Precision spraying method using agricultural drones in lychee orchards

Abstract

This invention relates to a precision spraying method for agricultural drones used in lychee orchards, belonging to the field of drone spraying technology. The invention obtains droplet charge characteristic prediction results through a droplet charge feature prediction model, and then evaluates and analyzes the charge of droplets sprayed by the agricultural drone based on these prediction results. The evaluation and analysis results are then used to generate a first control strategy or a second control strategy, and the operating parameters of the agricultural drone spraying equipment are dynamically adjusted based on these strategies. This invention analyzes the droplet charge under environmental conditions by combining the operating parameters of the agricultural drone spraying equipment with scenario analysis of the droplet charge, thereby improving the spraying accuracy of the agricultural drone in lychee orchards.

Description

Technical Field

[0001] This invention relates to the field of drone spraying technology, and more particularly to a precision spraying method for plant protection drones used in lychee orchards. Background Technology

[0002] Lychee, a prized fruit of Lingnan (southern China), boasts high nutritional and economic value, earning it the title of "King of Fruits." However, due to its growth in tropical and subtropical regions with consistently high temperatures and humidity, lychee is susceptible to frequent and severe pests and diseases, significantly impacting its quality and yield. Therefore, strengthening effective chemical control of lychee pests and diseases is crucial for ensuring and increasing lychee production. In my country, pesticide application methods primarily include manual spraying, ground machinery spraying, and aerial spraying. In orchards in hot regions, pest and disease control still mainly relies on manual spraying operations such as backpack sprayers, pedal sprayers, and motorized high-pressure spray guns. However, manual spraying is inefficient, the quality of spraying is difficult to guarantee, and it cannot achieve rapid, full-coverage plant protection spraying in a short time, potentially delaying the optimal control period. Furthermore, due to labor shortages and an older average workforce in orchards, there is an urgent need to develop highly automated, time-saving, and labor-saving efficient pesticide application technologies. Ground machinery spraying operations are costly, have low pesticide utilization rates, and are difficult to operate in complex terrains such as mountains and hills. Agricultural drones are highly efficient and mobile, effectively avoiding the drawbacks of manual and ground-based mechanical spraying. They can become a new technology for large-scale and rapid integrated prevention and control of fruit tree diseases and pests, helping fruit farmers to reduce costs and increase efficiency. However, electrostatic droplets are easily affected by environmental interference, which can lead to poor deposition and reduce the spraying accuracy in lychee orchards. Summary of the Invention

[0003] This invention overcomes the shortcomings of the prior art and provides a method for precise spraying of plant protection drones in lychee orchards.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of this invention provides a method for precision spraying using agricultural drones in lychee orchards, comprising the following steps:

[0006] Obtain operational parameter data of the agricultural drone spraying equipment, and construct a droplet charge characteristic prediction model based on the generative adversarial network and the operational parameter data of the agricultural drone spraying equipment.

[0007] Environmental characteristic data of the litchi orchard are obtained, and the predicted results of the fog droplet charge characteristics are obtained by combining the environmental characteristic data of the litchi orchard with the fog droplet charge characteristic prediction model.

[0008] Based on the droplet charge characteristic prediction results, the charged charge of the droplets sprayed by the current agricultural drone spraying equipment is evaluated and analyzed to obtain the evaluation and analysis results;

[0009] Based on the evaluation and analysis results, a first control strategy or a second control strategy is generated, and the operating parameters of the agricultural drone spraying equipment are dynamically adjusted based on the first control strategy or the second control strategy.

[0010] Furthermore, in this method, a predictive model for the charged characteristics of droplets is constructed based on the generative adversarial network and the operating parameter data of the agricultural drone spraying equipment, specifically as follows:

[0011] A predictive model for the charged characteristics of fog droplets is constructed based on generative adversarial networks, and the electric field strength characteristic data of the agricultural drone spraying equipment is obtained based on the operating parameters of the agricultural drone spraying equipment.

[0012] Based on the electric field strength characteristic data of the plant protection drone spraying equipment, the characteristic data of the charge of the droplets when the plant protection drone spraying equipment is working are obtained, and a dataset of the charge of the droplets under different operating parameters is constructed based on the characteristic data of the charge of the droplets when the plant protection drone spraying equipment is working.

[0013] The data set of droplet charged charge characteristics under different operating parameters is input into the droplet charged charge characteristic prediction model, and several environmental parameters are set as constraints.

[0014] The droplet charge feature prediction model is trained based on the constraints to obtain a droplet charge feature prediction model that meets the requirements.

[0015] Furthermore, in this method, environmental characteristic data of the litchi orchard is obtained, and the predicted result of the fog droplet charge characteristic is obtained by combining the environmental characteristic data of the litchi orchard with the fog droplet charge characteristic prediction model. Specifically:

[0016] The environmental characteristic data of the lychee orchard and the real-time operating parameters of the plant protection drone spraying equipment are obtained, and the real-time operating parameters of the plant protection drone spraying equipment and the environmental characteristic data of the lychee orchard are input into the droplet charge characteristic prediction model.

[0017] The generator generates an initial droplet charge feature prediction result based on the real-time operating parameter data of the agricultural drone spraying equipment, and inputs the initial droplet charge feature prediction result into the discriminator for judgment;

[0018] If the discriminator accepts the initial droplet charge feature prediction result, then the initial droplet charge feature prediction result is used as the final prediction result, and the final prediction result is output as the droplet charge feature prediction result.

[0019] If the discriminator does not accept the initial droplet charge feature prediction result, it generates the next droplet charge feature prediction result until the discriminator accepts the current droplet charge feature prediction result and outputs the final prediction result as the droplet charge feature prediction result.

[0020] Furthermore, in this method, the charged charge of the droplets sprayed by the current agricultural drone spraying equipment is evaluated and analyzed based on the droplet charge characteristic prediction results, and the evaluation and analysis results are obtained, specifically as follows:

[0021] Set a threshold for droplet charge feature data, statistically analyze the droplet charge feature prediction results, obtain the spatial distribution characteristics of the droplet charge feature prediction results, and construct a spatial distribution map of droplet charge based on the spatial distribution characteristics of the droplet charge feature prediction results.

[0022] The proportion of droplets with charge levels below the droplet charge feature data threshold in the spatial distribution map of droplet charge is statistically analyzed, and it is determined whether the proportion of droplets with charge levels below the droplet charge feature data threshold in the spatial distribution map of droplet charge is greater than a preset proportion data threshold.

[0023] When the percentage of droplets with charge below the droplet charge characteristic data threshold in the droplet charge spatial distribution map is greater than a preset percentage data threshold, an abnormal spray evaluation result is generated.

[0024] When the percentage of droplets with charge below the droplet charge characteristic data threshold in the droplet charge spatial distribution map is not greater than the preset percentage data threshold, a normal spray assessment result is generated, and an assessment analysis result is generated based on the abnormal spray assessment result or the normal spray assessment result.

[0025] Furthermore, in this method, a first control strategy or a second control strategy is generated based on the evaluation and analysis results, specifically as follows:

[0026] When the evaluation and analysis result is an abnormal spray evaluation result, a first control strategy is generated; when the evaluation and analysis result is a normal spray evaluation result, a second control strategy is generated.

[0027] The first control strategy or the second control strategy is output as the output result.

[0028] Furthermore, in this method, the operating parameters of the agricultural drone spraying equipment are dynamically adjusted based on the first control strategy or the second control strategy, specifically including:

[0029] When the second control strategy is adopted, control is performed according to the current operating parameters of the agricultural drone spraying equipment;

[0030] When the first control strategy is adopted, a genetic algorithm is introduced, the number of generations is set based on the genetic algorithm, and the genetic process is performed based on the number of generations to reset the operating parameters of the agricultural drone spraying equipment.

[0031] After obtaining the operating parameters of the agricultural drone spraying equipment, the percentage of droplets whose charge is lower than the droplet charge characteristic data threshold is obtained. When the percentage data is not greater than the preset percentage data threshold, the genetic process ends and the operating parameters of the agricultural drone spraying equipment are output.

[0032] When the percentage data is not greater than the preset percentage data threshold, the inheritance continues until the percentage data is not greater than the preset percentage data threshold. Then, the operating parameters of the plant protection drone spraying equipment are reset, and control is performed according to the reset operating parameters of the plant protection drone spraying equipment.

[0033] A second aspect of the present invention provides a precision spraying system for agricultural drones used in lychee orchards, including a memory and a processor. The memory includes a program for a precision spraying method for agricultural drones used in lychee orchards. When the processor executes the program for the precision spraying method for agricultural drones used in lychee orchards, it implements the steps of any of the precision spraying methods for agricultural drones used in lychee orchards described in the present invention.

[0034] A third aspect of the present invention provides a computer-readable storage medium including a program for a precision spraying method using a plant protection drone for a lychee orchard. When the program is executed by a processor, it implements the steps of the precision spraying method using a plant protection drone for a lychee orchard as described in any one of the present invention.

[0035] This invention addresses the shortcomings of the prior art and has the following beneficial effects:

[0036] This invention acquires operational parameter data of an agricultural drone spraying device, constructs a droplet charge characteristic prediction model based on generative adversarial networks and this data, then obtains environmental characteristic data of a lychee orchard, and uses the droplet charge characteristic prediction model to obtain droplet charge characteristic prediction results. Based on these prediction results, the invention evaluates and analyzes the charge of droplets sprayed by the agricultural drone, obtains the evaluation and analysis results, and finally generates a first or second control strategy based on these results. The invention then dynamically adjusts the operational parameter data of the agricultural drone spraying device based on these strategies. By combining operational parameter data of the agricultural drone spraying device with analysis of droplet charge under environmental parameters, this invention can perform scenario analysis of the spraying situation of agricultural drones in lychee orchards based on droplet charge, thereby improving the spraying accuracy of agricultural drones. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0038] Figure 1 A flowchart of a precision spraying method using agricultural drones for lychee orchards is shown.

[0039] Figure 2 A system block diagram of a precision spraying system using agricultural drones for lychee orchards is shown. Detailed Implementation

[0040] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0041] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0042] like Figure 1 As shown, the first aspect of the present invention provides a method for precision spraying of agricultural drones in lychee orchards, comprising the following steps:

[0043] S102: Obtain the operating parameter data of the agricultural drone spraying equipment, and construct a droplet charge characteristic prediction model based on the generative adversarial network and the operating parameter data of the agricultural drone spraying equipment.

[0044] For example, the operating parameters of agricultural drone spraying equipment include the drone's flight speed, flight altitude, electric field strength data of the spraying equipment, spraying speed of the spraying equipment, and rotation speed of the spraying equipment.

[0045] S104: Obtain environmental characteristic data in the litchi orchard, and combine the environmental characteristic data of the litchi orchard with the fog droplet charge characteristic prediction model to obtain the fog droplet charge characteristic prediction result;

[0046] S106: Based on the prediction results of droplet charge characteristics, evaluate and analyze the charge of droplets sprayed by the current agricultural drone spraying equipment, and obtain the evaluation and analysis results;

[0047] S108: Generate a first control strategy or a second control strategy based on the evaluation and analysis results, and dynamically adjust the operating parameters of the agricultural drone spraying equipment based on the first control strategy or the second control strategy.

[0048] It should be noted that this invention analyzes the spraying situation of agricultural drones in lychee orchards by combining the operating parameters of the agricultural drone spraying equipment with the charge of the droplets under environmental parameters. This allows for scenario analysis based on the charge of the droplets, thereby improving the spraying accuracy of agricultural drones in spraying tasks.

[0049] Furthermore, in this method, operational parameter data of the agricultural drone spraying equipment is obtained, and a droplet charge characteristic prediction model is constructed based on the generative adversarial network and the operational parameter data of the agricultural drone spraying equipment. Specifically:

[0050] A predictive model for the charge characteristics of fog droplets is constructed based on generative adversarial networks, and the electric field strength characteristic data of the agricultural drone spraying equipment is obtained based on the operating parameters of the agricultural drone spraying equipment.

[0051] Based on the electric field strength characteristic data of the agricultural drone spraying equipment, the system predicts and obtains the droplet charge characteristic data when the agricultural drone spraying equipment is working. Based on the droplet charge characteristic data when the agricultural drone spraying equipment is working, a droplet charge characteristic dataset under different operating parameters is constructed.

[0052] The droplet charge characteristic dataset under different operating parameters is input into the droplet charge characteristic prediction model, and several environmental parameters are set as constraints.

[0053] The model for predicting the charge characteristics of fog droplets is trained based on constraints to obtain a model that meets the desired requirements.

[0054] It should be noted that temperature, humidity, and the electric field strength of the spraying equipment all affect the charge of the droplets. If the electric field strength remains constant, the charge of the droplets will decrease under high humidity and high temperature conditions. This loss of charge will cause the droplets to deviate during application, thus affecting the spraying accuracy. A droplet charge feature prediction model is constructed by fusing generative adversarial neural networks to predict the charge of droplets sprayed by agricultural drone spraying equipment.

[0055] Furthermore, in this method, environmental characteristic data of the litchi orchard is obtained, and the predicted results of the fog droplet charge characteristics are obtained by combining the environmental characteristic data of the litchi orchard with the fog droplet charge characteristic prediction model. Specifically:

[0056] Obtain environmental characteristic data in the litchi orchard and real-time operating parameters of the plant protection drone spraying equipment, and input the real-time operating parameters of the plant protection drone spraying equipment and the environmental characteristic data of the litchi orchard into the droplet charge characteristic prediction model;

[0057] The generator generates an initial droplet charge feature prediction result based on the real-time operating parameter data of the agricultural drone spraying equipment, and inputs the initial droplet charge feature prediction result into the discriminator for judgment;

[0058] If the discriminator accepts the initial droplet charge feature prediction result, then the initial droplet charge feature prediction result is used as the final prediction result, and the final prediction result is output as the droplet charge feature prediction result.

[0059] If the discriminator does not accept the initial droplet charge feature prediction result, it generates the next droplet charge feature prediction result until the discriminator accepts the current droplet charge feature prediction result and outputs the final prediction result as the droplet charge feature prediction result.

[0060] It should be noted that due to the differences in temperature and humidity within the lychee orchards, the amount of charge on the droplets sprayed by the spraying equipment will vary. This method can predict the amount of charge on the droplets when each agricultural drone sprays in each lychee orchard.

[0061] Furthermore, in this method, the charged charge of droplets sprayed by the current agricultural drone spraying equipment is evaluated and analyzed based on the droplet charge characteristic prediction results, and the evaluation and analysis results are obtained as follows:

[0062] Set a threshold for droplet charge feature data, statistically analyze the droplet charge feature prediction results, obtain the spatial distribution characteristics of the droplet charge feature prediction results, and construct a spatial distribution map of droplet charge based on the spatial distribution characteristics of the droplet charge feature prediction results.

[0063] For example, the spatial distribution characteristics of the predicted droplet charge features are the distribution characteristics of the droplet charge within the spraying work area (such as circular areas, fan-shaped areas, etc.) of the UAV. This forms a spatial distribution map of droplet charge. The spatial distribution map of droplet charge can predict and display the droplet charge distribution in different inner areas such as circular areas and fan-shaped areas (similar to the orbital map of planets in the solar system, where the blank areas between planets represent the amount of charge). Since there may be differences in temperature and humidity in each location, the droplets at different locations will have different amounts of charge.

[0064] The proportion of droplets with charge values ​​below the droplet charge characteristic data threshold in the spatial distribution map of droplet charge is statistically analyzed, and it is determined whether the proportion of droplets with charge values ​​below the droplet charge characteristic data threshold in the spatial distribution map of droplet charge is greater than the preset proportion data threshold.

[0065] When the percentage of droplets with charge below the droplet charge characteristic data threshold in the droplet charge spatial distribution map is greater than the preset percentage data threshold, an abnormal spray evaluation result is generated.

[0066] When the percentage of droplets with charge below the droplet charge characteristic data threshold in the droplet charge spatial distribution map is not greater than the preset percentage data threshold, a normal spray assessment result is generated, and an assessment analysis result is generated based on the abnormal spray assessment result or the normal spray assessment result.

[0067] It should be noted that this method can assess the spatial distribution of charged droplets sprayed by agricultural drone spraying equipment.

[0068] Furthermore, in this method, a first control strategy or a second control strategy is generated based on the evaluation and analysis results, specifically as follows:

[0069] When the evaluation analysis result is an abnormal spray evaluation result, a first control strategy is generated; when the evaluation analysis result is a normal spray evaluation result, a second control strategy is generated.

[0070] The first control strategy or the second control strategy is output as the result.

[0071] Furthermore, in this method, the operating parameters of the agricultural drone spraying equipment are dynamically adjusted based on either the first or second control strategy, specifically including:

[0072] When the second control strategy is adopted, control is performed according to the current operating parameters of the agricultural drone spraying equipment;

[0073] When the first control strategy is adopted, a genetic algorithm is introduced, the number of generations is set based on the genetic algorithm, the genetic process is performed based on the number of generations, and the operating parameters of the agricultural drone spraying equipment are reset.

[0074] After obtaining the operating parameters of the agricultural drone spraying equipment, the percentage of droplets whose charge is lower than the droplet charge characteristic data threshold is obtained. When the percentage is not greater than the preset percentage data threshold, the genetic process ends and the operating parameters of the agricultural drone spraying equipment are output.

[0075] When the percentage data is not greater than the preset percentage data threshold, the genetic process continues until the percentage data is not greater than the preset percentage data threshold. Then, the operating parameters of the plant protection drone spraying equipment are reset, and control is performed according to the reset operating parameters of the plant protection drone spraying equipment.

[0076] It should be noted that this method can optimize the operating parameters of agricultural drone spraying equipment based on the charge of droplets, thereby improving the spraying accuracy of agricultural drones in lychee orchards.

[0077] In addition, this method also includes:

[0078] The electric field intensity degradation and change characteristics of the agricultural drone spraying equipment are obtained through big data, and an electric field intensity degradation characteristic prediction model is constructed based on a deep neural network. The electric field intensity degradation and change characteristics of the agricultural drone spraying equipment are input into the electric field intensity degradation characteristic prediction model for training.

[0079] Through training, a predictive model for electric field intensity degradation characteristics that meets expectations is obtained, and data on the changes in electric field intensity degradation of agricultural drone spraying equipment within a preset time period are acquired.

[0080] The electric field intensity degradation change characteristic data of the agricultural drone spraying equipment within the preset time is input into the electric field intensity degradation characteristic prediction model that meets the expectations for prediction, so as to obtain the electric field intensity characteristic data of the agricultural drone spraying equipment under different operating parameters.

[0081] The operating parameters of the agricultural drone spraying equipment are updated based on the electric field strength characteristic data under different operating parameters. It should be noted that the electrostatic system of agricultural drone spraying equipment degrades after a certain period of use, resulting in different electric field strength characteristic data under different operating parameters. This method can further optimize the control parameters of the agricultural drone spraying equipment, thereby improving the spraying accuracy.

[0082] In addition, this method also includes:

[0083] By setting up electric field sensors in the lychee orchard, the electric field intensity data of the lychee orchard is obtained through the electric field sensors, and the location information of the agricultural drone is obtained. Based on the electric field intensity data of the lychee orchard, the location information of the agricultural drone, and the charge characteristic data of the fog droplets, the motion trajectory is analyzed.

[0084] By analyzing the motion trajectory, the motion trajectory of the droplets is obtained, as well as the area and location information of the lychee trees to be sprayed in the lychee orchard. Then, a dynamic spraying demonstration is performed based on the area and location of the lychee trees to be sprayed in the lychee orchard and the motion trajectory of the droplets using virtual reality technology.

[0085] By using dynamic demonstration, the area ratio information of the sprayed area is obtained, and it is determined whether the area ratio information of the sprayed area is greater than the preset area ratio threshold.

[0086] When the area ratio of the sprayed area is not greater than the preset area ratio threshold, the operating parameters of the current agricultural drone spraying equipment remain unchanged; when the area ratio of the sprayed area is greater than the preset area ratio threshold, the operating parameters of the current agricultural drone spraying equipment are adjusted.

[0087] It should be noted that in certain areas, there may be electrical equipment, and the electric field strength around the electrical equipment may be a certain level, which may affect the spraying effect of agricultural drones. This method can fully consider the impact of external electric field strength on the spraying of agricultural drones, thereby improving the spraying accuracy of agricultural drones.

[0088] like Figure 2 As shown, the second aspect of the present invention provides a precision spraying system 4 for plant protection drones in lychee orchards, including a memory 41 and a processor 42. The memory 41 includes a program for a precision spraying method for plant protection drones in lychee orchards. When the program for a precision spraying method for plant protection drones in lychee orchards is executed by the processor 42, it implements any of the steps of the precision spraying method for plant protection drones in lychee orchards.

[0089] A third aspect of the present invention provides a computer-readable storage medium including a program for a precision spraying method using a plant protection drone for a lychee orchard. When the program is executed by a processor, it implements the steps of any one of the precision spraying methods for a plant protection drone in a lychee orchard.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0091] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0093] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0095] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A precision spraying method using agricultural drones for lychee orchards, characterized in that, Includes the following steps: Obtain operational parameter data of the agricultural drone spraying equipment, and construct a droplet charge characteristic prediction model based on the generative adversarial network and the operational parameter data of the agricultural drone spraying equipment. Environmental characteristic data of the litchi orchard are obtained, and the predicted results of the fog droplet charge characteristics are obtained by combining the environmental characteristic data of the litchi orchard with the fog droplet charge characteristic prediction model. Based on the droplet charge characteristic prediction results, the charged charge of the droplets sprayed by the current agricultural drone spraying equipment is evaluated and analyzed to obtain the evaluation and analysis results; Based on the evaluation and analysis results, a first control strategy or a second control strategy is generated, and the operating parameters of the agricultural drone spraying equipment are dynamically adjusted based on the first control strategy or the second control strategy. A model for predicting the charge characteristics of droplets is constructed based on generative adversarial networks and the operational parameter data of the agricultural drone spraying equipment. Specifically: A predictive model for the charged characteristics of fog droplets is constructed based on generative adversarial networks, and the electric field strength characteristic data of the agricultural drone spraying equipment is obtained based on the operating parameters of the agricultural drone spraying equipment. Based on the electric field strength characteristic data of the plant protection drone spraying equipment, the characteristic data of the charge of the droplets when the plant protection drone spraying equipment is working are obtained, and a dataset of the charge of the droplets under different operating parameters is constructed based on the characteristic data of the charge of the droplets when the plant protection drone spraying equipment is working. The data set of droplet charged charge characteristics under different operating parameters is input into the droplet charged charge characteristic prediction model, and several environmental parameters are set as constraints. The droplet charge feature prediction model is trained based on the constraints to obtain a droplet charge feature prediction model that meets the requirements. Based on the droplet charge characteristic prediction results, the charged charge of the droplets sprayed by the current agricultural drone spraying equipment is evaluated and analyzed to obtain the evaluation and analysis results, specifically: Set a threshold for droplet charge feature data, statistically analyze the droplet charge feature prediction results, obtain the spatial distribution characteristics of the droplet charge feature prediction results, and construct a spatial distribution map of droplet charge based on the spatial distribution characteristics of the droplet charge feature prediction results. The proportion of droplets with charge levels below the droplet charge feature data threshold in the spatial distribution map of droplet charge is statistically analyzed, and it is determined whether the proportion of droplets with charge levels below the droplet charge feature data threshold in the spatial distribution map of droplet charge is greater than a preset proportion data threshold. When the percentage of droplets with charge below the droplet charge characteristic data threshold in the droplet charge spatial distribution map is greater than a preset percentage data threshold, an abnormal spray evaluation result is generated. When the percentage of droplets with charge below the droplet charge characteristic data threshold in the droplet charge spatial distribution map is not greater than the preset percentage data threshold, a normal spray assessment result is generated, and an assessment analysis result is generated based on the abnormal spray assessment result or the normal spray assessment result.

2. The method for precision spraying of plant protection drones in lychee orchards according to claim 1, characterized in that, Environmental characteristic data of the lychee orchard is obtained, and the fog droplet charge characteristic prediction result is obtained by combining the environmental characteristic data of the lychee orchard with the fog droplet charge characteristic prediction model. Specifically: The environmental characteristic data of the lychee orchard and the real-time operating parameters of the plant protection drone spraying equipment are obtained, and the real-time operating parameters of the plant protection drone spraying equipment and the environmental characteristic data of the lychee orchard are input into the droplet charge characteristic prediction model. The generator generates an initial droplet charge feature prediction result based on the real-time operating parameter data of the agricultural drone spraying equipment, and inputs the initial droplet charge feature prediction result into the discriminator for judgment; If the discriminator accepts the initial droplet charge feature prediction result, then the initial droplet charge feature prediction result is used as the final prediction result, and the final prediction result is output as the droplet charge feature prediction result. If the discriminator does not accept the initial droplet charge feature prediction result, it generates the next droplet charge feature prediction result until the discriminator accepts the current droplet charge feature prediction result and outputs the final prediction result as the droplet charge feature prediction result.

3. The method for precision spraying of plant protection drones in lychee orchards according to claim 1, characterized in that, Based on the evaluation and analysis results, a first control strategy or a second control strategy is generated, specifically as follows: When the evaluation and analysis result is an abnormal spray evaluation result, a first control strategy is generated; when the evaluation and analysis result is a normal spray evaluation result, a second control strategy is generated. The first control strategy or the second control strategy is output as the output result.

4. The method for precision spraying of plant protection drones in lychee orchards according to claim 3, characterized in that, The operating parameters of the agricultural drone spraying equipment are dynamically adjusted based on the first or second control strategy, specifically including: When the second control strategy is adopted, control is performed according to the current operating parameters of the agricultural drone spraying equipment; When the first control strategy is adopted, a genetic algorithm is introduced, the number of generations is set based on the genetic algorithm, and the genetic process is performed based on the number of generations to reset the operating parameters of the agricultural drone spraying equipment. After obtaining the operating parameters of the agricultural drone spraying equipment, the percentage of droplets whose charge is lower than the droplet charge characteristic data threshold is obtained. When the percentage data is not greater than the preset percentage data threshold, the genetic process ends and the operating parameters of the agricultural drone spraying equipment are output. When the percentage data is not greater than the preset percentage data threshold, the inheritance continues until the percentage data is not greater than the preset percentage data threshold. Then, the operating parameters of the plant protection drone spraying equipment are reset, and control is performed according to the reset operating parameters of the plant protection drone spraying equipment.

5. A precision spraying system for agricultural drones used in lychee orchards, characterized in that, The device includes a memory and a processor. The memory includes a program for a precision spraying method using a plant protection drone for lychee orchards. When the processor executes the program for the precision spraying method using a plant protection drone for lychee orchards, it implements the steps of the precision spraying method using a plant protection drone for lychee orchards as described in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that, The method includes a precision spraying method program for agricultural drones used in lychee orchards. When the program is executed by a processor, it implements the steps of the precision spraying method for agricultural drones used in lychee orchards as described in any one of claims 1-4.

Images