A smart fertilizer spreading control system based on BeiDou positioning

Abstract

This invention discloses an intelligent fertilizer spreading control system based on BeiDou positioning, relating to the field of agricultural automation technology. The system includes: establishing a fertilizer spreading simulation model; obtaining data on the relationship between the spreader's opening size and the amount of fertilizer applied through experiments; collecting soil analysis data and remote sensing image data, and preprocessing them; generating a fertilizer prescription map based on the preprocessed soil analysis data and remote sensing image data using a multi-objective optimization model, according to the soil characteristics and crop requirements of the field; acquiring high-precision positioning data through the BeiDou positioning system, fusing the fertilizer prescription map with the high-precision positioning data to complete field spatial calibration; generating the optimal fertilizer spreading path using the Dijkstra algorithm based on the field spatial calibration results and the high-precision positioning data; dynamically adjusting the spreader's opening size by combining the tractor's real-time speed and the fertilizer spreading simulation model, and precisely controlling the amount of fertilizer applied using a control algorithm; and automatically shutting down the spreader by monitoring the tractor's turning angle and the field boundary in real time.

Description

Technical Field

[0001] This invention relates to the field of agricultural automation technology, and in particular to an intelligent fertilizer spreading control system based on BeiDou positioning. Background Technology

[0002] With the rapid growth of the global population, agricultural production is facing increasing pressure. Improving agricultural production efficiency, reducing resource waste, and protecting the environment have become important directions for the development of modern agriculture. Traditional fertilization methods, due to a lack of precise control, suffer from problems such as fertilizer waste and uneven fertilization, which not only affect crop growth but may also have adverse effects on soil and the environment.

[0003] Currently, existing agricultural technologies are developing towards intelligence and automation, with intelligent agricultural machinery and drone technology gradually being applied in field management. For example, automated seeders and precision fertilizer applicators are playing an increasingly important role in farmland operations. However, most of these devices rely on manual control or preset parameters, making it impossible to dynamically adapt to the actual conditions of the field and lacking flexible control methods. In particular, it is difficult to achieve precise control based on speed, terrain, and soil conditions during fertilizer application.

[0004] The BeiDou Navigation Satellite System, China's independently developed global navigation and positioning system, boasts significant advantages such as high precision and wide coverage, and has been gradually and widely applied in agriculture. Intelligent fertilizer application systems incorporating BeiDou positioning technology can effectively improve the automation and intelligence of agricultural machinery, reduce resource waste during fertilization, and improve agricultural production efficiency. Therefore, there is an urgent need to develop an intelligent fertilizer application control system based on BeiDou positioning technology to address the shortcomings of traditional fertilization methods and achieve efficient management of field operations. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] Therefore, this invention provides an intelligent fertilizer spreading control system based on Beidou positioning to solve the problems of being unable to dynamically adapt to the actual conditions of the field, lacking flexible control methods, and especially the difficulty in achieving precise control based on speed, terrain and soil conditions during the fertilizer spreading process.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] This invention provides an intelligent fertilizer application control system based on BeiDou positioning, comprising:

[0009] The system comprises the following modules: a model building module for establishing a fertilization simulation model and obtaining data on the relationship between the spreader's opening size and the amount of fertilizer applied; a data acquisition module for collecting and preprocessing soil analysis data and remote sensing image data; a prescription map generation module for generating a fertilization prescription map based on the soil characteristics and crop requirements of the field, using the preprocessed soil analysis data and remote sensing image data; a spatial calibration module for obtaining high-precision positioning data through the BeiDou Navigation Satellite System and fusing the fertilization prescription map with the high-precision positioning data to complete the field spatial calibration; a path generation module for generating the optimal fertilization path using the Dijkstra algorithm based on the field spatial calibration results and high-precision positioning data; an opening size control module for dynamically adjusting the spreader's opening size by combining the tractor's real-time speed and the fertilization simulation model, and using a control algorithm to precisely control the amount of fertilizer applied; and a status detection module for real-time monitoring of the tractor's turning angle and field boundaries, automatically shutting down the spreader.

[0010] As a preferred embodiment of the intelligent fertilizer spreading control system based on BeiDou positioning described in this invention, the following steps are included: establishing a fertilizer application simulation model and obtaining data on the relationship between the opening size of the fertilizer spreader and the amount of fertilizer applied through experiments.

[0011] An accurate physical model of the fertilizer spreader is established using existing fertilizer spreaders, and the physical model is then imported into EDEM software for simulation.

[0012] Determine the types of fertilizers suitable for the fertilizer spreader, and set the parameters for different types of fertilizers and the opening size of the fertilizer spreader in the simulation software;

[0013] Multiple simulation experiments were conducted under each opening size setting, and finite element simulation analysis was performed by changing the parameters of different fertilizer types each time to obtain comprehensive simulation results.

[0014] Record the total mass of fertilizer applied per unit time for each opening size, and calculate the corresponding fertilizer application rate;

[0015] Repeat the simulation experiment, take the average amount of fertilizer as the final data, and record the corresponding opening size;

[0016] Record the parameters during the simulation experiment, organize the simulation data into tables, use MATLAB to plot the relationship between opening size and fertilizer application, determine the preliminary trend and linear or nonlinear relationship, and select different mathematical models for fitting.

[0017] As a preferred embodiment of the intelligent fertilizer application control system based on BeiDou positioning described in this invention, the process includes: collecting soil analysis data and remote sensing image data, and performing preprocessing, comprising the following steps:

[0018] Soil samples were taken from the selected area and analyzed.

[0019] Use satellite remote sensing technology to acquire high-resolution remote sensing image data;

[0020] For the acquired remote sensing image data, geometric correction, atmospheric correction and radiometric correction were performed using MATLAB software;

[0021] Soil analysis data are cleaned and standardized.

[0022] As a preferred embodiment of the intelligent fertilizer application control system based on BeiDou positioning described in this invention, the following steps are included: Based on the soil characteristics and crop requirements of the field, and using preprocessed soil analysis data and remote sensing image data, a fertilizer prescription map is generated using a multi-objective optimization model.

[0023] Spatial registration was performed between soil analysis data and remote sensing image data. QGIS3 software was used to load the soil analysis data and remote sensing image data and perform spatial analysis to determine the soil characteristics and vegetation index of the field.

[0024] The field was divided into zones using the K-means clustering method, and a soil zoning map of the field was generated based on the clustering results.

[0025] The optimization objectives are determined by the accuracy of fertilization, the cost of fertilization, and the risk of fertilizer loss.

[0026] Set constraints that meet crop requirements, total fertilizer application limits, and fertilizer application limits for high-slope areas;

[0027] Particle swarm optimization algorithm is adopted as the optimization algorithm for the multi-objective optimization model;

[0028] Randomly generate a swarm of particles and assign an initial velocity and position to each particle;

[0029] Calculate the comprehensive objective function value for each particle, and record the individual optimal solution and the global optimal solution;

[0030] Update the particle's position and velocity according to the formula for updating the particle's current velocity and position;

[0031] The particles are repeatedly updated until the maximum number of iterations is reached or the change in the objective function value is less than a set threshold, at which point the optimal solution is output.

[0032] After optimization, the fertilization results of each grid cell are converted into a gridded fertilization prescription map, and the field map is overlaid to output the final fertilization prescription map.

[0033] As a preferred embodiment of the intelligent fertilizer application control system based on BeiDou positioning described in this invention, the following steps are included: acquiring high-precision positioning data through the BeiDou positioning system, fusing the fertilizer application prescription map with the high-precision positioning data, and completing field spatial calibration.

[0034] The BeiDou Navigation Satellite System is used to obtain high-precision coordinates of field boundaries and record the latitude and longitude information of boundary points; the coordinates of the center point of the field and the boundary area are obtained.

[0035] The pixel coordinates of the fertilizer prescription map are converted into geographic coordinates using a coordinate transformation formula.

[0036] The field boundaries and zone coordinates are overlaid onto the field map in the BeiDou system to check the spatial matching degree between the fertilizer prescription map and the actual field.

[0037] As a preferred embodiment of the intelligent fertilizer spreading control system based on BeiDou positioning described in this invention, the optimal fertilizer spreading path is generated using the Dijkstra algorithm based on field spatial calibration results and high-precision positioning data, including the following steps.

[0038] High-precision positioning data in the fields is obtained through the BeiDou positioning system, and the boundaries and feature points of the farmland are recorded.

[0039] Dijkstra's algorithm is used for fertilizer application path planning. High-precision location data is used to provide the start and end points for the fertilizer application path planning algorithm. Each point in the field is regarded as a node in the graph, and the boundary is regarded as an obstacle.

[0040] The fertilizer application path was smoothed using Bezier curves or spline curves, and the smoothed fertilizer application path was verified.

[0041] The system monitors the tractor's position in real time, corrects the path using high-precision positioning, and recalculates the path based on environmental changes.

[0042] As a preferred embodiment of the intelligent fertilizer spreading control system based on BeiDou positioning described in this invention, the following steps are included: dynamically adjusting the opening size of the fertilizer spreader by combining the real-time speed of the tractor and the fertilizer application simulation model, and precisely controlling the fertilizer application amount using a control algorithm.

[0043] A fertilizer spreader opening size controller based on servo motor rotation and linkage system is adopted. The lower computer realizes precise control of the servo motor, and the upper computer receives and processes the sent opening size data.

[0044] The lower-level machine is equipped with environmental sensors to detect environmental parameters during operation and transmit them to the upper-level machine. This ensures that the upper-level machine has sufficient data to accurately predict the opening size of the fertilizer spreader and plan its path.

[0045] The host computer obtains the real-time speed of the tractor through the Beidou positioning system and calculates the target opening size, target angle and speed based on the simulation model;

[0046] By adding absolute encoders to the front wheels and incremental encoders to the rear wheels, the tractor can be precisely controlled to operate as expected after receiving data.

[0047] The tractor's angle and speed are controlled by adjusting the PID control algorithm.

[0048] As a preferred embodiment of the intelligent fertilizer spreading control system based on BeiDou positioning described in this invention, the calculation formula of the PID control algorithm is as follows:

[0049]

[0050] in, To control the output, This represents the error value for the current angle or speed. This is the proportionality coefficient. The integral coefficient is... is the differential coefficient.

[0051] As a preferred embodiment of the intelligent fertilizer spreading control system based on Beidou positioning described in this invention, it further includes a status monitoring module, which is used to monitor the turning angle of the tractor and the boundary of the field in real time and automatically shut down the fertilizer spreader.

[0052] As a preferred embodiment of the intelligent fertilizer spreading control system based on BeiDou positioning described in this invention, the real-time monitoring of the tractor's turning angle and the field boundary, and the automatic shutdown of the fertilizer spreader, includes the following steps.

[0053] The BeiDou positioning system is used to obtain the real-time location and driving status of the tractor;

[0054] Set a turning angle threshold to determine whether the tractor's turning angle exceeds a specific angle;

[0055] When the tractor's turning angle is greater than the turning angle threshold, the fertilizer spreader will pause operation. When the tractor's turning angle is less than or equal to the turning angle and the direction is stable, the fertilizer spreading operation will resume.

[0056] The system monitors the tractor's location in real time, determines whether it is approaching the field boundary, and decides whether to turn off the fertilizer spreader based on the distance.

[0057] The beneficial effects of this invention are as follows: High-precision positioning data of tractors and fields is obtained through the BeiDou Navigation Satellite System. The fertilizer application prescription map is then fused with the positioning data to spatially calibrate the fields, clearly defining the field boundaries and the geographical locations of the fertilization areas. By combining the fertilizer application prescription map with BeiDou positioning data, the fertilization requirement points on the prescription map can be accurately mapped to the actual geographical locations of the fields, ensuring that the fertilizer spreader fertilizes at the correct time and location. Using Dijkstra's algorithm, an optimal path covering all fertilization areas of the field can be planned, avoiding duplicate operations or missed areas, thereby improving operational efficiency. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a block diagram of the intelligent fertilizer spreading control system based on BeiDou positioning in Example 1. Detailed Implementation

[0060] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0061] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0062] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0063] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides an intelligent fertilizer spreading control system based on BeiDou positioning, including the following steps:

[0064] The model building module is used to build a fertilization simulation model and obtain data on the relationship between the opening size of the fertilizer spreader and the amount of fertilizer applied through experiments.

[0065] An accurate physical model of the fertilizer spreader is established using existing fertilizer spreaders, and the physical model is then imported into EDEM software for simulation.

[0066] Determine the types of fertilizers suitable for the fertilizer spreader, and set the parameters for different types of fertilizers and the opening size of the fertilizer spreader in the simulation software;

[0067] Furthermore, based on the actual application of the fertilizer spreader, common granular fertilizers such as urea, diammonium phosphate, and compound fertilizers are preferentially selected, and the physical properties of particle size distribution, density, shape, and flowability are recorded for each fertilizer. Different types of fertilizer particles are created in EDEM, and the above physical property parameters are entered. For example, for urea, the particle size distribution is set to 0.5-2 mm, the density to 1330 kg / m³, the shape to be regular particles, and the angle of repose to be 30°.

[0068] Based on the physical properties of the fertilizer, adjust the particle properties in the simulation (such as friction coefficient, elastic modulus, adhesion force, etc.) to ensure the realism and reliability of the simulation results. For example, the friction coefficient of urea can be set to 0.2, and the elastic modulus to 1 GPa. Consider the different environmental conditions that may be encountered in actual operation (such as humidity and temperature), and set the corresponding environmental parameters in the simulation. For example, humidity can vary between 40% and 80%, and temperature can vary between 10°C and 30°C.

[0069] Based on different types of fertilizers, different data were set for each fertilizer, and multiple sets of experiments were conducted for each fertilizer type. Simulation experiments were carried out step-by-step, sequentially setting the opening size and chassis rotation speed. For example, setting the opening size to K1 and changing the chassis rotation speeds U1, U2, and U3 to conduct different simulation experiments and obtain simulation data; setting the chassis rotation speed U1 and changing different opening sizes K1, K2, and K3 to conduct different simulation experiments and obtain simulation data.

[0070] Furthermore, determine the setting of the fertilizer spreader's opening size (e.g., 20%, 30%, 50%), and prioritize granular fertilizers (e.g., urea, diammonium phosphate, compound fertilizer, etc.); record the physical characteristics of the fertilizer, including particle size distribution (unit: millimeters), density (unit: kg / m³), shape (regular or irregular particles), and flowability; based on the physical characteristics of the fertilizer, ensure that the fertilizer spreader can work normally and spread fertilizer evenly.

[0071] For each opening size setting, conduct multiple fertilizer application tests, record the total mass of fertilizer applied per unit time for each opening size, and calculate the corresponding fertilizer application rate; also record environmental parameters (such as speed, environmental conditions, etc.) during the fertilization process.

[0072] Repeat the simulation experiment, take the average amount of fertilizer as the final data, and record the corresponding opening size; in the actual operation, select different opening sizes and chassis speeds based on the obtained simulation experiment data.

[0073] Record the parameters during the simulation experiment, organize the simulation data into tables, use MATLAB to plot the relationship between opening size and fertilizer application, determine the preliminary trend and linear or nonlinear relationship, and select different mathematical models for fitting.

[0074] Furthermore, if the data exhibits a non-linear relationship, a multinomial regression model is used for fitting; if the data exhibits a linear trend, a linear regression model is used for fitting.

[0075] Multinomial regression:

[0076] Linear regression:

[0077] in, Fertilizer application rate (unit: kg / s). Size of the fertilizer spreader's opening (unit: cm2). , ,…, The regression coefficient is determined by the experimental data. The optimal value is determined based on the simulation experimental data to determine the fitting order. The linear correlation coefficient indicates the degree of influence of opening size on fertilizer application rate. The intercept represents the base fertilizer application rate (usually zero) when the fertilizer spreader is at zero opening.

[0078] It should be noted that simulation experiments were conducted to determine the relationship between the rotational speed of the fertilizer spreader chassis and the size of the spreader's opening plate on the area and mass of fertilizer applied per unit time when applying different types of fertilizer using existing fertilizer spreaders. The rotational speed of the fertilizer spreader chassis is related to the vehicle speed; the faster the vehicle speed, the faster the chassis rotational speed, and vice versa. However, even at a slow speed, there will still be a certain initial velocity. Through simulation, the trend and linear or nonlinear relationship between the chassis rotational speed and the opening plate size, as well as the fertilizer application rate and the area applied, can be accurately determined (vehicle speed is obtained based on the actual transmission ratio and does not require simulation). Different mathematical models were selected for fitting.

[0079] By establishing a precise fertilization simulation model, not only can the working status of the fertilizer spreader be monitored in real time, but it can also be precisely controlled, thereby significantly improving the accuracy of the system's fertilizer application. By predicting the fertilizer application rate under different parameter combinations through the simulation model, agricultural workers can select the optimal fertilizer spreader settings according to actual needs, ensuring that each field receives the appropriate amount of fertilizer. Precise fertilizer application control can avoid over-fertilization, reduce agricultural production costs, and minimize negative environmental impacts, aligning with the requirements of sustainable development.

[0080] The data acquisition module is used to collect soil analysis data and remote sensing image data, and to perform preprocessing.

[0081] Soil samples were taken from the selected area to analyze soil nutrients, pH, moisture, etc. Multiple samplings at different locations are recommended.

[0082] High-resolution images and data of farmland are acquired using satellite remote sensing technology. Before the flight, key parameters, such as flight altitude, shooting angle, and shooting time, are set to obtain optimal image quality. Other factors that may affect fertilization, such as weather conditions and crop growth stage, are also recorded.

[0083] For the acquired remote sensing image data, MATLAB software is used for geometric correction to eliminate geometric distortions caused by factors such as the curvature of the earth and the imaging angle of the sensor; radiometric correction and atmospheric correction are performed to eliminate the influence of environmental factors such as light intensity and cloud cover on the image, ensuring that the image reflects the true features of the ground.

[0084] Soil analysis data are cleaned and standardized to ensure data consistency.

[0085] It should be noted that by analyzing the main nutrients in the soil, the system can determine the specific fertilizer requirements of each area, avoiding excessive or insufficient application of any type of fertilizer. Based on soil data, the system can generate personalized fertilization plans to ensure that each field receives the most suitable fertilizer, avoiding resource waste and environmental pollution.

[0086] The prescription map generation module is used to generate fertilizer prescription maps based on the soil characteristics and crop requirements of the field, using preprocessed soil analysis data and remote sensing image data, and employing a multi-objective optimization model.

[0087] Spatial registration was performed between soil analysis data and remote sensing image data. QGIS3 software was used to load the soil analysis data and remote sensing image data and perform spatial analysis to determine the soil characteristics and vegetation index (NDVI) of the field.

[0088] The Normalized Difference Vegetation Index (NDVI) is calculated using remote sensing imagery. Based on the distribution of NDVI values, the growth status of crops in the field is determined, and regional differences are marked.

[0089] Spatial registration was performed between soil data and remote sensing image data to ensure coordinate consistency. Soil property data and NDVI data were then imported into MATLAB software to form a unified spatial dataset.

[0090] The field was divided into zones using the K-means clustering method. Soil zoning maps of the field were generated based on the clustering results, and each zone was labeled with fertilization requirements (fertilizer amount and fertilizer type).

[0091] The optimization objectives are determined by the accuracy of fertilization, the cost of fertilization, and the risk of fertilizer loss.

[0092] Specifically, the goal of optimizing the model is to comprehensively consider the following three aspects:

[0093] Precision fertilization ): Quantify the deviation between the actual amount of fertilizer applied and the crop's needs, with the goal of minimizing the deviation, ensuring that the amount of fertilizer applied matches the crop's needs, and reducing waste or insufficiency.

[0094] Fertilizer cost ( The goal is to minimize the total cost of fertilization while meeting the needs of the crop.

[0095] Fertilizer loss risk ( The risk of fertilizer runoff is related to slope and amount of fertilizer applied. The goal is to keep the risk of runoff within an acceptable range, control fertilizer runoff in high-slope areas, and reduce the impact on the environment.

[0096] The overall objective function is:

[0097]

[0098] in, For the comprehensive objective function, , , The target weighting coefficient is used to adjust the priority among accuracy, economy, and environmental protection.

[0099] Set constraints that meet crop requirements, total fertilizer application limits, and fertilizer application limits for high-slope areas;

[0100] Specifically, ensure that the fertilizer application in each grid cell meets the crop's minimum requirements. Ensure that the total fertilizer application in each grid cell is less than or equal to the total fertilizer resource limit. Limit fertilization in high-slope areas to the maximum allowable amount.

[0101] After constructing the model, the particle swarm optimization algorithm (PSO) is used for iterative calculation to find the optimal value of the objective function. It should be noted that PSO is an optimization algorithm that simulates swarm intelligence. Each particle represents a possible fertilization scheme. By updating the particle position and velocity, the global optimum is gradually approached.

[0102] Specifically, a swarm of particles is randomly generated, with each particle representing a fertilization scheme, and each particle is assigned an initial velocity and position;

[0103] Calculate the comprehensive objective function value for each particle. Record the individual optimal solution and the global optimal solution;

[0104] Update the particle's position and velocity according to the current velocity and position update formula, as shown in the expression:

[0105] ;

[0106] ;

[0107] in, For the first The particle in the first The speed on the dimension, updated to the 1st The value of the next iteration. For the first The particle in the first The velocity in the dimension is in the first The value of the next iteration. For the first The particle in the first The individual's historical optimal position on the dimension (i.e., the best solution obtained by the particle in the historical iteration). For inertial weights, , As a learning factor, , It is a random number. For the first The particle in the first The position above is in the dimensional position. The value of the next iteration (i.e., the current solution of the particle). For the first The global optimal position in the dimension (i.e., the historical best solution of the entire particle swarm in that dimension). For the first The particle in the first The position on the dimension has been updated to the current position. The value of the next iteration (i.e., the next solution for the particle).

[0108] The particles are repeatedly updated until the maximum number of iterations is reached or the change in the objective function value is less than a set threshold, at which point the optimal solution is output.

[0109] After optimization, the fertilization amount of each grid cell is converted into a gridded fertilization prescription map, and the field map is overlaid to generate a visualization result, outputting the final fertilization prescription map.

[0110] It should be noted that the output format is a gridded fertilization map, with each grid labeled with the application rate and type of each type of fertilizer (such as nitrogen, phosphorus, and potassium). The fertilization prescription map can be directly input into agricultural machinery to guide precision fertilization. The optimized model comprehensively considers soil characteristics, crop needs, and terrain features, making fertilization more scientific and rational. Through cost optimization, it reduces fertilizer waste and lowers production costs. It also adjusts fertilizer application rates based on slope data to reduce fertilizer runoff and environmental pollution. The model is highly visual and practical; the output gridded fertilization map is easy for agricultural machinery to execute directly, improving fertilization efficiency.

[0111] The spatial calibration module is used to acquire high-precision positioning data through the BeiDou positioning system, integrate the fertilizer prescription map with the high-precision positioning data, and complete the spatial calibration of the field.

[0112] The BeiDou Navigation Satellite System is used to obtain high-precision coordinates of field boundaries and record the latitude and longitude information of boundary points; the coordinates of the center point of the field and the boundary area are obtained.

[0113] The pixel coordinates of the fertilizer prescription map are converted into geographic coordinates using a coordinate transformation formula.

[0114] The field boundaries and zone coordinates are overlaid onto the field map in the BeiDou system to check the spatial matching degree between the fertilizer prescription map and the actual field, ensuring that there is no misalignment or offset.

[0115] It should be noted that the use of BeiDou high-precision positioning solves the problem of ambiguous field location in traditional fertilization, ensuring a perfect match between fertilizer application rate and location. This provides geographical reference for subsequent path planning and intelligent control of the fertilizer spreader, making fertilization operations more intelligent. By accurately marking field boundaries, it avoids the fertilizer spreader operating in non-target areas (such as outside the field or areas where fertilization is unnecessary), thus conserving fertilizer resources. Compared with other existing technologies, its spatially precise calibration and preprocessing of fertilizer prescription maps improve fertilization accuracy and fertilizer utilization.

[0116] The path generation module is used to generate the optimal fertilizer application path based on the field spatial calibration results and high-precision positioning data, using the Dijkstra algorithm.

[0117] By using the BeiDou positioning system, the coordinates of field boundary points are collected, the boundary point set is recorded, and the characteristic points of the field are determined, including the starting point, ending point, and coordinates of obstacle areas (such as crop growth areas) within the field.

[0118] Define the working width of the fertilizer spreader (Unit: meters) and spacing of fertilizer application path Satisfy the following formula:

[0119]

[0120] in, This is the overlap coefficient, which typically ranges from 0.8 to 1.0 (adjusted according to the requirements for uniform fertilizer application).

[0121] The area within the field boundary is divided into several parallel paths perpendicular to the fertilizer spreading direction; a fertilizer spreading path grid is generated based on the working width of the fertilizer spreader and the path spacing, and the grid point set is recorded; each point in the field is regarded as a node in the graph, and the boundary is regarded as an obstacle; grid points in the obstacle area are marked as impassable points to avoid the path passing through the obstacle area; the point set after the path is gridded is constructed into a weighted graph.

[0122] Set the fertilizer application start point as the starting point of the weighted graph and the end point as the ending point; use Dijkstra's algorithm to calculate the shortest path from the start point to the end point, while covering all walkable points in the grid to ensure no path is missed; output the generated path point set, where each point is connected sequentially to form the fertilizer application path.

[0123] Bezier curves or spline curves are used to smooth the fertilizer spreading path to ensure the stability of the fertilizer spreader during movement. The smoothed fertilizer spreading path is then verified to check whether the path completely covers all fertilizer spreading areas in the field. If any omissions or overlaps are found, the path grid spacing or fertilizer spreading direction is readjusted to ensure that the path is free of repetition and omissions.

[0124] The system monitors the tractor's position in real time, calculates the deviation between the tractor's current position and the planned path points, and triggers path correction if the deviation exceeds a set threshold (e.g., 0.5 meters). Based on the current position and the planned path points, it calculates the tractor's deviation direction and correction angle, adjusts the tractor's travel direction, and brings it back onto the planned path.

[0125] Real-time monitoring of changes in the field environment (such as obstacles, wetlands, etc.) is conducted. If new impassable areas are discovered, the path grid point set and path point set are dynamically updated. The optimal path is recalculated to adapt to the dynamic environment and ensure that the fertilizer spreader runs along the updated path.

[0126] It should be noted that optimal path planning reduces the tractor's unnecessary travel distance, shortens fertilizer application time, and improves operational efficiency. By optimizing the path, tractor fuel consumption and fertilizer waste are reduced, thus lowering agricultural production costs. Path optimization ensures that all fertilization areas in the field are accurately covered, without any omissions or duplicate applications.

[0127] The opening control module, which combines the tractor's real-time speed and the fertilizer application simulation model, dynamically adjusts the opening size of the fertilizer spreader and uses control algorithms to precisely control the fertilizer application rate. The module includes the following steps:

[0128] A fertilizer spreader opening size controller based on a servo motor rotation and linkage system is adopted. The rotation angle range of the servo motor should cover the adjustment range of the opening size.

[0129] Precise control of the servo motor is achieved through the lower-level computer. The control signal of the servo motor is sent by the lower-level computer, and the upper-level computer receives and processes the open disk size data sent by the upper-level computer.

[0130] The lower-level machine is equipped with an environmental sensor to detect environmental parameters during operation and transmit them to the upper-level machine, ensuring that the upper-level machine has enough data to accurately predict the opening size of the fertilizer spreader and plan the path.

[0131] The host computer obtains the real-time speed of the tractor through the Beidou positioning system and calculates the target opening size, target angle and speed based on the simulation model;

[0132] By adding absolute encoders to the front wheels and incremental encoders to the rear wheels, the tractor can be precisely controlled to operate as expected after receiving data.

[0133] The tractor's angle and speed are controlled by adjusting the PID control algorithm. The calculation formula is as follows:

[0134]

[0135] in, To control the output, This represents the error value for the current angle or velocity, indicating the difference between the current state (angle or velocity) and the target angle or velocity. This is a proportionality coefficient, reflecting the direct impact of errors on opening price adjustments. The integral coefficient reflects the impact of cumulative error on the adjustment. The differential coefficient reflects the effect of the rate of change of error on the adjustment;

[0136] Furthermore, the tractor speed is acquired in real time, and the target opening size is dynamically updated based on speed changes. The real-time status is compared with the target angle or speed to calculate the error. A PID control algorithm is then used to dynamically adjust the tractor angle and speed based on the error. If the speed change is significant (e.g., a sudden decrease or increase in tractor speed exceeding 10%), the controller response time should be less than 1 second to quickly adjust the opening size to adapt to the speed change. PID parameters are optimized by incorporating historical error data. , , This will further improve control precision.

[0137] It should be noted that the use of high-precision servo control allows for rapid response to system decisions, adjusting the opening size based on vehicle speed and fertilizer spreading data, resulting in better real-time control and improved fertilizer spreading accuracy. PID control can adjust the tractor's direction and speed in real time, ensuring it travels strictly along the predetermined path, reducing deviations and improving the accuracy and precision of fertilizer spreading.

[0138] The status monitoring module is used to monitor the tractor's turning angle and field boundary in real time and automatically shut down the fertilizer spreader;

[0139] The BeiDou positioning system is used to obtain the real-time location and driving status of the tractor, and this data provides a basis for judging the tractor's turning angle and boundary.

[0140] Set a turning angle threshold to determine whether the tractor turns an angle exceeding a specific angle (such as 15°). If the turning angle exceeds the threshold, it may cause uneven fertilization by the fertilizer spreader.

[0141] When the tractor's turning angle exceeds the turning angle threshold, the fertilizer spreader is paused. Once the tractor's turning angle is less than or equal to the turning angle and the direction is stable, the fertilizer spreading operation resumes. During the pause, the tractor's movement status is continuously monitored, and its real-time position and turning angle are recorded. When the tractor's turning is detected as complete (i.e., the tractor's turning angle is less than or equal to the turning angle threshold and the turning angle change tends to zero for several consecutive moments), an activation signal is sent to the fertilizer spreader controller to resume the fertilizer spreading operation. The fertilizer spreader's operating status is updated to ensure no area is missed.

[0142] The system monitors the tractor's location in real time, determines whether it is approaching the field boundary, and decides whether to turn off the fertilizer spreader based on the distance.

[0143] Furthermore, the polygonal range of the field boundary is defined based on the set of field boundary points, with the boundary points provided by the field spatial calibration results; the point-to-polygon distance formula is used to calculate the shortest distance from the tractor's current position to the field boundary;

[0144] Set a distance threshold Y=1m to determine whether the distance between the tractor and the boundary is less than or equal to the distance threshold Y. If the condition is met, trigger the fertilizer spreader to shut down; otherwise, the fertilizer spreader remains running.

[0145] When the distance between the tractor and the boundary is less than or equal to the distance threshold Y, a shutdown signal is sent to the fertilizer spreader controller to immediately stop fertilizer release; the time of fertilizer spreader shutdown and the position of the tractor are recorded for subsequent analysis and optimization of the operation process; when the tractor moves away from the boundary (i.e. the distance between the tractor and the boundary is greater than the distance threshold Y), the fertilizer spreader is restarted to resume fertilization operation.

[0146] In summary, this invention achieves the following: By establishing a fertilization simulation model, the fertilizer spreader can accurately calculate the real-time fertilization amount based on the size of the application area during actual operation. By collecting and preprocessing soil analysis data and remote sensing imagery of the field, precise input data is provided for generating the fertilization prescription map. The combination of multi-dimensional data ensures that the generated fertilization prescription map better meets the actual needs of the field. Through comprehensive analysis of soil nutrient distribution, crop growth, and terrain characteristics, the fertilization prescription map guides the fertilizer spreader to apply appropriate amounts of fertilizer in different areas. The fertilization prescription map is the core data output for precise fertilization, indicating the amount of fertilizer required in different areas of the field, achieving "on-demand fertilization." High-precision positioning data of the tractor and the field is obtained through the BeiDou positioning system. The fertilization prescription map is then fused with the positioning data to spatially calibrate the field, clarifying the field boundaries and the geographical location of the fertilization area. By combining the fertilization prescription map with BeiDou positioning data, the fertilization requirement points in the prescription map can be accurately mapped to the actual geographical location of the field, ensuring that the fertilizer spreader fertilizes at the correct time and location. Dijkstra's algorithm can be used to plan the optimal path covering all fertilization areas of a field, avoiding duplicate operations or missed areas, thereby improving operational efficiency.

[0147] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A Beidou positioning-based intelligent fertilizer spreading control system, characterized in that: include, The model building module is used to build a fertilization simulation model and obtain data on the relationship between the opening size of the fertilizer spreader and the amount of fertilizer applied through experiments; the opening size of the fertilizer spreader refers to the effective opening area of ​​the fertilizer spreader's discharge port. The data acquisition module is used to collect soil analysis data and remote sensing image data, and to perform preprocessing. The prescription map generation module is used to generate fertilizer prescription maps based on preprocessed soil analysis data and remote sensing image data, according to the soil characteristics and crop requirements of the field. The spatial calibration module is used to acquire high-precision positioning data through the BeiDou positioning system, integrate the fertilizer prescription map with the high-precision positioning data, and complete the spatial calibration of the field. The path generation module is used to generate the optimal fertilizer application path based on the field spatial calibration results and high-precision positioning data, using the Dijkstra algorithm. The opening control module is used to dynamically adjust the opening size of the fertilizer spreader by combining the real-time speed of the tractor and the fertilizer application simulation model, and to use the control algorithm to precisely control the amount of fertilizer applied. A fertilization simulation model was established, and data on the relationship between the size of the fertilizer spreader's opening and the amount of fertilizer applied were obtained through simulation experiments. This included the following steps. An accurate physical model of the fertilizer spreader is established using existing fertilizer spreaders, and the physical model is then imported into EDEM software for simulation. Determine the types of fertilizers suitable for the fertilizer spreader, and set the parameters for different types of fertilizers and the opening size of the fertilizer spreader in the simulation software; Multiple simulation experiments were conducted under each opening size setting, and finite element simulation analysis was performed by changing the parameters of different fertilizer types each time to obtain comprehensive simulation results. Record the total mass of fertilizer applied per unit time for each opening size, and calculate the corresponding fertilizer application rate; Repeat the simulation experiment, take the average amount of fertilizer as the final data, and record the corresponding opening size; Record the parameters during the simulation experiment, organize the simulation experiment data into tables, use MATLAB to draw the relationship between the opening size and the amount of fertilizer, determine the preliminary trend and linear or nonlinear relationship, and select different mathematical models for fitting. Based on the soil characteristics and crop requirements of the field, and using preprocessed soil analysis data and remote sensing imagery data, a multi-objective optimization model is employed to generate a fertilizer prescription map, including the following steps. Spatial registration was performed between soil analysis data and remote sensing image data. QGIS3 software was used to load the soil analysis data and remote sensing image data and perform spatial analysis to determine the soil characteristics and vegetation index of the field. The field was divided into zones using the K-means clustering method, and a soil zoning map of the field was generated based on the clustering results. The optimization objectives are determined by the accuracy of fertilization, the cost of fertilization, and the risk of fertilizer loss. Set constraints that meet crop requirements, total fertilizer application limits, and fertilizer application limits for high-slope areas; Particle swarm optimization algorithm is adopted as the optimization algorithm for the multi-objective optimization model; Randomly generate a swarm of particles and assign an initial velocity and position to each particle; Calculate the comprehensive objective function value for each particle, and record the individual optimal solution and the global optimal solution; Update the particle's position and velocity according to the formula for updating the particle's current velocity and position; The particle is repeatedly updated until a maximum iteration number is reached or a change in the objective function value is less than a set threshold, and an optimal solution is output; After optimization, the fertilizer application amount of each grid cell is converted into a grid-based fertilization prescription map, and the final fertilization prescription map is output by superimposing the field map.

2. The intelligent fertilization control system based on Beidou positioning according to claim 1, characterized in that: Collect soil analysis data and remote sensing image data, and preprocess them, including the following steps, Soil sampling is conducted in selected areas, and soil samples are analyzed; High-resolution remote sensing image data is obtained using satellite remote sensing technology; For the collected remote sensing image data, use MATLAB software for geometric correction, atmospheric correction and radiation correction; The soil analysis data is cleaned and standardized. 3.The intelligent fertilizing control system based on Beidou positioning of claim 2, characterized in that: High-precision positioning data is obtained through the Beidou positioning system, the fertilization prescription map is fused with the high-precision positioning data, and the field space is calibrated, including the following steps, High-precision coordinates of the field boundary are obtained using the Beidou positioning system, and the latitude and longitude information of the boundary points is recorded. The center point coordinates and boundary area of the field are obtained; The pixel coordinates of the fertilization prescription map are converted to geographic coordinates through the coordinate conversion formula; The field boundary and partition coordinates are superimposed on the field map in the Beidou system to check the spatial matching degree of the fertilization prescription map with the actual field.

4. The intelligent fertilization control system based on Beidou positioning according to claim 3, characterized in that: Based on the field space calibration results and high-precision positioning data, the Dijkstra algorithm is used to generate the optimal fertilization path, including the following steps, High-precision positioning data is obtained through the Beidou positioning system, and the boundary and feature points of the field are recorded; The Dijkstra algorithm is used for fertilization path planning, and high-precision position data is used to provide the starting point and ending point for the fertilization path planning algorithm. Each point in the field is regarded as a node in the graph, and the boundary is regarded as an obstacle; Bezier curves or spline curves are used to smooth the fertilization path, and the smoothed fertilization path is verified; Real-time monitoring of tractor position, path correction through high-precision positioning, recalculation of path according to environmental changes.

5. The intelligent fertilization control system based on Beidou positioning according to claim 4, characterized in that: Combine real-time speed of tractor and fertilization simulation model, dynamically control the opening size of fertilizer distributor, use control algorithm to accurately control the amount of fertilizer, including the following steps, Use the fertilizer distributor opening size controller based on the steering motor rotation and linkage system to accurately control the steering motor through the lower computer. The upper computer receives and processes the opening size data sent; The lower computer adds environmental sensors to detect environmental parameters during operation and transmits them to the upper computer, ensuring that the upper computer has enough data to accurately predict the opening size of the fertilizer distributor and plan the path; The upper computer obtains the real-time speed of the tractor through the Beidou positioning system, and calculates the target opening size, target angle and speed according to the simulation model; The tractor is equipped with an absolute value encoder on the front wheels and an incremental encoder on the rear wheels to ensure accurate control of the tractor according to the expected requirements. The PID control algorithm is used to control the angle and speed of the tractor.

6. The intelligent fertilization control system based on Beidou positioning according to claim 5, characterized in that: The calculation formula of the PID control algorithm is: wherein, for controlling the output, is an error value of the current angle or speed, is a proportional coefficient, is an integral coefficient, is a differential coefficient.

7. The intelligent fertilization control system based on Beidou positioning according to claim 6, characterized in that: It also includes a state monitoring module for real-time monitoring of tractor angle and field boundary, and automatic shutdown of the fertilizer distributor. 8.The intelligent fertilizing control system based on Beidou positioning of claim 7, wherein: Real-time monitoring of tractor corner and field boundary, automatic closing of fertilizer distributor, including the following steps, Using Beidou positioning system to obtain real-time position and driving state of tractor; Setting the corner threshold to determine whether the tractor corner exceeds a certain angle; When the corner of the tractor is greater than the corner threshold, the fertilizer distributor is triggered to suspend operation, and when the corner of the tractor is less than or equal to the corner and the direction is stable, the fertilizer distribution operation is resumed; Real-time monitoring of the position of the tractor to determine whether it is close to the field boundary, and whether to close the fertilizer distributor according to the distance.

Images