Agricultural automation linkage control method and control system based on artificial intelligence

Abstract

The invention provides an agricultural automation linkage control method and system based on artificial intelligence. Belongs to the technical field of agricultural automation control. The method comprises the following steps: acquiring agricultural environment data and agricultural automation equipment data, and preprocessing the acquired agricultural environment data; performing feature extraction on the preprocessed agricultural environment data, and constructing an agricultural environment feature vector; and performing agricultural environment state analysis according to the agricultural environment feature vector to obtain an agricultural environment state evaluation result. Through preliminary control simulation, reinforcement learning algorithm optimization control strategy and linkage control simulation, and continuous iteration optimization of equipment control strategy, the control strategy better meets the actual production requirement, the accuracy and effectiveness of agricultural automation equipment control are improved, manual intervention is reduced, and the production cost is reduced.

Description

Technical Field

[0001] The present invention proposes an agricultural automation linkage control method and control system based on artificial intelligence, belonging to the technical field of agricultural automation control. Background Art

[0002] Traditional agricultural production methods rely on manual experience for operations such as irrigation, fertilization, and environmental regulation. This is not only labor-intensive but also difficult to accurately grasp the various conditions required for plant growth, resulting in low production efficiency and serious waste of resources. While the development of automated agricultural equipment has improved production efficiency to a certain extent, the lack of effective coordination and linkage between various devices makes it impossible to comprehensively regulate and control according to the real-time needs of plant growth. Furthermore, existing technologies fail to fully consider the creation of a wild-like growth environment, which is not conducive to improving plant quality. Summary of the Invention

[0003] The present invention provides an agricultural automation linkage control method and control system based on artificial intelligence to solve the problems mentioned in the above background technology: The present invention proposes an agricultural automation linkage control method based on artificial intelligence, the method comprising: S1. Acquire agricultural environment data and agricultural automation equipment data, and preprocess the acquired agricultural environment data; perform feature extraction on the preprocessed agricultural environment data to construct an agricultural environment feature vector; perform agricultural environment status analysis based on the agricultural environment feature vector to obtain an agricultural environment status assessment result; perform agricultural automation equipment correlation analysis based on the agricultural environment status assessment result and agricultural automation equipment data to obtain an agricultural automation equipment correlation relationship; perform agricultural automation architecture integration based on the agricultural automation equipment correlation relationship to obtain an agricultural automation linkage control architecture; S2. Perform image recognition processing on agricultural environmental data, extract plant growth parameters based on the recognition results, and use machine learning algorithms to predict the growth trends of plant growth parameters based on the growth models and historical data of different plants to obtain plant growth trend data; perform environmental demand analysis on agricultural environmental data based on the plant growth trend data to obtain plant growth environmental demand data; classify the plant growth environmental demand data to obtain classified environmental demand data; S3. Conduct preliminary control simulation of agricultural automation equipment using the agricultural automation linkage control architecture to obtain preliminary equipment control simulation data; optimize the control strategy using a reinforcement learning algorithm based on the preliminary equipment control simulation data and plant growth environment requirement data to obtain an optimized equipment control strategy; conduct linkage control simulation of the agricultural automation equipment based on the optimized equipment control strategy to obtain equipment linkage control simulation data; S4. Obtain agricultural production target data, conduct a target feasibility assessment on the agricultural production target data based on the plant growth trend data, and obtain target feasibility assessment data; use a deep learning algorithm to predict resource demand based on the target feasibility assessment data and the classified environmental demand data, and obtain resource demand prediction data; combine the resource demand prediction data with the equipment linkage control simulation data to optimize resource allocation and obtain an optimized resource allocation plan; S5. Based on the optimized resource allocation plan and equipment linkage control simulation data, the agricultural production target data is decomposed into targets to obtain equipment control target data; based on the equipment control target data and the agricultural automation linkage control architecture, the genetic algorithm is used to optimize the equipment control parameters to obtain the optimized equipment control parameters; based on the optimized equipment control parameters, the agricultural automation linkage control strategy is generated, the control strategy is transmitted to the agricultural automation equipment control system, and the linkage control task is executed.

[0004] The artificial intelligence-based agricultural automation linkage control system proposed in the present invention includes: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement any one of the methods described above.

[0005] The beneficial effects of the present invention are as follows: by acquiring agricultural environmental data and preprocessing and extracting features to construct feature vectors, and then analyzing the agricultural environmental status, the agricultural environmental status can be fully and accurately grasped, providing a reliable basis for subsequent precise control, avoiding control errors caused by incomplete or inaccurate environmental information, and helping to improve the stability and quality of agricultural production.

[0006] By using image recognition technology to obtain plant growth parameters and combining growth models and historical data to predict growth trends, and then analyzing and classifying the plant growth environment requirements, we can understand the plant growth status and required environmental conditions in advance, provide strong support for precise control of the agricultural environment to meet plant growth needs, and help improve crop yield and quality.

[0007] Through preliminary control simulation, reinforcement learning algorithm optimization of control strategy and linkage control simulation, the equipment control strategy is continuously iterated and optimized to make the control strategy more in line with actual production needs, improve the accuracy and effectiveness of agricultural automation equipment control, reduce manual intervention and reduce production costs.

[0008] By evaluating the feasibility of agricultural production goals based on plant growth trends, using deep learning algorithms to predict resource needs, and combining equipment linkage control simulation data to optimize resource allocation, it is possible to rationally allocate agricultural production resources, avoid excessive or insufficient resource investment, improve resource utilization efficiency, and achieve sustainable development of agricultural production.

[0009] By breaking down agricultural production goals into smaller pieces and optimizing equipment control parameters with genetic algorithms, agricultural automation linkage control strategies are generated and executed. This can transform macro-production goals into specific and actionable equipment control instructions, ensuring that agricultural automation equipment operates according to optimal parameters, accurately achieving agricultural production goals, and improving the level of refined management of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a step diagram of the method of the present invention. DETAILED DESCRIPTION

[0011] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0012] One embodiment of the present invention, for example Figure 1 As shown, an agricultural automation linkage control method based on artificial intelligence, the method includes: S1: Acquire agricultural environment data and agricultural automation equipment data, wherein the agricultural environment data includes temperature, humidity, light intensity, air pressure, and air quality; the agricultural automation equipment data includes the operating status and parameters of drones, automatic irrigation systems, smart greenhouse control equipment, etc.; preprocess the acquired agricultural environment data; use a machine learning algorithm (such as a random forest algorithm) to extract features from the preprocessed agricultural environment data and construct an agricultural environment feature vector; perform agricultural environment status analysis based on the agricultural environment feature vector to obtain an agricultural environment status assessment result; based on the agricultural environment status assessment result and agricultural automation equipment data, perform agricultural automation equipment association analysis through a deep learning model (such as a graph neural network model) to obtain an agricultural automation equipment association relationship; integrate the agricultural automation architecture based on the agricultural automation equipment association relationship to obtain an agricultural automation linkage control architecture; S2: Perform image recognition processing on agricultural environmental data, using image recognition technology to identify the growth status, pest and disease conditions, etc. of different plants; extract plant growth parameters based on the recognition results, including plant height, leaf color, and number of flowers; combine growth models and historical data of different plants, and use machine learning algorithms (such as convolutional neural networks) to predict the growth trends of plant growth parameters to obtain plant growth trend data; perform environmental demand analysis on agricultural environmental data based on the plant growth trend data to obtain plant growth environment demand data; classify the plant growth environment demand data to distinguish between conventional environmental demands and special environmental demands, such as simulated wild environment demands such as dawn and dusk, fog, temperature difference between morning and evening, and air pressure difference, to obtain classified environmental demand data; S3: Conduct preliminary control simulations of agricultural automation equipment using the agricultural automation linkage control architecture, simulating the operation of different equipment under different control strategies to obtain preliminary equipment control simulation data. Based on the preliminary equipment control simulation data and plant growth environment requirement data, optimize the control strategy using a reinforcement learning algorithm (such as the policy gradient method), continuously adjust the equipment's control parameters to maximize plant growth requirements, and obtain the optimized equipment control strategy. Based on the optimized equipment control strategy, conduct linkage control simulations of agricultural automation equipment, taking into account the synergy and mutual influence between equipment, and obtain equipment linkage control simulation data. S4: Acquire agricultural production target data, including yield targets and quality targets; perform a target feasibility assessment on the agricultural production target data based on plant growth trend data, analyze the possibility of achieving the production target under current environmental conditions, and obtain target feasibility assessment data; use a deep learning algorithm to perform resource demand forecasting based on the target feasibility assessment data and the classified environmental demand data, including forecasting the amount of irrigation water, fertilizer application, energy consumption, and other resources required to achieve the production target, to obtain resource demand forecast data; combine the resource demand forecast data with the equipment linkage control simulation data to optimize resource allocation and obtain an optimized resource allocation plan; S5: Based on the optimized resource allocation plan and equipment linkage control simulation data, the agricultural production target data is broken down into detailed targets, and the overall target is broken down into specific control targets for each device to obtain the equipment control target data; based on the equipment control target data and the agricultural automation linkage control architecture, a genetic algorithm (such as an adaptive genetic algorithm) is used to optimize the equipment control parameters, and the operating parameters of the equipment are further adjusted to achieve precise control of the equipment and obtain the optimized equipment control parameters; based on the optimized equipment control parameters, an agricultural automation linkage control strategy is generated, and the control strategy is transmitted to the agricultural automation equipment control system to execute the linkage control task.

[0013] The working principle and effect of the above technical solution are as follows: through the precise analysis of agricultural environmental data and the application of optimized control strategies, it can achieve precise linkage of agricultural automation equipment, reduce resource waste, and improve agricultural production efficiency; the use of intelligent algorithms and automated control reduces dependence on manual operation, reduces labor costs, and can maintain a stable agricultural production environment for a long time; Through optimized equipment linkage control strategies, it is possible to simulate and achieve a wild-like plant growth environment, meet the special environmental needs of different plants, and improve plant growth quality and yield. Through resource demand prediction and resource allocation optimization, it ensures the precise allocation of resources such as irrigation and fertilization, avoids excessive resource use, reduces energy consumption, and improves resource utilization efficiency. Taking into account plant growth trends and environmental requirements, we conducted feasibility assessments of production targets and forecasted resource requirements, further enhancing the likelihood of achieving the targets and ensuring the realization of agricultural production goals. Through the combined application of deep learning and reinforcement learning algorithms, we enhanced the synergy between different agricultural automation equipment, avoided conflicts and redundant operations between devices, and ensured efficient collaboration of equipment. By precisely controlling the plant growth environment, we improved the growth and health of plants, thereby improving plant quality and market competitiveness while ensuring yield.

[0014] In one embodiment of the present invention, the S1 includes: S11. Real-time agricultural environmental data is collected through a distributed sensor network, where the distributed sensors are deployed in different areas of the field and greenhouse. The agricultural environmental data includes temperature (accurate to ±0.5°C), humidity (accurate to ±2%RH), light intensity (unit: lux), spectrum (unit: nm), photosynthetic photon flux density (unit: umol / m2 / s), air pressure (unit: hPa), and air quality (including indicators such as PM2.5 and CO2 concentration). Agricultural automation equipment data is collected through the device Internet of Things interface, where the Internet of Things interface includes LoRa and 4G modules. The agricultural automation equipment data includes the operating status and parameters of drones (flight trajectory, battery life, spray flow), automatic irrigation systems (water pump power, pipeline pressure, valve switch status), and smart greenhouse control equipment (ventilation fan speed, shade net opening and closing degree), forming a raw data set. S12. Perform multi-level cleaning on the collected agricultural environmental data to remove outliers caused by sensor failure (identified using the 3σ principle); interpolate missing values ​​(using linear interpolation for time series and kriging interpolation for spatial distribution); standardize data of different units and magnitudes (map them to the interval [0,1]); and finally generate a regular agricultural environmental preprocessed dataset. S13. Using a machine learning algorithm to extract features from the preprocessed agricultural environmental data, screening out core features strongly correlated with plant growth, including diurnal temperature range, duration of sustained high humidity, and cumulative light intensity; compressing the high-dimensional environmental data into low-dimensional feature vectors with an information retention rate of ≥90% through principal component analysis, where each vector contains 8-12 core environmental features, forming an agricultural environmental feature vector; S14. Based on the agricultural environmental characteristic vectors, a multi-dimensional assessment model is constructed, and environmental suitability indicators are set. The suitability indicators include the temperature range suitable for crop growth and the light threshold. The environmental suitability score corresponding to each characteristic vector is calculated using a fuzzy comprehensive evaluation method. The environmental status is classified into excellent, medium, and poor levels based on the correlation between historical environmental data for the same period and crop growth, and the agricultural environmental status assessment results (including specific indicator scores and comprehensive grades) are generated. S15. Input the agricultural environmental status assessment results and agricultural automation equipment data into a deep learning model. Construct a device association graph with the devices as nodes and the collaborative relationships between devices (e.g., the coordinated humidity adjustment between the irrigation system and greenhouse ventilation equipment) as edges. Through model training, learn the mapping relationship between environmental status changes and equipment operation, identify strong association rules between devices (e.g., the probability of coordinated activation of ventilation fans and sunshade nets at high temperatures), and obtain the association relationships between agricultural automation equipment. S16. Based on the device association relationship, design a hierarchical linkage control architecture, in which the bottom layer is the device execution layer (directly controlling device actions), the middle layer is the collaborative control layer (processing the linkage logic between devices), and the top layer is the decision-making layer (receiving environmental assessment results and issuing instructions); determine the data interaction protocol for each layer (for example, the MQTT protocol is used for communication between the device layer and the control layer), and integrate them to form an agricultural automation linkage control architecture that can dynamically adapt to environmental changes.

[0015] The working principle and effect of the above technical solution are as follows: by accurately collecting agricultural environmental data, the system can more efficiently adjust the equipment status, realize automated management, reduce manual intervention, optimize the production process, and thus improve overall production efficiency; through the intelligent control system, irrigation, fertilization and other resources can be allocated on demand, avoiding excessive use or waste, reducing energy consumption, and improving resource utilization efficiency.

[0016] The system optimizes the synergy between devices through deep learning algorithms, reduces conflicts and redundant operations between devices, and ensures efficient and coordinated operation of equipment; by adjusting environmental parameters in real time, it provides plants with a more suitable growth environment, thereby improving the health and quality of plants.

[0017] Because the system can automatically monitor and adjust the agricultural environment, it reduces dependence on manual operations, thereby reducing labor costs and ensuring long-term production stability; through the rational allocation of resources and optimization of equipment linkage, agricultural production can be carried out in a more sustainable mode, reducing excessive dependence on natural resources and contributing to environmental protection.

[0018] In one embodiment of the present invention, the S15 includes: Extract the specific indicator scores, comprehensive grades, and corresponding timestamps from the agricultural environmental status assessment results, and simultaneously extract the operating status parameters, equipment identification, and acquisition timestamps from the agricultural automation equipment data. Associate and bind the two types of data through timestamp matching, and eliminate data pairs with time dislocations exceeding 10 seconds to generate a fused dataset. Based on the agricultural environmental status assessment results (excellent, medium, poor) and specific environmental indicators (such as high temperature and high humidity), the equipment operating status in the fused dataset is classified into scenarios, such as "equipment collaboration in a high temperature environment (temperature ≥ 35°C)" and "equipment collaboration in a high humidity environment (humidity ≥ 85%)." This determines the list of equipment involved in collaboration in different scenarios. The agricultural automation equipment involved in the scenario is used as a node, and the device type and operating parameter range are used as node attributes. Based on the consistency of the device's actions at the same timestamp (for example, simultaneous startup and simultaneous parameter adjustment), initial association edges are established between nodes, and the initial association frequency of the edges is marked (the number of times the same action occurs simultaneously). The initial device association graph containing node attributes and initial association edges, along with the environmental state data for the corresponding scenario, is input into the graph neural network model. This model learns the mapping relationship between environmental state changes and the strength of device association edges, and optimizes the weights of the association edges through model iteration (the weights are dynamically adjusted based on the frequency of collaboration and the effect of environmental improvement). Based on the preset weight threshold (≥0.8), the associated edges with optimized weight values ​​that meet the standards are screened out. Combined with the environmental status of the corresponding scene, strong association rules between equipment are extracted. For example, "when the environmental assessment level is poor and the temperature index exceeds the standard, the coordinated activation weight of the ventilation fan and the sunshade net is ≥0.85", which is integrated to form the association relationship of agricultural automation equipment.

[0019] The working principle and effect of the above technical solution are as follows: through scene classification and the extraction of device association rules, the system can optimize device collaboration strategies according to different environmental conditions, making cooperation between devices more accurate and efficient, and reducing conflicts and incoordination in device operation. Through the learning and iterative optimization of the graph neural network model, the system can dynamically adjust the weight values ​​of device association edges to ensure the accuracy of device collaboration under different environmental conditions, thereby effectively reducing errors in the environmental control process. By utilizing optimized equipment association rules, the system can automatically trigger coordinated equipment adjustments when environmental factors such as temperature and humidity exceed standards, greatly improving the intelligent linkage of equipment and avoiding the individual operation of equipment under different environmental conditions, thereby improving the overall level of automated control. Through precise equipment collaboration and optimized scenario classification, the system can respond to environmental changes in real time, making the agricultural production environment more adaptable and helping to improve the stability and suitability of the crop growth environment.

[0020] In equipment collaboration, by screening and optimizing collaboration strategies, the system can reduce unnecessary equipment operation and energy consumption, further improve energy utilization efficiency, and reduce ineffective equipment consumption; through the optimized collaboration of automated equipment and intelligent analysis of environmental status assessment results, the degree of automation in agricultural production has been further improved, reducing the need for manual intervention and achieving efficient and intelligent agricultural environmental management.

[0021] In one embodiment of the present invention, the S2 includes: S21. Use high-definition cameras (deployed on field poles or carried by drones) to regularly collect plant growth images (once at 9:00 AM and once at 3:00 PM daily), covering plant panoramas, leaf close-ups, and details of flowers and fruits. Preprocess the images to generate a standardized growth image dataset. S22. Analyze the preprocessed image using a deep learning image recognition model. Identify plant growth status (leaf spread, stem uprightness) and pest and disease conditions (lesion shape, insect outline) using a trained feature extraction network (e.g., ResNet50). Output identification result labels (e.g., "yellowing leaves," "aphid infestation," "robust growth") and annotate the coordinates of abnormal areas. S23. Based on the image recognition results, extract quantitative growth parameters using an image measurement algorithm and calculate plant height using binocular vision (error ≤ 2 cm). Analyze leaf color using the HSV color space (extract the mean of the R, G, and B channels). Count the number of flowers using an object detection frame (accuracy ≥ 95%) to form a plant growth parameter set containing multi-dimensional parameters. S24. Input plant growth parameters, corresponding crop growth models (e.g., a logistic growth model for rice, an accumulated temperature growth model for tomatoes), and historical growth data (parameter change curves for the past three growth cycles) into a machine learning prediction model. The model is trained to learn how the parameters change over time (in days), and to predict growth parameter change trends (e.g., daily plant height growth, leaf color change) over the next 15 days to obtain plant growth trend data. S25. Based on growth trend data, a growth status-environmental factor association model is constructed. The environmental data corresponding to the "good growth" stage in the trend data is analyzed to determine the thresholds of key environmental factors for each growth stage (e.g., a suitable temperature of 20-25°C for the seedling stage and a suitable humidity of 60-70% for the flowering stage). Combined with crop physiological characteristics (e.g., the light requirement for photosynthesis), the environmental requirements that meet the growth trend are determined to generate plant growth environment requirement data. S26. Divide the plant growth environment requirement data into different dimensions, classifying those that meet the requirements of conventional agricultural production (such as constant temperature and stable humidity during the day) as conventional environmental requirements; classify the special conditions that simulate natural ecology (such as gradual changes in light intensity during the dawn and dusk periods, a temperature difference of 5-8°C between day and night, and humidity fluctuations in foggy environments) as simulated wild environment requirements; verify the classification boundaries using the K-means clustering algorithm to obtain the classified environmental requirement data.

[0022] The working principle and effect of the above technical solution are as follows: By combining high-definition cameras and deep learning image recognition models, the system can accurately obtain various plant growth data, especially in terms of pest and disease monitoring and growth status analysis, ensuring that key parameters in the plant growth process can be captured in real time and accurately. Traditional plant growth monitoring requires a lot of manual participation, but this technical solution significantly reduces labor costs through automated image acquisition and analysis, saving a lot of time and human resources, especially in large-scale agricultural production. By accurately identifying the growth status and pest and disease conditions of plants, the system can promptly detect potential problems and provide scientific decision-making support for agricultural managers, helping them to take measures in advance and reduce losses caused by pests and diseases. Combining plant growth models and historical growth data, the system can predict growth trends for the next 15 days, allowing farmers to prepare in advance for future environmental changes and plant growth needs, optimize resource allocation, and improve crop yield and quality.

[0023] By establishing a correlation model between growth status and environmental factors, the system can clearly identify the environmental requirements of different growth stages and adjust the environment based on real-time monitoring data, thereby ensuring that crops grow in the optimal environment and reducing the impact of adverse environmental factors on crop growth; based on the growth characteristics of different crops, the system can intelligently adjust environmental requirements and production methods, generate demand data for simulated wild environments and conventional environments, and help farmers develop personalized planting plans based on the needs of specific crops, thereby improving the level of intelligence in agricultural production.

[0024] In one embodiment of the present invention, the step S23 includes: Based on the recognition result labels and abnormal area coordinates obtained by S22, feature areas related to plant height, leaf color, and flower number are extracted from the standardized growth image dataset, and the specific location and range of each feature area in the image are determined; Using the stereo image of the characteristic area obtained by the binocular vision system, the spatial distance between the bottom and top of the plant in the characteristic area is calculated in combination with the principles of stereo geometry. Through the error correction mechanism (ensuring the error is ≤ 2cm), the quantitative data of the plant height is obtained; The image of the leaf in the feature area is converted from RGB color space to HSV color space, and the mean of the R, G, and B channels in this space is extracted as the quantization parameter of the leaf color; For flowers in the feature area, the number of flowers is counted using a counting algorithm based on the detection frame generated by the target detection algorithm, ensuring a statistical accuracy rate of ≥95% to obtain quantitative data on the number of flowers; The corresponding data including plant height, leaf color quantitative parameters and flower number are summarized and organized according to a unified data format to form a plant growth parameter set including multi-dimensional parameters.

[0025] The working principle and effect of the above technical solution are as follows: By combining a binocular vision system and the principles of stereo geometry, plant height can be accurately measured, ensuring that the error is controlled within 2 cm, greatly improving the accuracy and reliability of plant height data. Through the automated flower counting algorithm and leaf color quantification technology, the deviations that may occur in manual measurement are avoided, making flower number counting and leaf color analysis more objective and accurate, and reducing the impact of human error. Organizing multi-dimensional growth parameters into a standardized data format facilitates comparisons between crop growth data at different times and locations, improving data comparability and the depth of analysis. By extracting information from key characteristic areas of plants and quantifying it into standard data, not only can the growth status of plants be understood in real time, but their health status can also be accurately reflected, helping agricultural managers make timely adjustments and responses.

[0026] With the addition of these quantitative data, the accuracy of prediction models based on historical growth data and environmental factors can be further improved, helping farmers to more accurately predict future plant growth trends and optimize planting plans. By precisely quantifying each characteristic area, the impact of environmental changes on measurement results is reduced, ensuring the stability and availability of plant growth data, thereby making agricultural management decisions more scientific and reasonable.

[0027] In one embodiment of the present invention, S3 includes: S31. Based on the agricultural automation linkage control architecture, initial control parameters are set for each agricultural automation device. The initial control parameters include: the initial spray flow rate of the drone is set according to the crop density (e.g., 0.5 L per square meter); the initial watering cycle of the automatic irrigation system is set according to the soil moisture threshold (starting when it is below 60%); and the initial parameters of the smart greenhouse control device are set according to conventional environmental requirements (e.g., ventilation fans start when the temperature is greater than 30°C). This forms the initial control parameter set for the devices. S32. In a digital twin agricultural scenario (a three-dimensional virtual model based on actual farmland / greenhouse), input initial control parameters and classified environmental demand data, simulate the operation of different equipment under a single control strategy, set up multiple simulation scenarios (e.g., hot weather, rainy weather), record equipment operating status (e.g., whether drones spray according to their trajectory, whether irrigation systems start on time) and environmental responses (e.g., temperature change curves in the greenhouse), and obtain preliminary equipment control simulation data; S33. Use the preliminary control simulation data and plant growth environment requirement data as input to construct a reinforcement learning model (with "meeting environmental requirements" as the reward function and "equipment energy consumption" as the penalty function). By interacting with the environment (adjusting equipment control parameters), the strategy is continuously optimized (for example, when high temperatures are high, first activate the sunshade net and then reduce the ventilation fan speed to reduce energy consumption). After 500 rounds of iterative training, the strategy satisfaction (environmental requirement compliance rate) is ≥ 90%, and the optimized equipment control strategy is obtained. S34. Develop equipment linkage control rules based on the relationships between agricultural automation equipment, clarify the master-slave device logic (for example, when a drone detects pests and diseases, it automatically triggers a linkage instruction for the irrigation system to add pesticides); establish a priority mechanism (for example, in extreme weather, ventilation equipment takes precedence over irrigation equipment); and define parameter coordination thresholds (for example, when the light intensity is less than 5000 lux, the greenhouse fill light and ventilation fan are activated simultaneously), forming a linkage control rule library. S35. Load the optimized control strategy and linkage rules into the digital twin scenario to simulate the coordinated operation of multiple devices, the joint operation of drones and irrigation systems (drones locate pest and disease areas, and irrigation systems precisely apply pesticides), and the linkage of multiple devices in smart greenhouses (coordinated adjustment of temperature, humidity, and light). Record data such as the response delay between devices (≤2s), the compliance rate of environmental indicators, and resource consumption to obtain simulation data on device linkage control.

[0028] The working principle and effects of the above technical solution are as follows: Through the simulation of digital twin agricultural scenarios and the application of reinforcement learning models, equipment control strategies are optimized, enabling equipment to respond more accurately and efficiently to environmental demands, thereby significantly improving the operating efficiency of agricultural automation equipment and the level of crop growth management. Through the optimization of reinforcement learning models, equipment energy consumption is effectively controlled. By formulating equipment linkage control rules, the collaboration between different devices becomes smoother and more precise. For example, when a drone detects a pest, the irrigation system can automatically add pesticides, avoiding manual intervention and delays, and improving operational efficiency.

[0029] By testing the collaborative operation of equipment in simulated scenarios and recording the response delay between devices and the compliance rate of environmental indicators, we ensure the rapid response of equipment in complex environments, making agricultural management more flexible and efficient. Through automated equipment linkage and precise control, the errors and dependence of manual operations are reduced, ensuring the accuracy and consistency of each link in agricultural production and reducing the risk of human operational errors.

[0030] By building digital twin models and reinforcement learning optimization strategies, agricultural managers can obtain more scientific data support and adjust equipment parameters and management strategies in real time, thereby improving the rationality and effectiveness of decision-making.

[0031] In one embodiment of the present invention, the step S34 includes: S341. Based on the functional dependencies and collaboration strengths between agricultural automation equipment, divide the hierarchy of master and slave devices, clarify the logical conditions under which a master device triggers a slave device action (for example, a drone acting as a master device for pest detection automatically sends a pesticide-carrying irrigation instruction to the irrigation system (slave device) when it identifies a pest density of ≥ 5 insects / m2), and establish a master-slave device trigger mapping table. S342. Analyze the urgency of different agricultural production scenarios (e.g., extreme high temperatures (≥40°C) and rainstorm warnings are considered Level 1 emergency scenarios, while routine growth regulation is considered Level 3). Assign operational priorities to equipment based on the scenario level. In Level 1 scenarios, ventilation and cooling equipment takes precedence over irrigation equipment, while in Level 2 scenarios, irrigation equipment takes precedence over fertilization equipment. This establishes a scenario-priority mapping rule. S343: Based on the correlation between environmental factors (for example, the synergistic effect of temperature and humidity on transpiration), combined with plant growth environment requirement data, define equipment parameter synergy thresholds. When the greenhouse temperature is ≥32°C and the humidity is ≥80%, the ventilation fan and dehumidifier are activated simultaneously. When the light intensity is between 1000-3000 lux, the fill light and CO2 generator are activated in conjunction. Generate a parameter synergy threshold comparison table. S344. Structural integration of the results of the logical division of master and slave devices, the scenario priority mechanism, and the parameter coordination threshold is carried out. An IF-THEN rule format is adopted (for example, "IF the drone detects an area with pests and diseases THEN the irrigation system locates the area and applies pesticides at 1.2 times the standard concentration"), and a linkage control rule library is constructed through rule conflict detection (for example, device priority conflicts in different scenarios) and correction.

[0032] The working principle and effect of the above technical solution are as follows: by clarifying the relationship between master and slave devices and setting trigger conditions, the coordinated work of various devices is ensured; by assigning priorities to devices according to environmental scenarios, the system can respond quickly and effectively handle emergencies; by defining parameter thresholds for device collaboration, the system ensures that various devices such as fans, dehumidifiers, and CO2 generators work in a coordinated manner, reducing energy waste, maintaining an optimal environment for plant growth, and promoting more sustainable agricultural practices.

[0033] By integrating equipment operations through IF-THEN logic rules, conflicts between devices are avoided, ensuring that devices are started in order of priority, and improving the stability and reliability of the system; through automatic adjustment and refined equipment settings, errors in manual operation are reduced, ensuring more consistent and accurate crop care, thereby increasing crop yields; by providing real-time data and simulations, the system helps agricultural managers make more accurate decisions, optimize equipment operation, resource allocation and environmental adjustments, and improve the overall effectiveness of agricultural management strategies.

[0034] In one embodiment of the present invention, the step S344 includes: Standardize and analyze the results of the master-slave device logic division, scenario priority mechanism, and parameter coordination thresholds, and extract elements to generate a device-level element table, a scenario priority parameter set, and a coordination threshold element library. Perform association mapping on the device-level element table, scenario priority parameter set, and coordination threshold element library to construct a multi-dimensional rule element association matrix. The trigger conditions and execution actions in the multi-dimensional rule element association matrix are converted into rules to generate initial IF-THEN rule entries (for example, "IF the temperature in the greenhouse is ≥32°C and the humidity is ≥80% THEN the ventilation fan and dehumidifier are started simultaneously"). The initial IF-THEN rule entries are classified and coded to form rule clusters divided by device type and scenario level. Conflict patterns are identified for rule entries in the rule cluster. A rule conflict list is generated through scenario priority comparison and parameter threshold cross-validation (e.g., "priority conflict when ventilation and irrigation equipment are triggered simultaneously in a Level 1 emergency scenario"). Based on the conflict types in the rule conflict list and in conjunction with core agricultural production needs (e.g., prioritizing crop survival over growth regulation), rules are modified and prioritized to obtain a conflict-free rule set. Perform integrity check and redundancy elimination on the conflict-free rule set, supplement edge scenario rule entries (such as the superposition of extreme weather and equipment failure scenarios), and generate an optimized rule set; perform structured storage and index construction on the optimized rule set to form a linkage control rule library that can be dynamically called and expanded.

[0035] The working principle and effect of the above technical solution are as follows: through the standardized analysis and extraction of the logical division of master and slave devices, the scenario priority mechanism, and the collaborative threshold elements, a device hierarchical element table, a scenario priority parameter set, and a collaborative threshold element library are generated to ensure the standardization of device operation and system coordination; the multi-dimensional rule element association matrix is ​​converted into clear IF-THEN rule entries, and classified and coded according to device type and scenario level, making device operation more automated, reducing human intervention, and improving the system's response speed and accuracy; By identifying and correcting conflicting patterns within rule clusters and re-prioritizing them based on core agricultural needs, we avoid equipment conflicts and priority confusion, ensuring smooth system operation in complex scenarios and reducing potential equipment damage risks. By verifying the integrity of rule sets and eliminating redundancies, and supplementing edge scenario rule entries, we improve the rule system to cope with complex scenarios such as extreme weather and equipment failures, thereby enhancing the system's adaptability and operational flexibility. The optimized rule set is structured and indexed to form a linkage control rule base that can be dynamically called and expanded, ensuring that the rules can be efficiently called and providing a good foundation for future system expansion and upgrades. Through the dynamic expansion and intelligent management of the rule base, agricultural equipment can make optimal decisions based on different production scenarios and real-time environmental data, thereby improving the overall intelligence level and adaptability of the agricultural system and further improving the efficiency and sustainability of agricultural production.

[0036] In one embodiment of the present invention, the S4 includes: S41. Collect production target data through agricultural management systems, including yield targets (e.g., rice yield ≥ 600 kg per mu) and quality targets (e.g., tomato soluble solids ≥ 5%). Break down target data by growth stage (e.g., set sub-targets for the seedling stage, flowering stage, and fruiting stage), and quantify them into calculable indicators (e.g., converting yield targets into "number of fruits per plant × weight per fruit") to form a standardized production target dataset. S42. Build a feasibility assessment model based on plant growth trend data. Input the gap between the current growth trend (e.g., plant height growth rate) and the production target, and calculate the probability of achieving each stage's target (e.g., probability of achieving the flowering target = current number of flower buds / target number of flower buds). Analyze constraints (e.g., insufficient light may lead to a decrease in fruiting rate), and generate target feasibility assessment data (including the probability of achieving each stage and key constraints). S43. Based on the target feasibility assessment data and the classified environmental demand data, screen resource demand prediction factors, including basic factors (e.g., crop variety water requirements, fertilization standards during the growing season), dynamic factors (e.g., for every 1°C increase in ambient temperature, irrigation volume increases by 5%), and special factors (e.g., the additional water consumption required for fog simulation in simulated wild environments), to form a prediction factor library; S44. Input the prediction factor library and historical resource consumption data (irrigation water volume and fertilizer application rates over the same period of the past three years) into a deep learning prediction model. Train the model to understand the nonlinear relationship between learning factors and resource consumption. Optimize model parameters through cross-validation (validation set accuracy ≥ 85%). Based on the optimized model, predict the irrigation water volume (split by day / week), fertilizer application rate (by nitrogen, phosphorus, and potassium ratio), energy consumption (equipment operating power), and other data required to achieve production targets, thereby obtaining resource demand forecast data. S45. Combine resource demand forecast data and equipment linkage control simulation data to construct a resource allocation optimization model with the goals of "maximizing resource utilization" (for example, irrigation water utilization ≥ 80%) and "minimizing costs". The constraints are the maximum load of the equipment (for example, the maximum flow of the water pump) and the upper limit of the environmental carrying capacity (for example, the amount of fertilizer applied does not exceed the soil absorption threshold). The particle swarm optimization algorithm (PSO) is used to solve the optimal solution and output the resource allocation quota for each device (for example, the daily pesticide usage of drones and the watering time for each area of ​​the irrigation system) to obtain the optimized resource allocation plan.

[0037] The working principle and effect of the above technical solution are as follows: by breaking down production targets according to growth stages and quantifying them into calculable indicators, a standardized production target data set is formed to ensure the clarity and feasibility of production targets, making agricultural production management more scientific and operational; by combining plant growth trend data to construct a feasibility assessment model, the probability of achieving targets at each stage can be evaluated in real time, helping farm managers to promptly understand the possibility of achieving targets and make corresponding adjustments based on constraints, thereby improving the success rate of achieving production targets.

[0038] By building a predictive factor library and combining historical data with deep learning models, we can accurately predict the resource consumption required for each production target, reducing resource waste and providing a scientific basis for resource planning for the farm. By constructing a resource allocation optimization model and solving it using a particle swarm optimization algorithm, we ensure maximum resource utilization and minimized costs. The optimized resource allocation plan not only improves the efficiency of equipment and resource use, but also reduces unnecessary waste and lowers production costs.

[0039] Leveraging deep learning and particle swarm optimization algorithms, agricultural resource forecasting and allocation become more intelligent, avoiding the limitations of traditional manual forecasting and allocation, making the decision-making process more scientific, accurate, and efficient. By rationalizing resource allocation and maximizing resource utilization, energy and water waste is reduced, promoting environmental protection and sustainable development. Furthermore, optimized cost control makes agricultural production more economical and enhances the sustainability of the overall agricultural system.

[0040] Through real-time data analysis and dynamic model adjustment, the agricultural management system can adapt to changing environmental conditions and production goals, providing farm managers with greater flexibility and ensuring the stability and efficiency of the production process.

[0041] In one embodiment of the present invention, the step S45 includes: Heterogeneous data fusion processing is performed on resource demand forecast data (irrigation water volume, fertilizer application amount, etc.) and equipment linkage control simulation data (equipment collaborative operation parameters, load thresholds, etc.), unifying data dimensions and measurement standards (for example, converting "pesticide application amount" to "liters / hectare" units); core influencing parameters (such as resource consumption rate and equipment collaborative efficiency coefficient) are extracted to generate a standardized fusion parameter matrix; Based on the goals of maximizing resource utilization and minimizing costs, a dual-objective function model was constructed. The resource utilization function included sub-indicators such as the effective absorption rate of irrigation water and the fertilizer conversion rate, while the cost function covered factors such as water resource procurement costs and equipment energy consumption. The constraint boundary values ​​were clearly defined, including the physical limits of the equipment (e.g., the ventilation volume corresponding to the maximum fan speed) and environmental safety thresholds (e.g., the upper limit of soil nitrogen content), forming a constraint parameter library. A hierarchical modeling architecture is used to build a resource allocation optimization model. The bottom layer is the data input layer (connecting and integrating the parameter matrix and the constraint parameter library), the middle layer is the algorithm operation layer (integrating the core module of the particle swarm optimization algorithm), and the top layer is the result output layer (defining the resource allocation quota data format). Key algorithm parameters (such as particle swarm size and maximum number of iterations) are set to complete the model framework. Start the particle swarm optimization algorithm, initialize the particle swarm (each particle represents a set of resource allocation solutions), and calculate the quality of the particles through the fitness function (comprehensively evaluating the dual objective function value and constraint satisfaction). Then perform iterative optimization. The particles update their position and velocity by tracking the individual optimal solution and the global optimal solution. After each round of iteration, solutions that violate the constraints are eliminated until the convergence condition is met (the optimal solution does not change significantly after 5 consecutive rounds of iteration). The optimal solution output by the algorithm is verified in multiple scenarios (e.g., extreme weather, partial equipment failure scenarios) to evaluate the robustness of the solution. Based on the verification results, the resource allocation quota is fine-tuned (e.g., increasing irrigation redundancy by 10% to cope with drought risks) to generate a final optimization solution. The solution is broken down into device-level execution instructions (e.g., "Fertilizer applicator A runs from 9:00 to 11:00 daily, applying 2 kg of nitrogen per mu") to form a resource allocation execution list that can be directly called.

[0042] The working principle and effectiveness of the above technical solution are as follows: By integrating heterogeneous data from resource demand forecasts and equipment linkage control simulation data, not only are data dimensions and measurement standards unified, but core influencing parameters are also extracted to generate a standardized fusion parameter matrix. This greatly improves data comparability, ensuring the consistency and accuracy of the resource management system. Furthermore, the construction of a dual-objective function model considers both maximizing resource utilization (e.g., effective irrigation water absorption rate and fertilizer conversion rate) and minimizing costs (e.g., water procurement costs and equipment energy consumption costs). This goal-oriented optimization approach improves farm resource allocation and reduces overall operating costs.

[0043] The combination of the particle swarm optimization (PSO) algorithm and multi-scenario validation ensures the adaptability of the optimization solution in various environments. For example, the algorithm can adjust to extreme weather conditions or equipment failures, improving the solution's robustness and adaptability, ensuring efficient operation even in uncertain environments. Leveraging a hierarchical modeling architecture, the model automates the entire process from data input to output, reducing manual intervention and improving the intelligence of decision-making. Furthermore, the application of the particle swarm optimization algorithm in resource allocation avoids the inefficiencies and errors associated with traditional manual adjustments.

[0044] When generating the final optimization plan, the model addresses natural risks like drought by adding redundancy, ensuring sufficient resource supply and environmental sustainability. This helps farm managers prevent potential risks and maximize production stability and sustainability. By breaking down the final optimization plan into device-level execution instructions, resource allocation plans can be directly converted into operational instructions, improving execution accuracy and efficiency. The automation and refinement of this process allows resource management to move beyond theoretical models and effectively be implemented in real-world operations.

[0045] In one embodiment of the present invention, the S5 includes: S51. Develop rules for breaking down overall goals, assigning responsibility units based on equipment function (e.g., drones are responsible for pest control, while irrigation systems are responsible for water supply). Based on the contribution weight of each piece of equipment to the production target (e.g., the irrigation system contributes 30% to yield), break down overall yield and quality targets into specific control targets for each piece of equipment (e.g., drones must control pest incidence to ≤5%, while irrigation systems must maintain soil moisture at 60-70%). S52. Based on the decomposition rules and combined with plant growth trend data, quantify the control targets of each device. The drone control targets include "inspection coverage ≥ 98%" and "pesticide spraying error ≤ 5%." The automatic irrigation system control targets include "watering uniformity ≥ 90%" and "response delay ≤ 1 minute." This generates device control target data (including specific values ​​and time limits for achieving the targets). S53. Based on the equipment control target data and the agricultural automation linkage control architecture, define the optimization range of each equipment control parameter, including the drone parameter range (flight altitude 5-10m, spray flow rate 0.3-0.8L / m²); the smart greenhouse parameter range (ventilation fan speed 500-1500r / min, sunshade net opening and closing degree 0-100%). Ensure that the parameter range covers the adjustment space required to achieve the target, forming a parameter constraint range. S54. Input the parameter constraint interval and equipment control target data into the genetic algorithm. Using the target achievement rate as the fitness function, the algorithm selects the optimal control parameter combination (e.g., drone flight altitude 7m + spray flow rate 0.5L / m², greenhouse ventilation fan speed 800r / min) through multiple rounds of iterations through selection (retaining the parameter combinations with the top 30% fitness), crossover (parameter segment recombination), and mutation (randomly adjusting parameter values). The optimized equipment control parameters are then obtained. S55. Based on the optimized device control parameters and linkage rule base, a structured control strategy is generated, including trigger conditions (e.g., cooling linkage is activated when the temperature is > 35°C), execution steps (first opening the sunshade, then starting the fan, and finally adjusting the humidifier), and a feedback mechanism (environmental data is collected every 10 minutes to verify the effectiveness). The effectiveness of the strategy is verified through digital twin scenarios (target achievement rate ≥ 95%), forming the final control strategy. S56. The final control strategy is transmitted to the agricultural automation equipment control system (such as PLC controller, edge computing node) through the Internet of Things gateway to trigger the equipment linkage execution; at the same time, the equipment operation status (such as valve switch, motor speed) and environmental changes (such as temperature and humidity curve) are tracked in real time through the monitoring platform. When a deviation occurs (for example, the actual humidity is 5% lower than the target value), the backup parameter adjustment strategy is automatically called.

[0046] The working principle and effect of the above technical solution are as follows: By dividing responsibility units according to equipment functions, the overall output and quality goals are precisely broken down into specific control targets for each device, allowing each device to perform its tasks in a targeted manner. This not only improves the execution efficiency of each device, but also makes the role and contribution of each device more clear, thereby facilitating the effective allocation of resources. By combining plant growth trend data, quantitative control targets are set for each device, making the operating standards of equipment such as drones and irrigation systems clearer, ensuring that the performance of the equipment can be evaluated through specific indicators, and promoting the efficient operation of agricultural automation equipment. By applying genetic algorithms, we optimize equipment control parameters and select the optimal control combination through multiple rounds of iteration. This process ensures that each device can achieve the best results in actual operation, reduces resource waste, and improves production efficiency. The generated structured control strategy not only includes trigger conditions, execution steps, and feedback mechanisms, but also uses digital twin scenarios to verify the effectiveness of the strategy, ensuring that the control strategy can be smoothly executed under different environmental conditions. Through the linkage of intelligent devices, more accurate and real-time resource allocation and environmental regulation can be achieved, improving the overall level of intelligent agricultural production. IoT technology enables real-time tracking of equipment status and environmental changes, enabling dynamic adjustments to equipment. When environmental deviations occur, the system automatically invokes backup parameter adjustment strategies, ensuring the stability and continuity of agricultural production, reducing human intervention and errors, and improving the ability to respond to emergencies. By integrating equipment control strategies with IoT gateways, automated execution and autonomous decision-making in the production process are achieved, improving work efficiency while reducing human intervention and ensuring a high degree of automation and precise control of the production process.

[0047] The final control strategy and optimization solution can provide a flexible and efficient response plan for agricultural production, especially when facing risks such as extreme climate or equipment failure. It can ensure the stability of agricultural production through automatic adjustment and flexible response, and help promote the development of sustainable agriculture.

[0048] One embodiment of the present invention is an artificial intelligence-based agricultural automation linkage control system, comprising: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement any one of the methods described above.

[0049] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An agricultural automation linkage control method based on artificial intelligence, characterized in that: The method comprises: S1. Acquire agricultural environment data and agricultural automation equipment data, and preprocess the acquired agricultural environment data; perform feature extraction on the preprocessed agricultural environment data to construct an agricultural environment feature vector; perform agricultural environment status analysis based on the agricultural environment feature vector to obtain an agricultural environment status assessment result; perform agricultural automation equipment correlation analysis based on the agricultural environment status assessment result and agricultural automation equipment data to obtain an agricultural automation equipment correlation relationship; perform agricultural automation architecture integration based on the agricultural automation equipment correlation relationship to obtain an agricultural automation linkage control architecture; S2. Perform image recognition processing on agricultural environmental data, extract plant growth parameters based on the recognition results, and use machine learning algorithms to predict the growth trends of plant growth parameters based on the growth models and historical data of different plants to obtain plant growth trend data; perform environmental demand analysis on agricultural environmental data based on the plant growth trend data to obtain plant growth environmental demand data; classify the plant growth environmental demand data to obtain classified environmental demand data; S3. Conduct preliminary control simulation of agricultural automation equipment using the agricultural automation linkage control architecture to obtain preliminary equipment control simulation data; optimize the control strategy using a reinforcement learning algorithm based on the preliminary equipment control simulation data and plant growth environment requirement data to obtain an optimized equipment control strategy; conduct linkage control simulation of the agricultural automation equipment based on the optimized equipment control strategy to obtain equipment linkage control simulation data; S4. Obtain agricultural production target data, conduct a target feasibility assessment on the agricultural production target data based on the plant growth trend data, and obtain target feasibility assessment data; use a deep learning algorithm to predict resource demand based on the target feasibility assessment data and the classified environmental demand data, and obtain resource demand prediction data; combine the resource demand prediction data with the equipment linkage control simulation data to optimize resource allocation and obtain an optimized resource allocation plan; S5. Based on the optimized resource allocation plan and equipment linkage control simulation data, the agricultural production target data is decomposed into targets to obtain equipment control target data; based on the equipment control target data and the agricultural automation linkage control architecture, the genetic algorithm is used to optimize the equipment control parameters to obtain the optimized equipment control parameters; based on the optimized equipment control parameters, the agricultural automation linkage control strategy is generated, the control strategy is transmitted to the agricultural automation equipment control system, and the linkage control task is executed.

2. The agricultural automation linkage control method based on artificial intelligence according to claim 1 is characterized in that: Said S1 comprises: S11. Real-time collection of agricultural environmental data through a distributed sensor network; collection of agricultural automation equipment data through the device Internet of Things interface to form a raw data set; S12. Perform multi-level cleaning on the collected agricultural environmental data to remove outliers caused by sensor failures; interpolate missing values; standardize data of different units and magnitudes; and finally generate a regular agricultural environmental preprocessing dataset; S13. Use machine learning algorithms to extract features from the preprocessed agricultural environmental data, screen out core features that are strongly correlated with plant growth, and compress the high-dimensional environmental data into low-dimensional feature vectors with an information retention rate of ≥90% through principal component analysis. Each vector contains 8-12 core environmental features to form an agricultural environmental feature vector. S14. Based on the agricultural environmental characteristic vectors, a multi-dimensional assessment model is constructed, environmental suitability indicators are set, and the environmental suitability score corresponding to each characteristic vector is calculated using the fuzzy comprehensive evaluation method. The environmental status is graded based on the correlation between historical environmental data and crop growth, generating agricultural environmental status assessment results. S15. Input the agricultural environment status assessment results and agricultural automation equipment data into the deep learning model, construct a device association graph with the equipment as nodes and the collaborative relationships between equipment as edges; learn the mapping relationship between environmental status changes and equipment operation through model training, identify strong association rules between equipment, and obtain the association relationship of agricultural automation equipment; S16. Design a hierarchical linkage control architecture based on the equipment association relationship; determine the data interaction protocols at each level, and integrate them to form an agricultural automation linkage control architecture that can dynamically adapt to environmental changes.

3. The agricultural automation linkage control method based on artificial intelligence according to claim 1 is characterized in that: Said S2 comprises: S21. Regularly collect plant growth images using a high-definition camera; pre-process the images to generate a standardized growth image dataset; S22. Analyze the preprocessed image using a deep learning image recognition model, identify plant growth status and pest and disease conditions through a trained feature extraction network, output recognition result labels, and mark the coordinates of abnormal areas; S23. Based on the image recognition results, quantitative growth parameters are extracted using an image measurement algorithm, and plant height is calculated based on binocular vision. Leaf color is analyzed using the HSV color space. The number of flowers is counted using a target detection frame to form a plant growth parameter set containing multi-dimensional parameters. S24. Inputting the plant growth parameters, the growth model of the corresponding crop, and historical growth data into a machine learning prediction model, training the model to learn the change patterns of the parameters over time, predicting the growth parameter change trends in the next 15 days, and obtaining plant growth trend data; S25. Based on the growth trend data, a growth state-environmental factor correlation model is constructed; combined with the crop physiological characteristics, the environmental requirements that meet the growth trend are obtained, and plant growth environment requirement data is generated; S26. Divide the plant growth environment demand data into dimensions, classify the needs that meet conventional agricultural production as conventional environmental needs, and classify the special conditions that simulate natural ecology as simulated wild environment needs; verify the classification boundaries through the K-means clustering algorithm to obtain the classified environmental demand data.

4. The agricultural automation linkage control method based on artificial intelligence according to claim 3 is characterized in that: Said S23 comprises: Based on the recognition result labels and abnormal area coordinates obtained by S22, feature areas related to plant height, leaf color, and flower number are extracted from the standardized growth image dataset, and the specific location and range of each feature area in the image are determined; The stereo image of the characteristic area obtained by the binocular vision system is used to calculate the spatial distance between the bottom and top of the plant in the characteristic area in combination with the principles of stereo geometry. The quantitative data of the plant height is obtained through the error correction mechanism. The image of the leaf in the feature area is converted from RGB color space to HSV color space, and the mean of the R, G, and B channels in this space is extracted as the quantization parameter of the leaf color; For the flowers in the feature area, the detection frame generated by the target detection algorithm is used to count the number of flowers using a counting algorithm to obtain quantitative data on the number of flowers; The corresponding data are aggregated and organized according to a unified data format to form a plant growth parameter set containing multi-dimensional parameters.

5. The agricultural automation linkage control method based on artificial intelligence according to claim 1 is characterized in that: The S3 includes: S31. Based on the agricultural automation linkage control architecture, set initial control parameters for each agricultural automation device to form an initial control parameter set for the device; S32. In the digital twin agricultural scenario, input the initial control parameters and classified environmental demand data, simulate the operation of different equipment under a single control strategy, set up multiple sets of simulation scenarios, record the equipment operating status and environmental response, and obtain preliminary equipment control simulation data; S33, using the preliminary control simulation data and the plant growth environment requirement data as input, constructing a reinforcement learning model, and continuously optimizing the strategy through interaction with the environment to obtain an optimized device control strategy; S34. Based on the relationship between agricultural automation equipment, formulate equipment linkage control rules, clarify the master-slave equipment logic; set the priority mechanism; define the parameter coordination threshold, and form a linkage control rule library; S35. Load the optimized control strategy and linkage rules in the digital twin scenario, simulate the collaborative operation of multiple devices, and obtain device linkage control simulation data.

6. The agricultural automation linkage control method based on artificial intelligence according to claim 5 is characterized in that: Said S34 comprises: S341. Based on the functional dependency and collaboration strength between devices in the association relationship of agricultural automation equipment, divide the master device and slave device hierarchy, clarify the logical conditions for the master device to trigger the slave device action, and establish a master-slave device trigger mapping table; S342. Analyze the urgency of different agricultural production scenarios, assign operation priorities to equipment according to scenario levels, and form scenario-priority correspondence rules; S343. Based on the correlation between environmental factors and combined with the plant growth environment requirement data, define the equipment parameter coordination threshold value; and generate a parameter coordination threshold comparison table; S344. Structurally integrate the results of the master-slave device logic division, the scenario priority mechanism, and the parameter coordination threshold, adopt the IF-THEN rule format, and build a linkage control rule library through rule conflict detection and correction.

7. The agricultural automation linkage control method based on artificial intelligence according to claim 6 is characterized in that: The S344 includes: Standardize and analyze the results of the master-slave device logic division, scenario priority mechanism, and parameter coordination thresholds, and extract elements to generate a device-level element table, a scenario priority parameter set, and a coordination threshold element library. Perform association mapping on the device-level element table, scenario priority parameter set, and coordination threshold element library to construct a multi-dimensional rule element association matrix. The trigger conditions and execution actions in the multi-dimensional rule element association matrix are converted into rules to generate initial IF-THEN rule entries; the initial IF-THEN rule entries are classified and coded to form rule clusters divided by device type and scenario level; Conflict patterns are identified for rule entries in the rule cluster, and a rule conflict list is generated through scenario priority comparison and parameter threshold cross-validation. Based on the conflict types in the rule conflict list and in combination with the core needs of agricultural production, rules are modified and prioritized to obtain a set of conflict-free rules. Perform integrity check and redundancy elimination on the conflict-free rule set, supplement the edge scenario rule entries, and generate an optimized rule set; perform structured storage and index construction on the optimized rule set to form a linkage control rule library that can be dynamically called and expanded.

8. The agricultural automation linkage control method based on artificial intelligence according to claim 1 is characterized in that: Said S4 comprises: S41. Collect production target data through the agricultural management system, break down the target data by growth stage, and quantify it into calculable indicators to form a standardized production target data set; S42. Combine plant growth trend data to build a feasibility assessment model, input the gap between the current growth trend and the production target, and calculate the probability of achieving the target at each stage; analyze the constraints and generate target feasibility assessment data; S43. Screen resource demand prediction factors based on the target feasibility assessment data and the classified environmental demand data to form a prediction factor library; S44. Input the prediction factor library and historical resource consumption data into the deep learning prediction model, train the model to learn the nonlinear relationship between the factors and resource consumption, and optimize the model parameters through cross-validation; based on the optimized model, obtain resource demand prediction data; S45. Combine resource demand forecast data and equipment linkage control simulation data to build a resource allocation optimization model, use the particle swarm optimization algorithm to find the optimal solution, output the resource allocation quota for each device, and obtain the optimized resource allocation plan.

9. The agricultural automation linkage control method based on artificial intelligence according to claim 1 is characterized in that: Said S5 comprises: S51. Develop rules for breaking down overall goals and divide responsibility units by equipment function. Break down overall output and quality goals into specific control targets for each piece of equipment based on the weight of each piece of equipment's contribution to the production goal. S52. Based on the disassembly rules and combined with the plant growth trend data, the control target of each device is quantified to form the device control target data; S53, based on the equipment control target data and the agricultural automation linkage control architecture, defining the optimization range of each equipment control parameter to form a parameter constraint interval; S54, inputting the parameter constraint interval and the equipment control target data into the genetic algorithm, and screening out the optimal control parameter combination through selection, crossover, and mutation operations after multiple rounds of iteration to obtain the optimized equipment control parameters; S55. Generate a structured control strategy based on the optimized device control parameters and linkage rule base; verify the effectiveness of the strategy through the digital twin scenario to form the final control strategy; S56. The final control strategy is transmitted to the agricultural automation equipment control system through the Internet of Things gateway to trigger the equipment linkage execution; at the same time, the equipment operation status is tracked in real time through the monitoring platform. When deviations occur, the backup parameter adjustment strategy is automatically called.

10. Artificial intelligence-based agricultural automation linkage control system, including: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 9.

Images