Method, system, medium and product for integrating agricultural low-altitude operation and digital service

Abstract

The application relates to the field of low-altitude operation, and discloses an agricultural low-altitude operation and digital service integration method, system, medium and product. The method comprises the following steps: based on historical operation data, crop growth time sequence data recorded in a digital archive of a target land plot and real-time acquired environmental meteorological data, an operation scheme containing optimal flight trajectory parameters and material use parameters is generated; based on geographic information and operation requirements, a plurality of candidate pilots meeting the conditions are screened out from a pilot pool through a preset matching model, and the operation scheme is sent to a pilot terminal of a target pilot; process execution data containing actual flight trajectories and material consumption amounts uploaded by a drone bound to the target pilot in real time are received, the process execution data are compared and analyzed with the operation scheme, a deviation degree index representing operation execution conditions is quantitatively generated, and fund settlement of the target pilot is carried out based on the deviation degree index. The above technical scheme improves the efficiency and quality of agricultural low-altitude operation.

Description

Technical Field

[0001] This application relates to the field of low-altitude operations, and more particularly to a method, system, medium, and product that integrates agricultural low-altitude operations with digital services. Background Technology

[0002] With the development of agricultural modernization, the demand for efficient and precise agricultural operations is increasing. Low-altitude agricultural operations play a crucial role in pesticide spraying and sowing, improving operational efficiency and reducing labor costs. Simultaneously, the application of digital services in agriculture is becoming increasingly widespread, providing data support and decision-making basis for agricultural production, thus contributing to the scientific and intelligent management of agriculture. The combination of low-altitude agricultural operations and digital services is of great significance for improving the quality and efficiency of agricultural production.

[0003] In traditional low-altitude agricultural operations, flight trajectories and material usage are typically planned manually by operators based on experience. The selection of drone pilots often relies on manual contact or simple information matching, lacking a scientific and precise screening mechanism. During operations, it is difficult to acquire and analyze operational data in real time, making timely and accurate evaluation of operational effectiveness impossible. Furthermore, significant differences in the skill levels and experience of different operators lead to inconsistent operational quality. In addition, traditional methods lack effective digital management of the operational sites, making it difficult to fully utilize historical operational data and environmental meteorological data to optimize operational plans. Summary of the Invention

[0004] This application provides a method, system, medium, and product that integrates agricultural low-altitude operations with digital services. It combines agricultural low-altitude operations with digital services, accurately generates operation plans based on multiple data sources, efficiently matches drone pilots, quantifies operation deviations through comparative analysis, and settles funds reasonably, thereby improving the efficiency and quality of agricultural operations.

[0005] Firstly, this application provides an integrated method for agricultural low-altitude operations and digital services, the method comprising:

[0006] In response to receiving a job request containing geographic information of a target plot, create or update a digital profile for the target plot;

[0007] Based on the historical operation data, crop growth time series data, and real-time environmental meteorological data recorded in the digital archive, an operation plan containing optimal flight trajectory parameters and material usage parameters is generated.

[0008] Based on the geographic information and job requirements, multiple candidate pilots who meet the conditions are selected from the pilot pool through a preset matching model, and job invitations containing a summary of the job plan are pushed to the multiple candidate pilots. In response to the target pilot's confirmation of the job invitation, the job plan is sent to the target pilot's pilot terminal. The target pilot is any one of the multiple candidate pilots.

[0009] The system receives real-time process execution data, including actual flight trajectory and material consumption, uploaded by the drone bound to the target pilot. The process execution data is compared and analyzed with the operation plan to quantify and generate a deviation index that characterizes the operation execution. Based on the deviation index, the system performs financial settlement for the target pilot.

[0010] By employing the aforementioned technical solutions, optimal operational plans are generated through digital land archives and AI algorithms. Intelligent matching is then performed based on geographical location, pilot qualifications, and experience, transforming the process from manual experience-driven decision-making to data-driven precision scheduling. This significantly improves task matching efficiency and operational quality. From intelligent order dispatch and plan issuance to real-time data transmission from drones and automatic comparison and analysis to generate quantitative deviation indicators, a fully online and automated closed-loop management system is achieved, reducing manual intervention and enhancing operational transparency and traceability. Through digital pilot selection, standardized operational plan generation, objective recording of process data, and a settlement mechanism based on quantitative indicators, a verifiable and measurable agricultural low-altitude operation service standard system has been established, strengthening the foundation of trust between service providers and users.

[0011] In some embodiments, the step of selecting multiple candidate pilots from the pilot pool based on the geographic information and job requirements using a preset matching model, and then sending job invitations containing a summary of the job plan to the multiple candidate pilots, specifically includes:

[0012] The task requirements are analyzed, and key task features are extracted to construct a task demand vector. The key task features include the urgency of the task, the type of target crop, the type of target pest or disease, or the agricultural process.

[0013] Using the geographical location of the target plot as the center, select online drone pilots currently within a preset service radius from the drone pilot pool to form an initial candidate set;

[0014] For each candidate pilot in the initial candidate set, historical operation data is retrieved from the pilot database on the cloud server, and based on the key operation features, a capability feature vector for each candidate pilot is generated. The similarity between the capability feature vector and the operation requirement vector is calculated as the core matching score.

[0015] Based on the core matching score, combined with the pilot's qualification certification level and historical service evaluation score, a dynamic comprehensive recommendation index is calculated for each candidate pilot. The candidate pilots in the initial candidate set are ranked according to the dynamic comprehensive recommendation index, and the job invitation is pushed to a preset number of candidate pilots ranked first.

[0016] By employing the aforementioned technical solution, and quantifying job requirements and drone pilot capabilities into feature vectors and calculating similarity, the system achieves intelligent and precise matching based on multi-dimensional features such as specific job content, crop type, and urgency. This improves the rationality of task allocation and job success rate. A multi-level screening and ranking mechanism, along with a dynamic comprehensive recommendation index, optimizes the scheduling priority of service resources (drones) in real time according to different job requirements, ensuring the most suitable drone pilots respond first, thus improving the overall scheduling efficiency and resource utilization of the system. The matching model not only relies on real-time status (geographical location) but also deeply mines and applies drone pilots' historical job data and long-term service evaluations (qualifications, ratings), enabling the prediction and assurance of future service quality. This not only helps farmers select more reliable service providers but also provides a data foundation for establishing a long-term credit system for drone pilots.

[0017] In some embodiments, generating an operational plan that includes optimal flight trajectory parameters and material usage parameters based on historical operational data, crop growth time-series data, and real-time acquired environmental meteorological data recorded in the digital archive specifically includes:

[0018] By integrating the crop growth time-series data, historical operation data, and real-time environmental meteorological data recorded in the digital archive, the personalized demand characteristics of this operation are extracted. The personalized demand characteristics include at least the crop health level, the probability of occurrence of specific pests and diseases, and operation preferences based on historical results.

[0019] Based on the current date, the planting information of the target plot, the crop health level, and the probability of occurrence of the specific pests and diseases, determine the specific growth stage of the crop and the corresponding agricultural needs.

[0020] Based on the specific growth stage and the crop health level, the corresponding standard material type, basic usage range and benchmark operation mode are retrieved and called from the agricultural knowledge base in the cloud server;

[0021] Based on the operational preferences, the basic usage range is calibrated to generate material usage parameters for the target plot;

[0022] Based on the elevation data and crop canopy height information of the target plot, and combined with the flight trajectories executed on the target plot in the historical operation data, a basic flight trajectory is planned. According to real-time or forecasted environmental meteorological data, the flight altitude, speed and route spacing of the basic flight trajectory are adjusted to generate the optimal flight trajectory parameters.

[0023] The material usage parameters are integrated with the optimal flight trajectory parameters, and the corresponding operation time window is bound to generate the operation plan.

[0024] By employing the aforementioned technical solution and integrating multi-source information such as historical data of the plot, real-time crop growth, and environmental meteorological data, the extraction of "personalized demand characteristics" (such as health level, probability of pests and diseases, and historical preferences) for specific plots and crops has been achieved for the first time. This transforms the operational plan from a static standard template into a dynamic and precise prescription adapted to the current state of the plot. The plan generation process combines deterministic rules (based on growth stage and knowledge base retrieval standard procedures) with data-driven analysis (based on historical preference to calibrate dosage and on environmental adjustment trajectory), ensuring both the agricultural scientific nature of the plan and incorporating localized practical wisdom, thereby improving the effectiveness and acceptability of the plan. In trajectory planning, not only static geographical information of the plot (elevation, canopy) is considered, but also historical operational experience is comprehensively taken into account, and flight parameters are dynamically fine-tuned based on real-time / forecasted environmental meteorological data (such as wind speed, wind direction, and temperature). This gives the operational plan environmental adaptability, aiming to improve the stability, safety, and uniformity of application / operation effects in complex and variable field environments.

[0025] In some embodiments, calibrating the basic usage range based on the operational preferences to generate material usage parameters for the target plot specifically includes:

[0026] The job preference is encoded as a multidimensional preference control vector, where each dimension of the preference control vector corresponds to a calibration tendency. The magnitude of the calibration tendency indicates the tendency intensity. The calibration tendency includes effect enhancement tendency, cost saving tendency, and risk avoidance tendency.

[0027] The standard material types, basic dosage ranges, and benchmark operation modes retrieved from the agricultural knowledge base are analyzed to identify the controllable flexible parameters and allowable disturbance ranges. The flexible parameters include the proportion of mixed agents, the average application rate per acre, and the flight altitude during operation.

[0028] Based on the preference control vector, a perturbation strategy is generated for each of the flexible parameter items. The perturbation strategy defines the direction of change of the parameter value and the relationship of the change function within the perturbation interval.

[0029] Based on the disturbance strategy, all the flexible parameter items are calculated and adjusted synchronously to generate the material usage parameters.

[0030] By employing the above technical solution, the subjective operational preferences of farmers or managers are encoded into quantitative control vectors of specific dimensions. Abstract and conflicting value objectives are transformed into calculable and weighable engineering parameters, enabling the generated solution to clearly reflect and balance various practical demands. Unlike the traditional method of manually modifying parameters one by one, this embodiment uses a perturbation strategy to simultaneously and collaboratively calculate and adjust multiple flexible parameters (such as pesticide-liquid ratio, dosage per acre, and flight altitude). This systematic approach considers the interactions between parameters, thereby generating a logically self-consistent and overall optimal material usage plan, avoiding the pitfalls of focusing on one aspect at the expense of others.

[0031] In some embodiments, adjusting the flight altitude, speed, and route spacing of the basic flight trajectory based on real-time or forecasted environmental meteorological data to generate optimal flight trajectory parameters specifically includes:

[0032] With operational efficiency, spray uniformity, and energy consumption as optimization objectives, and the environmental meteorological data, crop canopy height information, and UAV performance parameters as constraints, a multi-objective function for trajectory parameter optimization is constructed.

[0033] Within the framework of the basic flight trajectory, multiple parameter combinations are generated for the three key parameters of flight altitude, flight speed and route spacing within their respective feasible value ranges, forming a candidate set of trajectory parameters;

[0034] Substitute each set of parameters in the candidate trajectory parameters into the multi-objective function to perform flight operation simulation, calculate their respective operation efficiency score, uniformity score and energy consumption score, and use a multi-objective optimization algorithm to solve for the Pareto optimal solution set under the constraints based on the operation efficiency score, the uniformity score and the energy consumption score.

[0035] From the Pareto optimal solution set, the final solution is selected according to the priority strategy of the current operation, and the optimal flight trajectory parameters are determined based on the final solution.

[0036] By adopting the above technical solution, the three core performance indicators of "operational efficiency," "spraying uniformity," and "energy consumption," which are mutually restrictive, are clearly defined as optimization objectives. Through the construction of a multi-objective function and the solving of the Pareto optimal solution set, the solution generation process no longer pursues extreme values ​​for a single indicator (such as maximum speed), but instead seeks a scientific and balanced optimal trade-off point among multiple key performance indicators, fundamentally improving the overall effectiveness of the operation plan. Simulation technology is used to pre-evaluate the operational effects of different parameter combinations (height, speed, spacing) under specific environmental constraints (weather, crop, UAV performance), and a multi-objective optimization algorithm is applied to automatically search for the optimal solution set. This process transforms complex, trial-and-error field decision-making that relies on expert experience into automated, precise calculations based on models and algorithms, which are repeatable and verifiable, greatly improving the scientific rigor and efficiency of decision-making.

[0037] In some embodiments, substituting each set of parameters from the candidate trajectory parameters into the multi-objective function for flight operation simulation includes:

[0038] Based on the model of the drone nozzle and the current operating parameters, a swarm of droplet particles with preset initial velocity and release angle is generated in a simulation environment.

[0039] Based on the crop canopy height information and three-dimensional structural model, and combined with the real-time environmental meteorological data, a canopy region fluid dynamic field is constructed in the simulation space. The canopy region fluid dynamic field characterizes the changes in wind speed and turbulence intensity inside and around the canopy with spatial location.

[0040] The fog droplet particle swarm is placed in the hydrodynamic field of the canopy region, and a simulation based on the coupling of computational fluid dynamics and discrete element method is run to calculate the trajectory of each fog droplet particle.

[0041] The number and size distribution of fog droplets adhering to a unit area of ​​the leaf are statistically analyzed to calculate the simulated deposition density and simulated coverage. The proportion of fog droplets that drift outside the target area is statistically analyzed as the simulated drift rate. Based on the simulated deposition density, the simulated coverage, and the simulated drift rate, the uniformity score and the energy consumption score are calculated.

[0042] By employing the aforementioned technical solution and integrating computational fluid dynamics and the discrete element method, the entire physical process from droplet formation and movement in turbulent wind fields to final deposition on complex canopy structures was meticulously simulated. This allows for a leap from relying on macroscopic empirical formulas to computable predictions based on microscopic physical mechanisms in the evaluation of operational quality (such as uniformity and drift rate), significantly improving the accuracy and scientific rigor of the assessment. By constructing a "canopy region fluid dynamics field" incorporating a three-dimensional structural model of the specific crop and real-time environmental wind field data, the simulation can quantitatively analyze how complex factors such as wind speed, turbulence, and canopy density collectively influence droplet penetration, distribution, and drift. This provides a powerful analytical tool for understanding and optimizing operational performance under real, non-ideal environments.

[0043] In some embodiments, comparing and analyzing the process execution data with the work plan to quantify and generate a deviation index characterizing the work execution status specifically includes:

[0044] The expected operational effect features are extracted from the operational plan, and the post-operation effect image of the target plot is obtained after the operation is completed. The expected operational effect features include the expected pesticide coverage distribution map, the expected pest and disease control target area map, or the expected vegetation index improvement area map.

[0045] The actual flight trajectory is spatially superimposed with the expected flight trajectory in the operation plan to calculate the trajectory overlap, the area of ​​the missed spray area, and the area of ​​the re-spray area.

[0046] The actual effect features extracted from the post-operation effect image are compared with the expected operation effect features to calculate the coverage difference between the actual coverage rate and the expected coverage rate, and / or calculate the overlap rate between the actual control area and the expected control target area.

[0047] Based on the trajectory overlap, the area of ​​the missed spray area, the area of ​​the re-spray area, the coverage difference and / or the prevention overlap rate, the deviation index is generated by weighted calculation according to preset rules.

[0048] By employing the aforementioned technical solution, and combining spatial trajectory analysis (overlap rate, missed / oversprayed area) with final effect comparison (coverage difference, control overlap rate), a complete evaluation chain from execution process to operational results is constructed, achieving a comprehensive, refined, and quantitative assessment of operational quality. By introducing and comparing post-operation effect images with expected operational effect characteristics, the core of acceptance is elevated from traditional "process compliance checks" (such as whether the preset flight path was completed) to "goal achievement verification" (such as whether the expected coverage or control effect was achieved). This makes service quality evaluation closer to the ultimate goal of agricultural production, promoting a potential shift in service models from "pricing per application / acre" to "pricing per effect."

[0049] In a second aspect, embodiments of this application provide a computer system including a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the steps of the method described in any possible implementation of the first aspect.

[0050] Thirdly, embodiments of this application provide a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method described in any possible implementation of the first aspect.

[0051] Fourthly, embodiments of this application provide a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described in any possible implementation of the first aspect.

[0052] It is understood that the computer system provided in the second aspect, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided in this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0053] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0054] 1. Transforming physical farmland into a calculable and traceable digital twin provides a unique and authoritative data foundation for all subsequent precise decision-making. Each operation enriches this archive, continuously increasing its historical value.

[0055] 2. By comprehensively utilizing historical data, real-time crop growth data, and environmental data, customized operational prescriptions are generated for specific problems at specific plots and times, elevating agricultural production decisions from relying on personal experience to data-driven scientific decisions, aiming to maximize operational effectiveness and minimize resource waste and environmental impact;

[0056] 3. Through algorithm models, the most suitable job tasks (orders) are automatically pushed to the most suitable service providers (drones) and executable technical instructions (job plans) are directly issued, which greatly reduces the time and cost of information search, negotiation and technical disclosure, and improves the operational efficiency of the entire service ecosystem.

[0057] 4. Through real-time data feedback and automated comparison, the performance of tasks is objectively quantified into a deviation index, and this quality index is directly linked to service compensation. This establishes a positive incentive mechanism where task quality determines economic returns, technically ensuring service quality and financial security, and building a digital trust system. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating the method for integrating low-altitude agricultural operations with digital services in an embodiment of this application.

[0059] Figure 2 This is a schematic diagram of the architecture of the integrated platform in the embodiments of this application;

[0060] Figure 3 This is a schematic diagram of the client interface in an embodiment of this application;

[0061] Figure 4 This is a schematic diagram of the drone operator's interface in an embodiment of this application;

[0062] Figure 5 This is a schematic diagram of an exemplary hardware structure of a computer system in an embodiment of this application. Detailed Implementation

[0063] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0064] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0065] The following is combined Figure 1 The method of the embodiments of this application will be described below.

[0066] Figure 1 This is a flowchart illustrating the integrated method of agricultural low-altitude operations and digital services in an embodiment of this application, as shown below. Figure 1 As shown, the method includes the following steps:

[0067] S101. In response to receiving a job request containing geographic information of a target plot, create or update a digital profile for the target plot;

[0068] S102. Based on the historical operation data, crop growth time series data and real-time acquired environmental meteorological data recorded in the digital archive, generate an operation plan that includes optimal flight trajectory parameters and material usage parameters.

[0069] S103. Based on the geographic information and job requirements, multiple candidate pilots who meet the conditions are selected from the pilot pool through a preset matching model, and a job invitation containing a summary of the job plan is pushed to the multiple candidate pilots. In response to the target pilot's confirmation operation of the job invitation, the job plan is sent to the pilot terminal of the target pilot, where the target pilot is any one of the multiple candidate pilots.

[0070] S104. Receive process execution data uploaded in real time by the drone bound to the target pilot, which includes the actual flight trajectory and material consumption. Compare and analyze the process execution data with the operation plan to quantify and generate a deviation index that characterizes the operation execution. Settle funds for the target pilot based on the deviation index.

[0071] Figure 2 This is a schematic diagram of the integrated platform architecture in the embodiments of this application, as shown below. Figure 2 As shown, the integrated platform includes a client, a drone operator's app, a management backend, and third-party fund custody.

[0072] Figure 3 This is a schematic diagram of the client interface in an embodiment of this application, such as... Figure 3 As shown, the client can view pilot qualifications, historical job evaluations, nearest job radius, and online consultation in real time through the pilot screening system. The client can visualize orders, such as uploading plot location, area, and crop variety; scheduling job times; automatically generating 3D aerial survey maps; matching pesticide application plans; and supporting online price comparison.

[0073] Figure 4 This is a schematic diagram of the drone operator's interface in an embodiment of this application, such as... Figure 4 As shown, the drone operator's app can push high-priority orders based on LBS location through the intelligent dispatch system, display customer historical transaction reviews, show prepayment guarantee indicators, and also support online consultation. The app can also manage operations, such as viewing the content of the operation target, confirming the operation time, automatically generating 3D aerial survey maps, matching pesticide application plans, and uploading operation videos in real time for acceptance review.

[0074] The following is combined Figures 2-4 The technical solution of this application is described.

[0075] When farmers place orders online through the integrated platform's client (farmer / cooperative), they need to upload or confirm the boundary information of the plot to be treated (e.g., by selecting on a map or uploading boundary coordinates), and the system will then receive the job request. Creating or updating digital profiles is the data cornerstone of the platform's operation. These digital profiles integrate the following multi-source data: job data, soil data, crop growth data, production data, and geographic information. Job data: Historical drone operation trajectories, pesticide dosage, and duration. Soil data: Soil moisture. Crop growth data: Crop growth data collected by the drone's multispectral sensors. Production data: Planting area and historical yield reported by the farmer. Geographic information: After uploading the plot location, the system generates a 3D aerial survey map. This digital profile is not static; it is updated with each job. For example, after a plant protection operation, the pesticide dosage, operation time, and crop images at that time (which may reflect pest and disease conditions) will be recorded in the plot's profile, forming historical job data and crop growth time-series data, providing a basis for future decision-making. The platform leverages its established data foundation, combined with real-time data, to generate a highly customized, directly executable operational plan for this task. Based on the crop's (e.g., pomelo) growth cycle, the platform pre-sets standardized operational nodes (e.g., 15 operations in total, including budding and flowering stages). The system determines the current stage (e.g., flowering stage) based on the current date and crop growth. It then calls upon the corresponding standardized plan for that stage, such as using a combination of spirotetramat and bifenazate with a basic dosage during flowering. Fine-tuning is then performed based on historical results (operational preferences) in the digital archive and current pest and disease warnings to generate the final formulation and dosage per acre. Based on a 3D aerial survey map (including elevation and canopy information) generated for the target plot, a basic contour flight path is planned. Further adjustments to flight altitude and speed are made based on real-time meteorological data (e.g., wind speed and direction) to generate the optimal flight path. The matching logic of the matching model is as follows: Based on LBS (Location-Based Services), priority is given to displaying drone operators within a 3-kilometer radius to ensure timely response. During the selection process, drone operators are required to have national certification and demonstrate their historical five-star ratings and service cases. Hard criteria can be set, such as "rating ≥ 4.5 stars". For specific crops (such as pomelos), local drone operators with "≥ 10 operations within a 3-kilometer radius" are given priority. The system pushes job invitations containing summaries such as plot location, area, crop type, and operation stage to the selected candidate drone operators' terminals. After any candidate drone operator confirms the order, a detailed operation plan (including precise flight maps, pesticide dosage ratios, and key operational points) is sent to the corresponding drone operator's terminal to guide their operation. During drone operations, operation videos and data (trajectory, actual pesticide consumption) are uploaded in real time. The monitoring backend monitors the county's plant protection progress and pesticide usage in real time, and uses AI to verify operation coverage and missed spraying rates. Here, "coverage rate" and "missed spraying rate" are components of the deviation index.Specifically, by overlaying the actual flight trajectory with the planned trajectory, the overlap rate can be calculated, and areas of missed or excessive spraying can be identified. The actual pesticide coverage rate can be calculated by analyzing operation videos or post-application images using AI. Acceptance is determined jointly by "farmer satisfaction evaluation (70% weight) + AI image recognition (30% weight) quality verification." These two components combine to form a quantifiable deviation index. Farmers prepay fees to a bank escrow account, and after acceptance, funds are released to the drone operator's account within 24 hours. If the deviation index indicates unsatisfactory operation quality (e.g., coverage rate not meeting standards), it can trigger a suspension of payment, deduction of funds, or dispute resolution procedures.

[0076] In some embodiments, the step of selecting multiple candidate pilots from the pilot pool based on the geographic information and job requirements using a preset matching model, and then sending job invitations containing a summary of the job plan to the multiple candidate pilots, specifically includes:

[0077] The task requirements are analyzed, and key task features are extracted to construct a task demand vector. The key task features include the urgency of the task, the type of target crop, the type of target pest or disease, or the agricultural process.

[0078] Using the geographical location of the target plot as the center, select online drone pilots currently within a preset service radius from the drone pilot pool to form an initial candidate set;

[0079] For each candidate pilot in the initial candidate set, historical operation data is retrieved from the pilot database on the cloud server, and based on the key operation features, a capability feature vector for each candidate pilot is generated. The similarity between the capability feature vector and the operation requirement vector is calculated as the core matching score.

[0080] Based on the core matching score, combined with the pilot's qualification certification level and historical service evaluation score, a dynamic comprehensive recommendation index is calculated for each candidate pilot. The candidate pilots in the initial candidate set are ranked according to the dynamic comprehensive recommendation index, and the job invitation is pushed to a preset number of candidate pilots ranked first.

[0081] The system performs natural language processing or structured parsing on orders submitted by farmers. It extracts key operational features, including the urgency level, target crop type, target pest / disease type, and agricultural procedure. Urgency level: For example, sudden pest control (such as locust plagues) is considered urgent, while routine periodic plant protection is considered normal. This directly affects subsequent order priority and drone operator response time requirements. Target crop type: Such as Guanxi pomelo, rice, and wheat; different crops have significantly different canopy structures, common pests / diseases, and operational procedures. Target pest / disease type / agricultural procedure: Such as "red spider mite control," "fruit protection during flowering," and "harvesting and transport." This determines the required expertise, pesticide knowledge, and equipment type. These key operational features are transformed into a structured, computable data vector. For example, [urgency level: 0.9, crop type: pomelo, agricultural procedure: flowering control]. This transforms abstract requirements into quantifiable inputs that the algorithm can understand. A circular service radius (e.g., 3 kilometers) is defined on the map centered on the latitude and longitude of the target plot. Only drone pilots whose current GPS location is within this radius are selected to ensure rapid response and reduce empty-run costs; only pilots who are logged into the platform, have online devices, and are marked as "available for orders" are selected. This is the first layer of efficient filtering, narrowing the global "pilot pool" to an "initial candidate set" with basic service conditions, significantly reducing the amount of data for subsequent complex calculations, and prioritizing service timeliness. Historical job data for each candidate pilot is retrieved from the pilot database, including but not limited to: a list of successfully completed crop types, records of pest / disease types / agricultural processes handled, and the scale and complexity of each job. Based on the same "key job feature" dimension, a capability vector is generated for each pilot. For example, for a job requirement of "controlling red spider mites during pomelo flowering," the system analyzes pilot A's historical data: he has completed 50 "pomelo" jobs, 30 of which involved the "flowering period," and 15 successfully "controlled red spider mites." Therefore, his capability vector will have a high numerical weight in the corresponding dimension. Cosine similarity, Euclidean distance, or other machine learning models are used to calculate the similarity between the pilot's capability feature vector and the job requirement vector. A higher score indicates a better match between the pilot's past experience and the requirements of the current task. This directly predicts the probability of the pilot completing the task with high quality. Certification level: such as "National Certification" or "Provincial Certification". Historical service evaluation score: the average of five-star reviews from previous farmers. This reflects the market's recognition of their service attitude and reliability, serving as a credit indicator. A weighted formula is used to comprehensively calculate the core matching score, qualification level coefficient, and historical score. The weights of the formula can be dynamically adjusted according to the platform's strategy (e.g., prioritizing qualifications initially, and prioritizing user reviews and matching in a mature stage). All initial candidate pilots are ranked in descending order based on a dynamic comprehensive recommendation index. Task invitations are simultaneously or sequentially sent to the top 3 or 5 pilots.The invitation information includes a summary of the job plan (location, area, crop, core requirements, estimated compensation) to facilitate quick decision-making. It typically employs either a "bidding" or "assignment" model. In the bidding model, multiple drone pilots receive an invitation, and the first to confirm wins the bid. In the priority assignment model, the platform may directly assign the order to the pilot with the highest search volume and require them to respond within a certain timeframe.

[0082] In some embodiments, generating an operational plan that includes optimal flight trajectory parameters and material usage parameters based on historical operational data, crop growth time-series data, and real-time acquired environmental meteorological data recorded in the digital archive specifically includes:

[0083] By integrating the crop growth time-series data, historical operation data, and real-time environmental meteorological data recorded in the digital archive, the personalized demand characteristics of this operation are extracted. The personalized demand characteristics include at least the crop health level, the probability of occurrence of specific pests and diseases, and operation preferences based on historical results.

[0084] Based on the current date, the planting information of the target plot, the crop health level, and the probability of occurrence of the specific pests and diseases, determine the specific growth stage of the crop and the corresponding agricultural needs.

[0085] Based on the specific growth stage and the crop health level, the corresponding standard material type, basic usage range and benchmark operation mode are retrieved and called from the agricultural knowledge base in the cloud server;

[0086] Based on the operational preferences, the basic usage range is calibrated to generate material usage parameters for the target plot;

[0087] Based on the elevation data and crop canopy height information of the target plot, and combined with the flight trajectories executed on the target plot in the historical operation data, a basic flight trajectory is planned. According to real-time or forecasted environmental meteorological data, the flight altitude, speed and route spacing of the basic flight trajectory are adjusted to generate the optimal flight trajectory parameters.

[0088] The material usage parameters are integrated with the optimal flight trajectory parameters, and the corresponding operation time window is bound to generate the operation plan.

[0089] Crop growth time-series data (from historical inspections by multispectral drones): By analyzing the changing trends of leaf area index, normalized vegetation index, etc., the crop health level (e.g., healthy, sub-healthy, latent disease period) is assessed. Historical operation data: Analysis of the pesticides used, dosages, and effect feedback (assessed through later yield or disease control rate) for similar operations in the past on this plot (e.g., pest and disease control in the same period of previous years). This forms an operation preference based on historical effects (e.g., formula A was more effective than formula B last time; a certain drone operator achieved higher coverage in a specific area). Real-time environmental meteorological data (from weather stations or forecasts): Combining current temperature and humidity, recent precipitation, and future forecasts, the probability of occurrence of specific pests and diseases (e.g., anthracnose, spider mites) is calculated using pest and disease occurrence models. Based on planting information (e.g., planting date and variety of pomelo) and the current date, the theoretical growth stage (e.g., budding stage, flowering stage, fruit expansion stage) is determined. Crop health level and pest and disease occurrence probability will correct this judgment. For example, theoretically it's the fruit expansion stage, but the crop's health level is poor and the probability of anthracnose is high. The system might classify it as a "critical period for disease control during the fruit expansion stage," and the agricultural needs would shift from simple nutrient supplementation to a comprehensive need of "nutrition + treatment + protection." This results in a precise, status-based growth stage label and the core agricultural objectives derived from it (such as "flower and fruit preservation," "disease control," and "nutrient supplementation"). The system then queries the cloud-based agricultural knowledge base using the determined "specific growth stage" and "crop health level" as indexes. The knowledge base includes standard material types, basic dosage ranges, and benchmark operating modes. Standard material types: Recommended pesticide / fertilizer types for this stage (e.g., "spirotetramat + bifenazate" recommended during the flowering period). Basic dosage range: Upper and lower limits of the standard dosage per acre (e.g., 30-50 ml of spirotetramat per acre). Benchmark operating modes: Includes recommended drone flight modes (e.g., "Z" shaped flight path), spray width, basic flight altitude, and other general parameters. If historical data indicates that a certain pesticide is more effective at the upper-middle range of its dosage range without causing phytotoxicity under similar conditions, the system will tend to calibrate upwards. If historical data indicates that the plot is sensitive to a certain pesticide, or that minor burns occurred after the last operation, the system will tend to calibrate downwards. If the preference points to a specific pilot's operating pattern (e.g., a pilot prefers a slower flight speed to ensure deposition), this information will be retained for reference in the next trajectory planning step. A fine-tuned final material usage parameter (precise pesticide ratio and dosage per acre) is obtained for this plot. Planning the basic flight trajectory: Input elevation data (from 3D aerial survey maps) and crop canopy height information to plan contour lines that ensure safe distances. Referencing historical flight trajectories, priority is given to flight patterns that have performed well on this plot in the past (e.g., specific flight path spacing or turning logic) to form the initial basic flight trajectory. Generating optimal flight trajectory parameters: The core input is real-time or forecast environmental meteorological data, especially wind speed and direction.Flight Altitude: Lower the altitude appropriately in strong winds to reduce drift; raise it appropriately in calm conditions and with dense canopy cover to increase droplet penetration. Flight Speed: Increase the speed appropriately in high winds to ensure operational efficiency, but this must be coordinated with altitude and nozzle flow rate to ensure the required deposition rate per unit area is achieved. Flight Line Spacing: In strong crosswinds, it may be necessary to reduce flight line spacing and adjust flight line angles to compensate for wind-induced spray width deviation and avoid missed spraying. Obtain a set of optimal flight trajectory parameters adapted to the current environmental conditions, aiming to ensure operational effectiveness (uniformity, coverage) while also considering operational safety and efficiency. Operation Time Window: Based on weather forecasts (avoiding periods before and after rainfall), agricultural requirements (such as pollen activity time), and available pilots, suggest or specify a specific operation time period (e.g., "It is recommended to operate between 09:00 and 11:00 on May 10th"). Encapsulate all the above outputs and the operation time window into an executable instruction package to obtain the operation plan.

[0090] In some embodiments, calibrating the basic usage range based on the operational preferences to generate material usage parameters for the target plot specifically includes:

[0091] The job preference is encoded as a multidimensional preference control vector, where each dimension of the preference control vector corresponds to a calibration tendency. The magnitude of the calibration tendency indicates the tendency intensity. The calibration tendency includes effect enhancement tendency, cost saving tendency, and risk avoidance tendency.

[0092] The standard material types, basic dosage ranges, and benchmark operation modes retrieved from the agricultural knowledge base are analyzed to identify the controllable flexible parameters and allowable disturbance ranges. The flexible parameters include the proportion of mixed agents, the average application rate per acre, and the flight altitude during operation.

[0093] Based on the preference control vector, a perturbation strategy is generated for each of the flexible parameter items. The perturbation strategy defines the direction of change of the parameter value and the relationship of the change function within the perturbation interval.

[0094] Based on the disturbance strategy, all the flexible parameter items are calculated and adjusted synchronously to generate the material usage parameters.

[0095] The system quantifies and encodes the collected operational preferences. Several core calibration tendency dimensions are established, which may have trade-offs with each other. For example, enhancement tendency: a high weight indicates that the primary goal of this operation is to pursue the best control effect or crop response, with reduced sensitivity to cost. Cost-saving tendency: a high weight indicates that, while ensuring basic effectiveness, the consumption of pesticides, fertilizers, or energy should be minimized as much as possible. Risk-averse tendency: a high weight indicates that potential risks such as pesticide damage, drift, and crop stress are prioritized, favoring a more conservative approach. Each tendency is assigned a numerical value (e.g., between 0 and 1) to represent its strength. For example, the preference vector could be [enhanced effect: 0.8, cost-saving: 0.2, risk-averse: 0.6]. This indicates that this operation highly prioritizes effectiveness, moderately avoids risk, and does not consider cost savings much. Flexible parameter items are identified: not all parameters can be adjusted arbitrarily. The system identifies those parameters that have a significant impact on the final effect and can be flexibly varied within a certain range. For example, the ratio of mixed agents: the mixing ratio of two recommended agents (e.g., A and B) (e.g., A:B can be adjusted from 3:7 to 7:3). The average application rate per acre: the specific value within the baseline application rate range (e.g., 30-50 ml / acre). The flight altitude during operation: adjustments near the baseline altitude (e.g., 1.5-2.5 meters above the canopy). Determining the disturbance range: setting a scientifically permissible adjustment range for each flexible parameter. This range is predefined based on agronomic knowledge, agent characteristics, equipment performance, and safety regulations, ensuring the safety of adjustments. Direction of change: determining how each flexible parameter should change in response to different tendencies. For example, when the "effect enhancement tendency" is strong, the average application rate per acre may need to be adjusted towards the upper limit; when the risk aversion tendency is strong, it may need to be adjusted towards the lower limit to prevent phytotoxicity. Functional relationship of change: defining the mathematical relationship of the adjustment. This can be a simple linear weighting or a more complex nonlinear function. For example, the final dosage per acre, D, can be calculated as: D = D_base + w_effect × Δ_max - w_risk × Δ_min, where D_base is the base dosage, w_effect and w_risk are the weights of the effect bias and risk bias, and Δ_max and Δ_min are the maximum upward and downward adjustments (not exceeding the perturbation range). The strategy also needs to consider the coupling relationships between parameters. For example, increasing the flight altitude (for risk avoidance) may require simultaneous fine-tuning of the flight speed or a slight increase in dosage per acre to compensate for a potential decrease in droplet deposition density (affecting the effect). A complete perturbation strategy will contain a set of such coordinated adjustment equations or rules. The system calculates the specific target value for each flexible parameter based on the current preference vector. Due to potential coupling between parameters (as in the example of altitude and dosage above), this step requires simultaneous calculation or iterative solution to ensure that the adjusted parameter combination is internally consistent.All calculations are performed within a preset perturbation range to ensure that the output results are within a scientifically safe range. This yields a set of personalized, collaboratively optimized final material usage parameters. It is no longer a fixed value or a simple range, but a defined set of values ​​adapted to specific current preferences (e.g., agent A and B mixed in a 5:5 ratio, dosage 42 ml per acre, recommended operating height 2.2 meters).

[0096] In some embodiments, adjusting the flight altitude, speed, and route spacing of the basic flight trajectory based on real-time or forecasted environmental meteorological data to generate optimal flight trajectory parameters specifically includes:

[0097] With operational efficiency, spray uniformity, and energy consumption as optimization objectives, and the environmental meteorological data, crop canopy height information, and UAV performance parameters as constraints, a multi-objective function for trajectory parameter optimization is constructed.

[0098] Within the framework of the basic flight trajectory, multiple parameter combinations are generated for the three key parameters of flight altitude, flight speed and route spacing within their respective feasible value ranges, forming a candidate set of trajectory parameters;

[0099] Substitute each set of parameters in the candidate trajectory parameters into the multi-objective function to perform flight operation simulation, calculate their respective operation efficiency score, uniformity score and energy consumption score, and use a multi-objective optimization algorithm to solve for the Pareto optimal solution set under the constraints based on the operation efficiency score, the uniformity score and the energy consumption score.

[0100] From the Pareto optimal solution set, the final solution is selected according to the priority strategy of the current operation, and the optimal flight trajectory parameters are determined based on the final solution.

[0101] Operational efficiency: Generally inversely proportional to total operation time, aiming for faster flight speed and greater flight path spacing. Spray uniformity: Aiming for the minimum coefficient of variation in droplet deposition distribution on the crop canopy to ensure consistent results. Energy consumption: Aiming for the lowest total power consumption, directly related to operational costs and drone endurance. These objectives conflict with each other: increasing speed may reduce uniformity and increase energy consumption; reducing flight path spacing improves uniformity but reduces efficiency. Environmental meteorological data: For example, wind speed must be below a safe threshold (e.g., ≤5 level wind), crosswinds will affect the effective spray width, and temperature will affect droplet evaporation. Crop canopy height information: Flight altitude must be above a safe distance (e.g., more than 1 meter above the canopy), but not too high to avoid excessive drift. Drone performance parameters: Hardware limitations such as maximum payload, maximum climb rate, maximum speed, battery capacity, and nozzle flow range. Expressing the above objectives and constraints mathematically, a multi-objective optimization function F(x) = [efficiency (x), uniformity (x), energy consumption (x)] is formed, where x is the parameter vector to be optimized [altitude, speed, spacing]. Based on the constraints, a dynamic feasible range is determined for each parameter. For example, altitude: the range is determined as [H_min, H_max] based on the current wind speed and canopy height. Velocity: the range is determined as [V_min, V_max] based on the UAV performance and nozzle flow rate. Spacing: the range is determined as [S_min, S_max] based on the wind speed and nozzle type. Within the three-dimensional parameter space, hundreds or thousands of different parameter combinations [h_i, v_i, s_i] are generated using methods such as uniform sampling, Latin hypercube sampling, or grid search, forming a candidate set of trajectory parameters, where h_i, v_i, and s_i are any values ​​within their respective defined ranges. Flight operation simulation: For each parameter combination in the candidate set, a physics-based digital twin simulation is run. The simulation model simulates the UAV flying according to that set of parameters. The simulation also simulates the movement, evaporation, and deposition processes of droplets after they are ejected from the nozzle under the influence of the current specific environmental wind field (from real-time data). Finally, the simulation simulates the interaction between droplets and the specific three-dimensional structure of the crop canopy. Output the virtual job results under this set of parameters, such as completion time (used to calculate job efficiency score), deposition uniformity index (uniformity score), and estimated power consumption (energy consumption score). Apply a multi-objective optimization algorithm (such as NSGA-II, MOEA / D): The algorithm receives the three scores of all candidate solutions, and its task is to find a Pareto optimal solution set. In this solution set, no solution can further improve any objective without compromising at least one other objective. That is, these solutions represent all possible optimal trade-offs between "efficiency, uniformity, and energy consumption". Select a solution for actual execution from the mathematically optimal solution set. Define a priority strategy: This is the step of injecting business logic into technical decisions. Strategies such as: "Seize the farming season" mode: Prioritize the solution with the highest efficiency score. "Maintain effectiveness" mode: Prioritize the solution with the highest uniformity score.Cost Reduction Mode: Prioritizes the solution with the highest energy consumption score. Balanced Mode: Selects the solution with the largest weighted sum of the three objective scores, with weights set by the administrator. Based on a preset or real-time specified priority strategy, the system selects the most suitable solution from the Pareto optimal solution set. For example, in "Effect Guarantee Mode," the system will select the parameter combination with the highest uniformity score. The specific [altitude, speed, spacing] values ​​corresponding to the final solution are the generated optimal flight trajectory parameters.

[0102] In some embodiments, substituting each set of parameters from the candidate trajectory parameters into the multi-objective function for flight operation simulation includes:

[0103] Based on the model of the drone nozzle and the current operating parameters, a swarm of droplet particles with preset initial velocity and release angle is generated in a simulation environment.

[0104] Based on the crop canopy height information and three-dimensional structural model, and combined with the real-time environmental meteorological data, a canopy region fluid dynamic field is constructed in the simulation space. The canopy region fluid dynamic field characterizes the changes in wind speed and turbulence intensity inside and around the canopy with spatial location.

[0105] The fog droplet particle swarm is placed in the hydrodynamic field of the canopy region, and a simulation based on the coupling of computational fluid dynamics and discrete element method is run to calculate the trajectory of each fog droplet particle.

[0106] The number and size distribution of fog droplets adhering to a unit area of ​​the leaf are statistically analyzed to calculate the simulated deposition density and simulated coverage. The proportion of fog droplets that drift outside the target area is statistically analyzed as the simulated drift rate. Based on the simulated deposition density, the simulated coverage, and the simulated drift rate, the uniformity score and the energy consumption score are calculated.

[0107] In the simulation, particles representing pesticide droplets are initialized. Nozzle type: determines the initial droplet size distribution (e.g., VMD - median diameter) and initial diffusion angle. Different nozzles (e.g., hollow cone, fan-shaped) have different characteristics. Flight altitude and velocity: determine the initial spatial position and initial horizontal velocity of the released particle swarm. Nozzle flow rate: determines the total number of particles released per unit time, affecting deposition density. Release angle: some nozzles are adjustable, affecting the direction of the spray cone. In the 3D simulation space, a series of droplet particles with initial particle size, initial velocity vector (combining UAV flight speed and nozzle spray velocity), and initial angle are continuously generated from the UAV nozzle position. These particles are the basic objects for subsequent simulations. Environmental meteorological data: mainly real-time or forecasted wind speed and direction, serving as boundary conditions or background fields for the simulation. 3D structure model of the crop canopy: this is crucial for the simulation. The model includes not only height but also fine structures such as plant spacing, leaf density, and canopy porosity reconstructed through lidar or multi-view images. Based on the above inputs, the system solves the Navier-Stokes equations to simulate the motion of air flowing through a complex porous medium (crop canopy), generating a three-dimensional hydrodynamic field. This field not only includes the spatial distribution of wind speed and direction, but more importantly, it simulates the turbulence within the canopy (reduced wind speed, enhanced eddies) and the shear layer at the top of the canopy (drastic wind speed changes). This accurately reflects the complex airflow environment that droplets experience when passing through the canopy. Computational fluid dynamics (CFD) provides the instantaneous spatial flow field (wind speed, pressure). The discrete element method (DEM) tracks the forces acting on each droplet particle (considered as a discrete sphere or simplified model) in the flow field, including: air resistance (related to relative velocity and particle size), gravity, and (optionally) virtual collision forces (particle collisions, considered in dense spraying). Coupled solution: Within each extremely short time step, the DEM module calculates the forces and motions of particles based on the current flow field; the particle motion, in turn, locally affects the flow field (two-way or one-way coupling). This iterative process gradually calculates the complete three-dimensional trajectory of each particle from ejection to final settling or drifting away. Statistical analysis of the simulation results is performed, converting them into quantitative scores that can be used for optimization. All droplet particles that ultimately come into contact with and are "captured" by the crop canopy 3D model surface are identified. Simulated deposition density: The number of droplet particles adhering to a unit area of ​​leaf surface is counted. Particle size distribution analysis: The particle size of the adhered particles is counted to assess the quality of the effective deposition portion. Simulated coverage: The percentage of leaf surface area covered by at least one droplet is counted. All droplet particles whose trajectory ends outside the target plot boundary are identified. Simulated drift rate: The proportion of drifting particles to the total number of released particles is calculated. Uniformity score: Typically calculated by combining the coefficient of variation of the simulated deposition density (lower is more uniform) and the simulated coverage rate (higher is better), and may penalize high drift rates.Energy consumption score: Estimated by calling the UAV energy consumption model based on flight parameters (speed, altitude changes) and operation time. More directly, energy consumption can be indirectly characterized by the flight time required to complete the operation per unit area.

[0108] In some embodiments, comparing and analyzing the process execution data with the work plan to quantify and generate a deviation index characterizing the work execution status specifically includes:

[0109] The expected operational effect features are extracted from the operational plan, and the post-operation effect image of the target plot is obtained after the operation is completed. The expected operational effect features include the expected pesticide coverage distribution map, the expected pest and disease control target area map, or the expected vegetation index improvement area map.

[0110] The actual flight trajectory is spatially superimposed with the expected flight trajectory in the operation plan to calculate the trajectory overlap, the area of ​​the missed spray area, and the area of ​​the re-spray area.

[0111] The actual effect features extracted from the post-operation effect image are compared with the expected operation effect features to calculate the coverage difference between the actual coverage rate and the expected coverage rate, and / or calculate the overlap rate between the actual control area and the expected control target area.

[0112] Based on the trajectory overlap, the area of ​​the missed spray area, the area of ​​the re-spray area, the coverage difference and / or the prevention overlap rate, the deviation index is generated by weighted calculation according to preset rules.

[0113] Extract the pre-defined, quantifiable effect targets from the operational plan issued to the drone pilots. Expected pesticide coverage distribution map: A heat map indicating the expected percentage of spray coverage (e.g., 90% or more) in different areas within the plot. Expected pest and disease control target area map: Based on prior multispectral reconnaissance, the key areas of pest and disease occurrence are marked, representing the core areas that must be covered. Expected vegetation index improvement area map: For nutrient operations, areas where a significant increase in chlorophyll or biomass is expected are marked. After the operation is completed (e.g., a few hours later or the next day), conduct rapid aerial photography or remote sensing imaging of the target plot using the same drone or other methods. Image types can include visible light images and multispectral / hyperspectral images. Visible light images: Used to analyze the uniformity of spray coverage (through reflective or dye tracers). Multispectral / hyperspectral images: Used to directly assess the effectiveness of pest and disease control (changes in pest and disease stress index) or changes in vegetation growth (changes in vegetation index). Assess the degree of consistency between the actual drone flight and the planned path. In the GIS platform, the actual flight trajectory (a timestamped line) transmitted by the drone is overlaid with the expected flight trajectory in the operation plan. The spatial similarity between the two trajectories is calculated. Buffer analysis can be used to statistically analyze the proportion of the actual trajectory falling within a certain tolerance range (e.g., ±1 meter) of the expected trajectory. The higher the overlap, the more accurate the execution. Areas missed during spraying: Identify areas covered by the expected trajectory but not effectively covered by the actual trajectory. This is usually due to equipment failure, improper path planning, or pilot error, resulting in operational vacuum zones that directly affect the control effect. Areas over-sprayed: Identify areas repeatedly covered by the actual trajectory (more than the planned number of times). This can lead to localized over-dosage, potentially causing phytotoxicity, increasing costs, and causing pollution. From the post-operation effect images, image processing algorithms (such as threshold segmentation and machine learning models) are used to extract actual effect features. For example, identifying leaf areas covered by pesticide solution from visible light images generates an actual pesticide coverage map; calculating vegetation index changes from multispectral images generates an actual control effect map or vegetation improvement map. Calculate the coverage difference: Compare the actual pesticide coverage map with the expected pesticide coverage distribution map at the pixel level or region level. The difference can be calculated as the absolute difference or root mean square error of the area-weighted coverage rates of the two maps. For example, if the expected average coverage rate is 90% and the actual measurement is 85%, the difference is 5%. Calculate the control overlap rate: Overlay the actual control effect map (e.g., areas where the pest index decreases) with the expected pest control target area map. Calculate the proportion of the area of ​​the intersection of the two maps to the total area of ​​the expected target area. The higher the overlap rate, the more precise and effective the attack on the core problem. Pre-defined rules (weighting system): Platform administrators or agricultural experts will pre-define a set of scoring rules. For example: Deviation index = 100 - (Trajectory overlap × W1 + Missed spray penalty × W2 + Overspray penalty × W3 + Coverage difference penalty × W4 + Control overlap rate × W5). Where W1 to W5 are the weights of each item.Missed spraying, overspraying, and coverage discrepancies are typically penalized with deductions, while trajectory overlap and control overlap rates are considered beneficial and worthwhile. The weighting can be adjusted based on the type of operation (e.g., plant protection prioritizes coverage and overlap rates, while fertilization prioritizes uniformity). Substitute all the calculated raw indicator values ​​into the weighted calculation formula above. Output a deviation index value (usually a score between 0 and 100). A higher score indicates a smaller deviation between the actual execution and the expected plan, and thus a higher quality operation. This index directly and quantitatively reflects the overall execution level of the operation.

[0114] The above describes the integrated method of agricultural low-altitude operation and digital service in the embodiments of this application. The computer system in the embodiments of this application will be described in detail below in conjunction with the above integrated method of agricultural low-altitude operation and digital service.

[0115] Please see Figure 5 This is a schematic diagram of an exemplary hardware structure of a computer system in an embodiment of this application.

[0116] In some embodiments, the computer system 500 includes a computer device, which may be a terminal device. The computer device includes a processor 501, a memory 502, a sensor module 503, a communication module 504, an input device 505, and an output device 506 connected via a system bus. The processor 501 of the computer device provides computing and control capabilities. The memory 502 of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database is used to store data.

[0117] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0118] In some embodiments of this application, a computer-readable storage medium is provided, including instructions that, when executed on a computer system 500, cause the computer system 500 to perform the integrated agricultural low-altitude operation and digital service method of the embodiments of this application.

[0119] In some embodiments of this application, a computer program product is also provided, which, when run on a computer system 500, causes the computer system 500 to execute the integrated agricultural low-altitude operation and digital service method of the embodiments of this application.

[0120] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0121] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0122] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for integrating low-altitude agricultural operations with digital services, characterized in that, include: In response to receiving a job request containing geographic information of a target plot, create or update a digital profile for the target plot; Based on the historical operation data, crop growth time series data, and real-time environmental meteorological data recorded in the digital archive, an operation plan containing optimal flight trajectory parameters and material usage parameters is generated. Based on the geographic information and job requirements, multiple candidate pilots who meet the conditions are selected from the pilot pool through a preset matching model, and job invitations containing a summary of the job plan are pushed to the multiple candidate pilots. In response to the target pilot's confirmation of the job invitation, the job plan is sent to the target pilot's pilot terminal. The target pilot is any one of the multiple candidate pilots. The system receives real-time process execution data, including actual flight trajectory and material consumption, uploaded by the drone bound to the target pilot. This data is compared and analyzed with the work plan to quantify and generate a deviation index characterizing the work execution. Based on this deviation index, financial settlement is performed for the target pilot. Based on the geographic information and job requirements, a preset matching model is used to select multiple candidate pilots from the pilot pool who meet the criteria, and job invitations containing summaries of the job plan are sent to these multiple candidate pilots. Specifically, this includes: The task requirements are analyzed, and key task features are extracted to construct a task demand vector. The key task features include the urgency of the task, the type of target crop, the type of target pest or disease, or the agricultural process. Using the geographical location of the target plot as the center, select online drone pilots currently within a preset service radius from the drone pilot pool to form an initial candidate set; For each candidate pilot in the initial candidate set, historical operation data is retrieved from the pilot database on the cloud server, and based on the key operation features, a capability feature vector for each candidate pilot is generated. The similarity between the capability feature vector and the operation requirement vector is calculated as the core matching score. Based on the core matching score, combined with the pilot's qualification certification level and historical service evaluation score, a dynamic comprehensive recommendation index is calculated for each candidate pilot. The candidate pilots in the initial candidate set are ranked according to the dynamic comprehensive recommendation index, and the job invitation is pushed to a preset number of candidate pilots ranked first.

2. The integrated method for agricultural low-altitude operations and digital services according to claim 1, characterized in that, The process of generating an operational plan based on historical operational data, crop growth time-series data, and real-time acquired environmental and meteorological data recorded in the digital archive, including optimal flight trajectory parameters and material usage parameters, specifically includes: By integrating the crop growth time-series data, historical operation data, and real-time environmental meteorological data recorded in the digital archive, the personalized demand characteristics of this operation are extracted. The personalized demand characteristics include at least the crop health level, the probability of occurrence of specific pests and diseases, and operation preferences based on historical results. Based on the current date, the planting information of the target plot, the crop health level, and the probability of occurrence of the specific pests and diseases, determine the specific growth stage of the crop and the corresponding agricultural needs. Based on the specific growth stage and the crop health level, the corresponding standard material type, basic usage range and benchmark operation mode are retrieved and called from the agricultural knowledge base in the cloud server; Based on the operational preferences, the basic usage range is calibrated to generate material usage parameters for the target plot; Based on the elevation data and crop canopy height information of the target plot, and combined with the flight trajectories executed on the target plot in the historical operation data, a basic flight trajectory is planned. According to real-time or forecasted environmental meteorological data, the flight altitude, speed and route spacing of the basic flight trajectory are adjusted to generate the optimal flight trajectory parameters. The material usage parameters are integrated with the optimal flight trajectory parameters, and the corresponding operation time window is bound to generate the operation plan.

3. The integrated method for agricultural low-altitude operations and digital services according to claim 2, characterized in that, The step of calibrating the basic usage range based on the operational preferences to generate material usage parameters for the target plot specifically includes: The job preference is encoded as a multidimensional preference control vector, where each dimension of the preference control vector corresponds to a calibration tendency. The magnitude of the calibration tendency indicates the tendency intensity. The calibration tendency includes effect enhancement tendency, cost saving tendency, and risk avoidance tendency. The standard material types, basic dosage ranges, and benchmark operation modes retrieved from the agricultural knowledge base are analyzed to identify the controllable flexible parameters and allowable disturbance ranges. The flexible parameters include the proportion of mixed agents, the average application rate per acre, and the flight altitude during operation. Based on the preference control vector, a perturbation strategy is generated for each of the flexible parameter items. The perturbation strategy defines the direction of change of the parameter value and the relationship of the change function within the perturbation interval. Based on the disturbance strategy, all the flexible parameter items are calculated and adjusted synchronously to generate the material usage parameters.

4. The integrated method for agricultural low-altitude operations and digital services according to claim 2, characterized in that, The step of adjusting the flight altitude, speed, and route spacing of the basic flight trajectory based on real-time or forecasted environmental meteorological data to generate optimal flight trajectory parameters specifically includes: With operational efficiency, spray uniformity, and energy consumption as optimization objectives, and the environmental meteorological data, crop canopy height information, and UAV performance parameters as constraints, a multi-objective function for trajectory parameter optimization is constructed. Within the framework of the basic flight trajectory, multiple parameter combinations are generated for the three key parameters of flight altitude, flight speed and route spacing within their respective feasible value ranges, forming a candidate set of trajectory parameters; Substitute each set of parameters in the candidate trajectory parameters into the multi-objective function to perform flight operation simulation, calculate their respective operation efficiency score, uniformity score and energy consumption score, and use a multi-objective optimization algorithm to solve for the Pareto optimal solution set under the constraints based on the operation efficiency score, the uniformity score and the energy consumption score. From the Pareto optimal solution set, the final solution is selected according to the priority strategy of the current operation, and the optimal flight trajectory parameters are determined based on the final solution.

5. The integrated method for agricultural low-altitude operations and digital services according to claim 4, characterized in that, The step of substituting each set of parameters from the candidate trajectory parameters set into the multi-objective function for flight operation simulation includes: Based on the model of the drone nozzle and the current operating parameters, a swarm of droplet particles with preset initial velocity and release angle is generated in a simulation environment. Based on the crop canopy height information and three-dimensional structural model, and combined with the real-time environmental meteorological data, a canopy region fluid dynamic field is constructed in the simulation space. The canopy region fluid dynamic field characterizes the changes in wind speed and turbulence intensity inside and around the canopy with spatial location. The fog droplet particle swarm is placed in the hydrodynamic field of the canopy region, and a simulation based on the coupling of computational fluid dynamics and discrete element method is run to calculate the trajectory of each fog droplet particle. The number and size distribution of fog droplets adhering to a unit area of ​​the leaf are statistically analyzed to calculate the simulated deposition density and simulated coverage. The proportion of fog droplets that drift outside the target area is statistically analyzed as the simulated drift rate. Based on the simulated deposition density, the simulated coverage, and the simulated drift rate, the uniformity score and the energy consumption score are calculated.

6. The integrated method for agricultural low-altitude operations and digital services according to claim 1, characterized in that, The step of comparing and analyzing the process execution data with the work plan to quantify and generate a deviation index characterizing the work execution status specifically includes: The expected operational effect features are extracted from the operational plan, and the post-operation effect image of the target plot is obtained after the operation is completed. The expected operational effect features include the expected pesticide coverage distribution map, the expected pest and disease control target area map, or the expected vegetation index improvement area map. The actual flight trajectory is spatially superimposed with the expected flight trajectory in the operation plan to calculate the trajectory overlap, the area of ​​the missed spray area, and the area of ​​the re-spray area. The actual effect features extracted from the post-operation effect image are compared with the expected operation effect features to calculate the coverage difference between the actual coverage rate and the expected coverage rate, and / or calculate the overlap rate between the actual control area and the expected control target area. Based on the trajectory overlap, the area of ​​the missed spray area, the area of ​​the re-spray area, the coverage difference and / or the prevention overlap rate, the deviation index is generated by weighted calculation according to preset rules.

7. A computer system comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-6.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

Images