgrid are selected from a preset soil density table to determine the soil density of each subgrid. The soil densities of each subgrid are then integrated to determine the soil density at different locations within the crop root zone (i.e., primary soil density distribution).

[0067] Third, based on the historical movement trajectory, the number of movements and the movement position offset of the follower robot within a preset time period are determined. A first historical movement trajectory of the follower robot within the preset time period is obtained. The number of trajectory points in the first historical movement trajectory is counted to obtain the number of movements of the follower robot. For each trajectory point, the vertical and horizontal offsets between the trajectory point and the previous trajectory point are calculated to obtain the offset of the follower robot at each movement.

[0068] Fourth, based on the number of movements and movement position offsets, the biological activity in the biological activity area is predicted. First, the movement number and movement position offset data of the follower robot are analyzed. The movement number is divided into multiple time windows according to the time series. The mean number of movements and the standard deviation of the offset values ​​within each time window are calculated to construct the robot's movement activity index and movement stability index. Then, the preset biological activity level corresponding to the movement activity index and movement stability index is obtained to determine the biological activity status of the biological activity area.

[0069] Fifth, based on the biological activity, historical movement trajectory, and the impact of the soil data sensor's movement on the preset density of the soil within the preset area, determine the second soil density distribution in the robot's movement area and the third soil density distribution in the biological activity area.

[0070] For the robot's movement area, the change in soil density caused by sensor or robot movement is calculated for each location within the area by combining the movement of the soil data sensor and the preset soil density influence coefficients during robot movement (e.g., empirical values ​​for the soil density change resulting from a single sensor movement). For each trajectory point passed by the soil data sensor, the first telescopic length of the telescopic mechanism corresponding to that trajectory point and the second telescopic length of the trajectory point immediately preceding that trajectory point are obtained. If the first telescopic length is greater than the second telescopic length, a table of first preset density influence coefficients is obtained. The first preset density influence coefficient corresponding to the distance between that trajectory point and each location in the table of first preset density influence coefficients is added to the soil density stored at each location within the preset range of that trajectory point. If the first telescopic length is less than the second telescopic length, a table of second preset density influence coefficients is obtained. The second preset density influence coefficient corresponding to the distance between that trajectory point and each location in the table of second preset density influence coefficients is subtracted from the soil density stored at each location within the preset range of that trajectory point. For each trajectory point passed by the follower robot, the corresponding offset is obtained. Based on the offset values ​​in the vertical and horizontal directions in the offset condition, a table of first preset range and third preset density influence coefficients and a table of second preset range and fourth preset density influence coefficients corresponding to the offset values ​​are obtained. The third preset density influence coefficient corresponding to the distance between the trajectory point and each position point in the third preset density influence coefficient table is added to the soil density stored at each position point within the first preset range of the trajectory point, and the fourth preset density influence coefficient corresponding to the distance between the trajectory point and each position point in the fourth preset density influence coefficient table is subtracted from the soil density stored at each position point within the first preset range of the trajectory point.

[0071] Furthermore, considering the loosening or compacting effects of biological activity on the soil (for example, earthworm activity can reduce soil density), the soil density at locations experiencing biological activity is modified based on the biological activity level. These locations are determined based on the range of biological activity. Finally, the soil density at each location is integrated to generate a second soil density distribution within the robot's movement area.

[0072] For biologically active areas, the change in soil density due to biological activity is calculated for each sub-area (defined by the activity zones of different organisms) based on the preset influence coefficients of biological activity on soil density. Finally, the density change for each sub-area is combined with the corresponding stored soil density data to generate a third soil density distribution for the biologically active area.

[0073] Sixth, the soil density distribution of the target planting area is obtained by integrating the first soil density distribution of the crop root area, the second soil density distribution of the robot moving area, and the third soil density distribution of the biological activity area.

[0074] Seventh, based on the movement routes between the soil data sensor and each location, the soil density distribution, and the impact of the soil data sensor's movement on the preset soil density within the preset area, the target soil density distribution when the soil data sensor arrives at each location is determined. A predicted distance is obtained for each location. Determine whether any of the predicted distances has a target predicted distance less than zero. If so, remove the locations in the location set that the soil data sensor passed through after reaching the target location. Based on the current location and movement rules of the soil data sensor, a movement route is generated for each location. The movement route contains multiple trajectory points. For each location, target movement trajectory data is first generated based on the movement route and second movement trajectory data. Then, based on the target movement trajectory data, the area is re-divided to obtain a new robot movement area and a new biological activity area. The soil density distribution of the new robot movement area and the new biological activity area is re-determined using the above determination method to obtain a new second soil density distribution and a new third soil density distribution. The first soil density distribution, the new second soil density distribution, and the new third soil density distribution are integrated to obtain the target soil density distribution when the soil data sensor arrives at that location.

[0075] Eighth, based on the soil density distribution and the target soil density distribution, determine the target impact of soil density changes on crop growth. Perform position matching on the soil density distribution of the target planting area and the target soil density distribution, and calculate the soil density change value of each location point. Then, for positions where the density change is not zero, obtain the corresponding collapse soil density change threshold from the table based on the variety and growth stage of the corresponding crop planted above the position. Compare the density change of each grid with the collapse soil density change threshold to determine the number of positions where the density change exceeds the collapse soil density change threshold. Divide this number by the total number of location points in the target planting area to obtain the target impact value.

[0076] Ninth, obtain a set of target locations whose target impacts are within a preset impact range. Filter out a set of target impact values ​​within the preset impact range from all target impact values. Obtain the location corresponding to each target impact in the set of target impact values ​​to obtain a set of target locations.

[0077] Finally, based on the predicted distance of each target location in the target location set and the corresponding preset ideal distance, the target location corresponding to the minimum distance difference is used as the collection location. The minimum distance difference is the smallest distance difference among all target locations, and the distance difference is the difference between the predicted distance and the corresponding ideal distance. For each target location, the root distribution data of the grid where the target location is located and the total root density of different types of roots in the grid are obtained. The total root density is obtained by adding the root number density corresponding to each sub-grid in the grid. Obtain the root type corresponding to the deepest depth in the grid. Find the ideal distance corresponding to the root type and the total root density of different types of roots in the preset distance table, and calculate the difference between the predicted distance and the ideal distance to obtain the distance difference. The target location corresponding to the minimum distance difference is used as the collection location.

[0078] S205: Send the collected position to the follow-up robot.

[0079] Specifically, the telescopic length of the telescopic device corresponding to the acquisition location is obtained. Based on the telescopic length, a control instruction is generated. The control instruction is transmitted to the follower robot via a communication gateway. After receiving the control instruction, the follower robot adjusts the actual telescopic length of the telescopic device to the telescopic length specified in the instruction.

[0080] S206. After the follower robot moves the soil data sensor to the collection position, obtain the soil data collected by the soil data sensor at the collection position.

[0081] After receiving the feedback information of the operation execution completion transmitted by the follow-up robot, the soil data collected by the soil data sensor at the collection position is obtained.

[0082] Among them, soil data includes parameters such as temperature, humidity, pH value and electrical conductivity.

[0083] S207: Based on the soil data, crop leaf images, and growth status videos, the growth status of crops in the target planting area is evaluated using a preset evaluation model.

[0084] Among them, growth status includes current health status, future growth potential and potential growth risks.

[0085] Specifically, for crop leaf images, edge detection and morphological operations are used to segment the leaves and accurately extract the leaf area. Next, features are extracted from the leaf images, such as leaf color (by calculating statistics such as the mean and standard deviation of different color channels), shape (such as the aspect ratio, perimeter, and area), and texture (using methods such as the gray-level co-occurrence matrix to calculate texture parameters).

[0086] For videos showing crop growth patterns, the video is broken down into consecutive image frames. Object detection algorithms, such as the YOLO algorithm, are then used to identify crop plants in the video and track their growth. By analyzing image frames at different time points, growth indicators such as plant height change, leaf count, and tiller number can be calculated.

[0087] Soil data, leaf image features, and growth status video analysis results are input into a preset evaluation model. This evaluation model is based on machine learning or deep learning models, such as neural network models. During the model training phase, the model is trained using a large amount of data including soil data, crop leaf images (labeled with healthy leaves, leaves with different pest and disease symptoms, leaves with nutrient deficiency symptoms, etc.), growth status videos (covering crop growth process records at different growth stages and under different growth conditions), and actual crop growth status labels (current health status grading, such as healthy, mildly abnormal, and severely abnormal; actual yield data and growth rate data for the future period; the types and severity of actual pests and diseases, and growth risk events caused by environmental factors). This allows the model to learn the complex mapping relationship between soil data, leaf image features, growth status information, and the crop's current health status, future growth potential, and potential growth risks.

[0088] To assess the current health of a crop, the model considers input data, soil nutrient levels, leaf health, and crop growth dynamics to assess its current health. For example, if the soil nutrient level is low, leaves show signs of pests and diseases, and growth indicators are increasing slowly, the crop's current health is considered poor.

[0089] To assess future growth potential, the model combines historical data with current growth conditions to create predictions. Based on soil nutrient reserves, crop growth trends, and forecasts of environmental factors, it predicts crop growth rate and yield potential over the next period of time. For example, if soil phosphorus levels are high and current crop growth is good, the model predicts that future crop root and fruit development are likely to be better, indicating greater growth potential.

[0090] The model considers multiple factors to assess potential growth risks. First, it identifies pest and disease symptoms from leaf images, combines them with their transmission patterns and environmental conditions to predict the risk of pest spread. Second, it analyzes abnormalities in soil data, such as abnormal pH levels and excessive heavy metal content, to assess their potential impact on crop growth. Furthermore, it considers external factors such as meteorological data, such as forecasts of impending extreme weather events like heavy rain and drought, to assess their impact on crop growth.

[0091] Ultimately, the model outputs growth status, and the assessment results include the current health status of the crop, future growth potential, and potential growth risks.

[0092] S208. Based on the growth situation, control the preset intelligent maintenance equipment to perform corresponding maintenance tasks.

[0093] Specifically, if crop growth conditions indicate abnormalities (such as a "mild abnormality" or "severe abnormality" health status) or a high potential growth risk (such as a significant risk of pest and disease spread or insufficient soil nutrients), the target crop area is determined to require maintenance. Based on the abnormal parameters corresponding to the current health status and the risk factors corresponding to the potential growth risk, the required maintenance type (such as irrigation, fertilization, or spraying) is determined. If irrigation is the maintenance type, soil moisture data for the target crop area is first retrieved. Combined with the crop's water requirements at the current growth stage, the difference between the actual and ideal soil moisture content is calculated to determine the amount of supplemental water required. Irrigation levels are also adjusted based on weather forecast data. If fertilization is the maintenance type, the type, ratio, and amount of fertilizer required are determined based on the actual nitrogen, phosphorus, and potassium content of nutrients detected in soil testing data and the crop's nutrient requirements at the current growth stage. If spraying is the maintenance type, image recognition and data analysis are used to determine the type and severity of pests and diseases. Then, effective pesticides targeting the pest are screened from the pesticide database. Based on the pesticide's characteristics and instructions, the optimal dilution ratio and spray dosage are calculated. Weather conditions are also considered to select the appropriate spraying time. Based on the maintenance type and the corresponding maintenance parameter data, the corresponding maintenance task is generated.

[0094] Then, a task allocation algorithm matches the most suitable intelligent maintenance device for each task based on the current location and operating status of each maintenance device. Once assigned, the server sends a task instruction to the selected intelligent maintenance device. The instruction includes detailed information such as the operation path planning (using a path optimization algorithm to combine information such as farmland topography and obstacle distribution to generate the shortest and most efficient driving route), execution parameters, and task deadline.

[0095] In embodiments of the present application, a follower robot dynamically adjusts the position of a soil data sensor, enabling dynamic sensor movement to better capture sensor data in sync with seasonal changes in plant roots. By acquiring the follower robot's real-time position and posture, the set of accessible locations for the soil data sensor is determined. Collection locations are then selected based on crop root growth predictions, movement routes, and soil density distribution. Soil data is collected at the optimal sensor-accessible location, ensuring that the collected data more closely matches data collected at the preset locations. This reduces the impact of follower robot position offset (i.e., sensor position offset) on data collection accuracy, thereby improving soil data collection accuracy. Furthermore, by integrating soil data, crop leaf images, and growth status videos to assess crop growth, generate maintenance tasks, and control intelligent maintenance equipment to execute these tasks, the system achieves automated crop maintenance and improves the efficiency and quality of agricultural production.

[0096] The following combination Figure 3 To further illustrate the method of the embodiment of the present application.

[0097] See also Figure 3 , is another flow chart of the intelligent crop maintenance method in the embodiment of this application.

[0098] S301: Acquire the real-time robot posture and real-time sensor position within the target planting area.

[0099] S302: Determine a set of locations that are reachable by the soil data sensor.

[0100] S303: Based on the predicted root growth of the crop corresponding to each position in the position set, determine the predicted distance between each position and the corresponding crop root system.

[0101] S304: Obtain soil density distribution in the target planting area

[0102] S305: Determine the target soil density distribution when the soil data sensor arrives at each location.

[0103] S306: Determine the target impact of soil density change on crop growth based on the soil density distribution and the target soil density distribution.

[0104] S307: Acquire a target position set whose target influence degrees are within a preset influence degree range.

[0105] S308 : Based on the predicted distance of each target position in the target position set and the corresponding preset ideal distance, the target position corresponding to the minimum distance difference is used as the collection position.

[0106] S309: Send the collected position to the follow-up robot.

[0107] S310: After the follower robot moves the soil data sensor to the collection position, obtain soil data collected by the soil data sensor at the collection position.

[0108] Steps S301-S310 and Figure 2 Steps S201 to S206 in the illustrated embodiment are similar, and reference may be made to the description of steps S201 to S206 , which will not be repeated here.

[0109] S311. Determine crop growth difference areas and boundary areas between different crop types based on the growth conditions and crop types of each crop in the target planting area.

[0110] Specifically, the target planting area is first divided into multiple grids according to specific rules. For each grid, crop growth data and crop species information are retrieved. By comparing crop growth data between adjacent grids, the differences in various growth indicators are calculated. If the difference exceeds a preset threshold, these grids are marked as areas of crop growth difference.

[0111] To identify the boundary between different crop types, we traverse all grids and, based on the crop type information within the grid, determine whether the grid contains multiple crop types. If so, we mark the grid as a boundary.

[0112] S312. Determine a root-intensive area based on the predicted root growth of each crop in the target planting area.

[0113] Specifically, the root density of different root types within each subgrid, calculated in step S204, is obtained. The root density of all different root types within each subgrid is summed to obtain the total root density of the different root types within the grid. Then, all grids are traversed, and grids with a total root density greater than a threshold are marked as candidate root-dense areas.

[0114] S313 , screening out target horizontal coordinate positions whose horizontal coordinate positions are in the growth difference area, the boundary area, and the root-dense area.

[0115] Specifically, the horizontal coordinates of each grid in the growth difference region are traversed to determine whether the first horizontal coordinate of a first grid in the boundary region is within a preset range. If so, the second horizontal coordinate of a second grid in the root-dense region is determined to be within a preset range. If the second horizontal coordinate of a second grid in the root-dense region is within a preset range, the second horizontal coordinate is used as the target horizontal coordinate. After the traversal is completed, a target horizontal coordinate set is obtained.

[0116] S314 , calculating the target vertical coordinate position based on the ideal distance between the crop root system and the sensor and the crop root system depth corresponding to the target horizontal coordinate position.

[0117] Specifically, for each target horizontal coordinate location, the root distribution data for the grid where the target horizontal coordinate location is located, as well as the total root density of different root types within the grid, are obtained. The root type corresponding to the deepest depth in the grid is obtained. The ideal distance corresponding to the root type and total root density of different root types is searched in a preset distance table. This ideal distance is added to the deepest depth to obtain the target vertical coordinate location corresponding to the target horizontal coordinate location.

[0118] S315 : Obtain a first ideal acquisition position based on the target horizontal coordinate position and the target vertical coordinate position.

[0119] The target horizontal coordinate position and the corresponding target vertical coordinate position are integrated into three-dimensional coordinates to obtain a first ideal acquisition position.

[0120] S316. When a soil data sensor fails, a second position set is selected from the first position set reachable by the first soil data sensor, the second position set being within a preset distance range from the first ideal collection position.

[0121] Among them, the first soil data sensor can reach the target planting area.

[0122] Specifically, data features are first constructed for each soil parameter in the soil data, such as soil moisture, temperature, pH, nutrient content, etc. For each parameter, the mean deviation and standard deviation of the current data and the historical data of the same period are calculated. If the deviation exceeds the preset deviation value, statistical methods and machine learning models are used to detect anomalies. Unsupervised learning algorithms such as isolation forest and One-Class SVM are used to train the model on historical soil data to learn the normal data distribution. The current soil data is input into the model and the anomaly score is calculated. If the score exceeds the set threshold, a fault is determined to exist. At the same time, taking into account the correlation between soil parameters, dimensionality reduction methods such as principal component analysis (PCA) are used to explore the potential relationship between data. If the anomalies of multiple parameters violate the laws of soil physical and chemical changes at the same time (such as a sudden increase in soil temperature but no corresponding decrease in humidity), a fault is confirmed to exist.

[0123] If there is a fault in the soil data sensor, first obtain a set of soil data sensors that are in normal working condition within the preset range of the soil data sensor. Then, based on the sensor position and robot posture of each soil data sensor in the soil data sensor set and the root distribution corresponding to the first planting area, determine the first position set that each soil data sensor can reach and will not damage the crop root system (the distance between the soil data sensor and the crop root system is greater than the preset value). Next, filter out the first soil data sensor in the soil data sensor set. In the first position set of the first soil data sensor, there is a first position within the target planting area. Afterwards, obtain the first position set of the first soil data sensor, and calculate the distance between each first position and the first ideal collection position. Finally, filter out the second position set in the first position set whose distance to the first ideal collection position is within the preset distance range.

[0124] If the soil data sensor is not faulty, determine whether the first ideal collection position and the collection position are within a preset deviation range. If so, execute step S322; if not, execute step S320.

[0125] S317. Based on the first moving route of the first soil data sensor, the soil density distribution, and the fourth soil density distribution in the first planting area, determine the degree of influence of the soil density change corresponding to each first moving route on crop growth.

[0126] Specifically, first, a third set of locations within the first planting area of ​​the first soil data sensor is obtained from the first set of locations. For each first location, a determination is made as to whether the first location is within the first planting area of ​​the corresponding first soil data sensor. If so, the location is added to the third set of locations.

[0127] Next, based on the second ideal acquisition position of the first planting area, the data confidence level of each third position is calculated. The second ideal acquisition position of the first planting area is obtained. The horizontal and vertical deviation values ​​between each third position and the second ideal acquisition position are calculated. A preset first scoring value corresponding to the horizontal deviation value and a preset second scoring value corresponding to the vertical deviation value are obtained. The data confidence level of each third position is determined through weighted calculation based on the preset horizontal and vertical weights, the first and second scoring values.

[0128] Third, a fourth set of positions is selected from the set of third positions, where the data confidence level exceeds a preset confidence threshold. For each third position, the data confidence level is compared with the preset confidence threshold. If the data confidence level exceeds the preset confidence threshold, the third position is added to the fourth set of positions.

[0129] Fourth, a first movement route is planned for the first soil data sensor, from its current position to a second position, and then from the second position to a fourth position. For each second position, a second movement route is generated from the current position to the second position based on the current position of the first soil data sensor and a preset movement rule. A set of third movement routes is then generated from the second position to each fourth position based on the current position of the first soil data sensor and the preset movement rule. The second movement route is combined with each third movement route in the third movement route set to generate a first movement route set.

[0130] Fifth, the first moving route is divided into a fourth moving route and a fifth moving route, wherein each track point in the fourth moving route is within the target planting area, and each track point in the fifth moving route is within the first planting area.

[0131] Sixth, based on each fourth movement route, the soil density distribution, and the impact of the movement of the soil data sensor on the preset density of the soil within the preset area, a first target soil density distribution is determined when the first soil data sensor arrives at each fourth position. Simultaneously, based on each fifth movement route, the fourth soil density distribution in the first planting area, and the impact of the movement of the soil data sensor on the preset density of the soil within the preset area, a second target soil density distribution is determined when the first soil data sensor arrives at each fourth position.

[0132] Finally, based on the soil density distribution and the first target soil density distribution and the second target soil density distribution, the target impact degree of the soil density change on crop growth is determined.

[0133] The implementation steps for determining the degree of influence in step S317 are the same as Figure 2 The steps for determining the target impact degree in step S204 in the illustrated embodiment are similar, and reference may be made to the description in step S204 , which will not be repeated here.

[0134] S318. When the impact degree is within the preset impact degree range, based on the new collection position corresponding to the second moving route with the minimum impact degree, control the first soil data sensor to reach the new collection position to collect new soil data.

[0135] Specifically, when the impact level is within a preset impact level range, a second position in the second movement route corresponding to the minimum impact level is set as a new collection position. The new collection position is transmitted to a target follower robot corresponding to the second movement route. After the target follower robot moves the first soil data sensor to the new collection position, new soil data collected by the first soil data sensor at the new collection position is acquired.

[0136] S319: Based on the new collection location and the new soil data, update the collection location and the soil data.

[0137] The data contents in the collection location and soil data are replaced with the data contents in the new collection location and new soil data.

[0138] In some embodiments, if, when comparing and analyzing soil data with historical soil data, it is detected that only some of the parameter values ​​are abnormal, only the parameter values ​​of the parameters with the abnormal collection in the soil data are replaced, and the content of the collection location corresponding to the parameter with the abnormal collection is replaced with the new collection location.

[0139] S320: If the first ideal collection position and the collection position are not within a preset deviation range, determine a correction value of each soil parameter in the soil data.

[0140] If the first ideal collection position and the collection position are not within a preset deviation range, a correction value of each soil parameter in the soil data is determined according to the deviation between the first ideal position and the collection position and the soil density distribution.

[0141] Specifically, the straight-line distance between the first ideal acquisition position and the acquisition position, the offset values ​​of the coordinates in each direction and other parameters are calculated first to determine whether they are within a preset deviation range.

[0142] If not, a multidimensional analysis model is constructed to conduct an in-depth coupled analysis of the deviation and soil density distribution. First, based on geographic information system (GIS) data, the first ideal collection location and the collection location are projected onto a three-dimensional terrain model. The spatial characteristics of the deviation are refined by combining terrain slope, altitude, and other information.

[0143] Next, the server retrieves the soil density distribution and extracts soil density data within a certain range around the collection location. This data is stored in a grid format, with each grid corresponding to a specific soil density value and related attributes. Using an interpolation algorithm, the server converts the discrete soil density data into a continuous density field, visualizing the soil density distribution and locating the soil density gradient at the collection location.

[0144] Subsequently, for each soil parameter, a corresponding deviation-density correction function is constructed based on historical experimental data and expert experience. Taking soil moisture as an example, the correction function takes straight-line distance, coordinate offsets in various directions, and soil density gradient changes as input variables and outputs a correction coefficient. When constructing this function, the server uses a machine learning algorithm, trained on a large amount of labeled soil parameter measurement data, to optimize the parameter weights in the function and ensure that the correction function accurately reflects the impact of actual environmental changes on soil parameters.

[0145] When calculating the correction value, the original measured value of each soil parameter is multiplied by the corresponding correction coefficient to obtain a preliminary correction value. To improve correction accuracy, the server also introduces a dynamic compensation mechanism, combining real-time meteorological data (such as wind speed and rainfall) to make secondary adjustments to the preliminary correction result. For example, if rainfall occurs during data collection, the server will compensate for the soil moisture correction value based on the amount and duration of rainfall, thereby avoiding correction errors caused by environmental factors.

[0146] S321. Update the parameter value of each soil parameter in the soil data based on the correction value.

[0147] Add the corresponding correction value to the parameter value of each soil parameter in the soil data.

[0148] S322. Acquire crop leaf images and growth status videos collected by the camera carried by the drone in the target planting area.

[0149] Specifically, control commands are sent to the drone. Upon receiving the commands, the drone first plans a route from its current location to the target planting area using a pre-set path planning algorithm. It then moves along this route to the target planting area. Once the drone reaches the target planting area, control commands are sent to the drone's onboard camera. The camera captures images of crop leaves and video of their growth patterns in the target planting area, and transmits the captured data to a server via a communication gateway.

[0150] S323: Input the crop leaf image, growth status video, and soil data into a preset evaluation model to obtain the growth status of the crops in the target planting area.

[0151] S324. Based on the growth situation, control the preset intelligent maintenance equipment to perform corresponding maintenance tasks.

[0152] Steps S323-S324 and Figure 2 Steps S207 and S208 in the illustrated embodiment are similar, and the descriptions of steps S207 and S208 may be referred to, which will not be repeated here.

[0153] In an embodiment of the present application, when a soil data sensor fails, a first soil data sensor is selected by screening other planting areas around the failed sensor to reach the target planting area of ​​the failed sensor, and whose arrival position in the target planting area is within a preset range from the ideal collection position of the target planting area (i.e., the data collected at the arrival position can reflect the soil data in the target planting area), and evaluating the impact of the movement route of the first soil data sensor on the soil environment, thereby selecting the optimal replacement solution. Soil data for the target planting area is collected according to the optimal replacement solution, avoiding damage to crops caused by repairing or replacing sensors, ensuring the accuracy and continuity of soil data collection, and improving the fault response capability and overall operational efficiency of the smart agricultural system. At the same time, by comparing the first ideal collection position with the actual collection position, when the deviation between the two exceeds the preset range, the soil parameter correction value is determined in combination with the soil density distribution, and the soil data parameters are updated, thereby compensating for the soil data error caused by the collection position deviation and making the soil data more accurately reflect the actual soil conditions of the target planting area.

[0154] The above describes the smart crop maintenance method in the embodiment of the present application. The following describes in detail the data transmission process in the smart crop maintenance method.

[0155] See also Figure 4 , which is another structural diagram of the applicable system architecture of the intelligent crop maintenance method in the embodiment of the present application.

[0156] Figure 4 It includes servers, Beidou ground stations, Beidou satellites, Beidou gateways, soil data sensors and follow-up robots.

[0157] Data collected by soil data sensors is first transmitted to the Beidou gateway. The Beidou gateway acts as a data relay, performing preliminary processing and aggregation of the information collected by the soil data sensors. After receiving the data, the Beidou gateway transmits it to the Beidou satellite. The Beidou satellite, acting as an airborne communication hub, receives data from multiple Beidou gateways and transmits it to the Beidou ground station. The Beidou ground station receives the data from the satellites, further processes and converts it, and transmits the data to the server.

[0158] The above describes the transmission process of data to the server. The transmission process of the server to the follower robot or soil data sensor is the opposite of the above transmission process.

[0159] The server in the embodiment of the present application is described in detail below in conjunction with the above-mentioned intelligent crop maintenance method.

[0160] See also Figure 5 , is a schematic diagram of an exemplary hardware structure of a server in an embodiment of the present application.

[0161] In some embodiments, the server 500 includes a computer device, which can be a terminal device. The computer device includes a processor 501, memory 502, a communication module 503, an input device 504, and an output device 505, all connected via a system bus. The processor 501 of the computer device provides computing and control capabilities. The memory 502 of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operating system and computer programs in the non-volatile storage medium to run. The database is used to store data. The communication module 503 of the computer device is used to transmit collected soil data, crop leaf images, and growth status videos to the server, and to transmit control instructions to follower robots and drones. The input device 504 of the computer device is used to receive collected soil data, crop leaf images, and growth status videos. The output device 505 of the computer device is used to display crop growth status, etc. When executed by the processor 501, the computer program implements the intelligent crop maintenance method of the embodiments of the present application.

[0162] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0163] In some embodiments of the present application, a computer-readable storage medium is provided, including instructions. When the instructions are executed on the server 500, the server 500 can execute the intelligent crop maintenance method in the embodiments of the present application.

[0164] In some embodiments of the present application, a computer program product is further provided. When the computer program product is run on the server 500, the server 500 executes the intelligent crop maintenance method in the embodiments of the present application.

[0165] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0166] As used in the above embodiments, the term “when” may be interpreted to mean “if” or “after” or “in response to determining that” or “in response to detecting that”, depending on the context. Similarly, the phrases “upon determining that” or “if (stated condition or event) is detected” may be interpreted to mean “if determining that” or “in response to determining that” or “upon detecting (stated condition or event)” or “in response to detecting (stated condition or event)”, depending on the context.

[0167] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, hard disk, tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive).

[0168] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for intelligent crop maintenance, characterized in that: A server applied to a smart agricultural system, the smart agricultural system further comprising a follower robot, a soil data sensor, and a communication gateway, the follower robot being disposed in the soil below the crop root system, the soil data sensor being disposed above the follower robot, the soil data sensor being used to collect soil data below the crop root system, the follower robot being used to control a built-in telescopic device to move up and down to adjust the position of the soil data sensor, the server being connected to the soil data sensor and the follower robot respectively via the communication gateway, the method comprising: Acquiring a real-time robot posture of the follower robot and a real-time sensor position of the soil data sensor within a target planting area, wherein the follower robot cannot autonomously adjust its position and posture; Determining a set of positions that the soil data sensor can reach based on the real-time robot posture, the real-time sensor position, and the telescopic length range of the telescopic device; Determining a predicted distance between each location and the corresponding crop root system based on the predicted root growth of the crop corresponding to each location in the location set; Determining, based on the predicted distance, the movement routes between the soil data sensor and each location, and the soil density distribution in the target planting area, a collection location corresponding to a movement route that does not affect crop growth; Sending the collected position to the follower robot; After the follower robot moves the soil data sensor to the collection position, acquiring soil data collected by the soil data sensor at the collection position; Based on the soil data, crop leaf images, and growth status videos, evaluating the growth status of crops in the target planting area using a preset evaluation model; Based on the growth conditions, the preset intelligent maintenance equipment is controlled to perform corresponding maintenance tasks.

2. The method according to claim 1, characterized in that The step of determining a collection position corresponding to a movement route that does not affect crop growth based on the predicted distance, the movement route between the soil data sensor and each position, and the soil density distribution in the target planting area specifically includes: Obtaining soil density distribution of the target planting area, the soil density distribution including a first soil density distribution of the crop root area, a second soil density distribution of the robot movement area, and a third soil density distribution of the biological activity area; Determining a target soil density distribution when the soil data sensor arrives at each location based on a movement route between the soil data sensor and each location, the soil density distribution, and the impact of the movement of the soil data sensor on a preset density of soil within a preset area; determining a target impact degree of soil density change on crop growth based on the soil density distribution and the target soil density distribution; Obtain a set of target locations whose target impact degrees are within a preset impact degree range; Based on the predicted distance of each target position in the target position set and the corresponding preset ideal distance, the target position corresponding to the minimum distance difference is used as the collection position, the minimum distance difference is the smallest among the distance differences of all target positions, and the distance difference is the difference between the predicted distance and the corresponding ideal distance.

3. The method according to claim 2, characterized in that The obtaining of the soil density distribution of the target planting area specifically includes: Based on the predicted root growth of each crop in the target planting area, the historical movement trajectories of the follower robot and the soil data sensor, and the range of soil biological activity, the target planting area is divided into a crop root area, a robot movement area, and a biological activity area; determining a first soil density distribution condition in the crop root area based on the predicted root growth condition; Determining the number of movements and the movement position offset of the follower robot within a preset time period based on the historical movement trajectory; and predicting the biological activity of the biological activity area based on the number of movements and the movement position offset; Based on the biological activity, the historical movement trajectory, and the impact of the soil data sensor on the preset density of the soil within the preset area when it moves, the second soil density distribution in the robot movement area and the third soil density distribution in the biological activity area are determined.

4. The method according to claim 1, wherein After the follower robot moves the soil data sensor to the collection position and acquires the soil data collected by the soil data sensor at the collection position, the method further includes: Acquire a first ideal acquisition position of the target planting area; When the soil data sensor fails, a second set of positions, whose distances from the first ideal collection position are within a preset distance range from the first set of positions reachable by the first soil data sensor, are selected, and the first soil data sensor can reach the target planting area; determining, based on the first movement route of the first soil data sensor, the soil density distribution, and the fourth soil density distribution in the first planting area of ​​the first soil data sensor, a degree of influence of the soil density change corresponding to each first movement route on crop growth; When the impact degree is within a preset impact degree range, based on a new collection position corresponding to a second moving route with a minimum impact degree, controlling the first soil data sensor to arrive at the new collection position to collect new soil data; Based on the new acquisition position and the new soil data, the acquisition position and the soil data are updated.

5. The method according to claim 4, characterized in that The obtaining of the first ideal acquisition position of the target planting area specifically includes: Determining crop growth difference areas and boundary areas of different crop types based on the growth conditions and crop types of each crop in the target planting area; Determining a root-intensive area based on predicted root growth of each crop in the target planting area; Filter out target horizontal coordinate positions whose horizontal coordinate positions are in the growth difference area, the boundary area, and the root-intensive area; Calculating a target vertical coordinate position based on an ideal distance between the crop root system and the sensor and the crop root system depth corresponding to the target horizontal coordinate position; A first ideal acquisition position is obtained based on the target horizontal coordinate position and the target vertical coordinate position.

6. The method according to claim 4, characterized in that After the step of screening out a second set of positions from the first set of positions reachable by the first soil data sensor, the distance from the first ideal collection position being within a preset distance range, when the soil data sensor fails, the method further comprises: Acquire a third position set within the first planting area of ​​the first soil data sensor from the first position set; Calculating data confidence at each third position based on the second ideal collection position of the first planting area; Filtering out a fourth position set from the third position set, the fourth position set having data confidence higher than a preset confidence threshold; A first moving route of the first soil data sensor is planned from the current position to the second position, and then from the second position to the fourth position.

7. The method according to claim 4, characterized in that After the step of updating the acquisition location and the soil data based on the new acquisition location and the new soil data, the method further includes: If the first ideal collection position and the collection position are not within a preset deviation range, a correction value of each soil parameter in the soil data is determined based on the deviation between the first ideal position and the collection position and the soil density distribution; and based on the correction value, the parameter value of each soil parameter in the soil data is updated.

8. The method according to claim 1, characterized in that The smart agriculture system further includes a drone equipped with a camera for collecting images and videos of crops. The drone is used to move the camera to a collection location and perform maintenance tasks. The server is connected to the drone via the communication gateway. The system evaluates the growth of crops in the target planting area based on the soil data, crop leaf images, and growth status videos using a preset evaluation model, specifically including: The crop leaf images and growth status videos collected by the camera carried by the drone in the target planting area are obtained; the crop leaf images, growth status videos and soil data are input into the preset evaluation model to obtain the growth status of crops in the target planting area.

9. A server, characterized in that: include: one or more processors and memory; The memory is coupled to the one or more processors, and is configured to store computer program codes, where the computer program codes include computer instructions. The one or more processors call the computer instructions to enable the server to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed on a server, the server is caused to perform the method according to any one of claims 1 to 7.

Images