An agricultural robot precision pesticide application method and system based on adaptive control

Abstract

The application relates to the technical field of intelligent crop pesticide application, and relates to a precision pesticide application method and system of an agricultural robot based on adaptive control, which comprises the following steps: crop pest state score identification is performed on a crop information set to obtain a pest state matrix and acquire a pest matrix change sequence; a risk area set in the pest matrix change sequence is identified according to a pest threshold value, and the total pest intensity of the risk area set is acquired; when the total pest intensity is greater than a pesticide application threshold value, the spraying suitability of an environment information set is identified; when the spraying suitability is greater than an implementation condition threshold value, drug type identification is performed on the crop information set to obtain first pesticide configuration information and second pesticide configuration information; a pest development trend identification operation is performed on the pest matrix change sequence to obtain a pest trend range; and the first pesticide is proportioned according to the first pesticide configuration information and the risk area set, and the second pesticide is proportioned according to the second pesticide configuration information and the pest trend range. The application can improve the accuracy of crop pesticide application.

Description

Technical Field

[0001] This invention relates to the field of intelligent crop spraying technology, and in particular to a method and system for precise spraying of agricultural robots based on adaptive control. Background Technology

[0002] With the development of technology, pesticide application equipment such as fixed rails and drones are gradually replacing manual labor for large-scale pesticide application, which is labor-saving and efficient. However, even today, automated pesticide application using drones still operates under subjective human control. Whether pesticide application is necessary, whether it is suitable, and the amount of pesticide to be applied are all subject to subjective human assessment, resulting in low application accuracy and limitations in freeing up manpower.

[0003] Furthermore, the uniform application of pesticides by drones is a large-scale and rough operation. The prevention and control of crop pests are not separated, resulting in a large amount of unnecessary pesticide application even if the amount of pesticides used is reduced, which leads to soil and water pollution. Summary of the Invention

[0004] This invention provides a method for precise pesticide application using agricultural robots based on adaptive control, the main purpose of which is to improve the accuracy of pesticide application to crops.

[0005] To achieve the above objectives, the present invention provides a method for precise pesticide application using an agricultural robot based on adaptive control, comprising:

[0006] An agricultural robot base station is obtained, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a drug application robot;

[0007] Using a pre-built sensor cluster, environmental information sets and crop information sets of farmland in the target area are acquired. The edge processor is then used to identify crop damage status scores in the crop information set to obtain a damage status matrix. The historical damage status matrix within a preset pest and disease development time period is also acquired to obtain a damage matrix change sequence.

[0008] Based on a preset agricultural damage threshold, identify the set of risk regions in the agricultural damage matrix change sequence, and obtain the total agricultural damage intensity of the set of risk regions;

[0009] Determine whether the total agricultural pest intensity is greater than a preset pesticide application threshold. When the total agricultural pest intensity is greater than the pesticide application threshold, identify the spraying suitability of the environmental information set.

[0010] When the spraying suitability is greater than the preset implementation condition threshold, the crop information set is used to identify the pesticide type to obtain the first pesticide configuration information and the second pesticide configuration information.

[0011] Perform a pest development trend identification operation on the pest matrix change sequence to obtain the pest trend range;

[0012] Using the drug storage device, a drug ratio operation is performed based on the first pesticide configuration information and the risk area set to obtain a first agent, and a drug ratio operation is performed based on the second pesticide configuration information and the range of agricultural pest trends to obtain a second agent;

[0013] The first and second pesticides are sent to the application robot, and the application robot is used to apply pesticides to the risk area set and the range of agricultural pest trends.

[0014] Optionally, the step of using the edge processor to perform crop pest status score identification on the crop information set to obtain a pest status matrix includes:

[0015] Using the edge processor, growth nodes are identified in the crop information set to obtain crop growth nodes;

[0016] Based on the crop growth nodes, abnormal growth progress is identified to obtain an abnormal growth progress score matrix;

[0017] Insect identification is performed on the crop information set to obtain an insect quantity set and an insect type set. Based on the pre-configured insect type weight table, the influence coefficient corresponding to each insect type in the insect type set is queried to obtain the influence parameter set.

[0018] The set of influencing parameters and the set of insect numbers are weighted according to insect type to obtain the insect interference score matrix;

[0019] The agricultural damage state matrix is ​​obtained by summing the abnormal growth progress score matrix and the insect disturbance score matrix.

[0020] Optionally, the step of identifying a set of risk regions in the agricultural damage matrix change sequence based on a preset agricultural damage threshold, and obtaining the total agricultural damage intensity of the risk region set, includes:

[0021] Identify regions in the crop damage status matrix whose crop damage status scores are greater than a preset damage threshold to obtain a set of risk regions;

[0022] The sum of crop damage status scores in each risk region of the risk region set is calculated to obtain the damage intensity set.

[0023] The total agricultural damage intensity is obtained by summing the intensity of each agricultural damage in the set of agricultural damage intensities.

[0024] Optionally, identifying the spraying suitability of the environmental information set includes:

[0025] Using pre-constructed hierarchical rules, each piece of environmental information in the environmental information set is evaluated based on a good-medium-bad level, resulting in a set of environmental assessment variables.

[0026] Based on the pre-built set of membership functions, the pre-built Mamdani inference method is used to calculate the membership values ​​of the environmental assessment variable set based on the pre-built fuzzy rule base, so as to obtain the rule membership set.

[0027] Identify the fuzzy rule with the highest membership value in the rule membership set to obtain the target fuzzy rule, and output the pre-constructed output decision in the target fuzzy rule to obtain the spraying suitability.

[0028] Optionally, after the total pesticide damage intensity exceeds the application threshold, the method further includes:

[0029] Obtain the application time interval;

[0030] When the drug administration time interval is greater than the preset treatment period, the drug administration time interval is weighted according to the preset delay coefficient to obtain the time influence bias;

[0031] Based on the time influence bias, time spraying rules are constructed and filled into the fuzzy rule base to obtain an updated fuzzy rule base.

[0032] Optionally, the step of identifying the pesticide type in the crop information set to obtain first pesticide configuration information and second pesticide configuration information includes:

[0033] The crop information set is subjected to crop pest type identification to obtain crop pest type labels;

[0034] Using a pre-constructed pesticide configuration table, a query operation is performed based on the agricultural pest type label to obtain the first pesticide configuration information;

[0035] Based on the preset preventive pesticide application strategy, the first pesticide application information is fine-tuned to obtain the second pesticide application information.

[0036] Optionally, the step of identifying crop pest types from the crop information set to obtain crop pest type labels includes:

[0037] Obtain historical records of agricultural damage scenarios, wherein the historical records of agricultural damage scenarios include historical crop information and historical assessment tags;

[0038] Using a pre-built regression network model, machine learning is performed on the historical agricultural damage scene records to obtain an agricultural damage type identification network;

[0039] Using the aforementioned agricultural pest type identification network, the crop information set is subjected to text quantization processing to obtain a crop information vector set;

[0040] The crop information vector set is subjected to a clustering operation based on the similarity of each crop pest type pre-learned in the crop pest type identification network to obtain the crop pest similarity score corresponding to each crop pest type.

[0041] Extract the crop type corresponding to the crop type with the highest crop similarity score from the crop similarity scores of each crop type to obtain the crop type label.

[0042] Optionally, the step of using the spraying robot to perform spraying operations on the set of risk areas and the range of agricultural pest trends includes:

[0043] Using the spraying robot, a cruise operation is carried out on the target area farmland according to a preset flight route, and the location information of the spraying robot is obtained during the cruise operation.

[0044] Based on the set of risk areas and the range of agricultural pest trends, identify the reagent type corresponding to the location information;

[0045] The location information is obtained in the crop damage status matrix, and the target status score is obtained. Based on the target status score, the spraying speed of the pre-constructed spraying valve is adjusted to obtain the target spraying speed.

[0046] Based on the target spraying speed and reagent type, the first or second agent is selectively sprayed onto the location information.

[0047] Optionally, after using the spraying robot to apply pesticides to the set of risk areas and the range of agricultural pest trends, the method further includes:

[0048] Obtain the pesticide application records within the preset planting time period to obtain a set of pesticide application records;

[0049] The set of risk areas and the range of agricultural pest trends corresponding to each application are obtained from the application record set to obtain the applied areas;

[0050] Obtain the crop types in the farmland of the target area;

[0051] When the crop type is a preset cultivation type, superior plants in the target area farmland are obtained according to the area where the pesticide has been applied;

[0052] When the crop type is a preset commodity type, pesticide-free products in the target area farmland are obtained based on the area where pesticides have been applied.

[0053] To achieve the above objectives, the present invention also provides an agricultural robot precision spraying system based on adaptive control, comprising:

[0054] The device acquisition module is used to acquire agricultural robot base stations, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a spraying robot;

[0055] The crop damage status identification module is used to acquire environmental information set and crop information set of farmland in the target area using a pre-constructed sensor cluster, and to use the edge processor to identify crop damage status score of crop information set to obtain crop damage status matrix, and to acquire historical crop damage status matrix within a preset pest and disease development time period to obtain crop damage matrix change sequence, and to identify risk area set in the crop damage matrix change sequence according to preset crop damage threshold, and to obtain the total crop damage intensity of the risk area set;

[0056] The pesticide application suitability identification module is used to determine whether the total pesticide intensity is greater than a preset pesticide application threshold. When the total pesticide intensity is greater than the pesticide application threshold, the spraying suitability of the environmental information set is identified.

[0057] An adaptive pesticide application module is used to identify pesticide types in the crop information set when the spraying suitability exceeds a preset implementation condition threshold, obtain first pesticide configuration information and second pesticide configuration information, identify the agricultural pest development trend in the agricultural pest matrix change sequence to obtain the agricultural pest trend range, and use the pesticide storage device to perform pesticide ratio operation based on the first pesticide configuration information and risk area set to obtain a first agent, and perform pesticide ratio operation based on the second pesticide configuration information and agricultural pest trend range to obtain a second agent, and send the first agent and second agent to the application robot, and use the application robot to perform pesticide application operation on the risk area set and agricultural pest trend range.

[0058] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0059] Memory, storing at least one instruction;

[0060] The processor executes the instructions stored in the memory to implement the above-described method for precise pesticide application using an agricultural robot based on adaptive control.

[0061] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned method for precise pesticide application using an agricultural robot based on adaptive control.

[0062] To address the problems described in the background section, this invention first uses artificial intelligence to identify crop information sets, understanding basic information such as crop pests and growth status, and obtaining a pest matrix change sequence. Then, it identifies areas where pests occur using thresholds, obtaining a set of risk areas. This invention can control the application conditions of pesticides using fuzzy algorithms, thereby determining the appropriate timing for application. Furthermore, it transforms traditional large-scale pesticide application into three levels: therapeutic application, preventative application, and no application, thus significantly reducing the amount of pesticides used and achieving automated management for early detection and treatment. Therefore, this invention can improve the accuracy of pesticide application for crops. Attached Figure Description

[0063] Figure 1 This is a flowchart illustrating a method for precise pesticide application using an agricultural robot based on adaptive control, according to an embodiment of the present invention.

[0064] Figure 2 A functional block diagram of an agricultural robot precision spraying system based on adaptive control provided in an embodiment of the present invention;

[0065] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the precise pesticide application method of an agricultural robot based on adaptive control, according to an embodiment of the present invention.

[0066] Explanation of reference numerals in the attached figures:

[0067] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0068] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0069] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0070] This application provides a method for precise pesticide application using an agricultural robot based on adaptive control. The execution entity of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software may be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0071] Reference Figure 1The diagram shown is a flowchart illustrating a precision pesticide application method using an agricultural robot based on adaptive control, according to an embodiment of the present invention. In this embodiment, the precision pesticide application method using an agricultural robot based on adaptive control includes:

[0072] S1. Obtain an agricultural robot base station, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a drug application robot.

[0073] The agricultural robot base station refers to equipment used for complex management of pesticide application processes, including storage of equipment and pesticides.

[0074] The edge processor refers to a server with local computing capabilities that can execute preset automated programs.

[0075] The drug storage equipment refers to equipment for storing and mixing commonly used crop drugs (hormones, pesticides, fungicides, and soil conditioners, etc.).

[0076] The term "drug-dispensing robot" refers to a mobile device capable of spraying drugs, such as rail-mounted, tracked, or drone-borne devices. This invention illustrates the use of a drone as a drug-dispensing robot.

[0077] Specifically, in this embodiment of the invention, the three devices—edge processor, drug storage device, and drug delivery robot—can be deployed independently, or they can be integrated and placed in the center, edge, or a custom area of ​​the farm.

[0078] S2. Using a pre-built sensor cluster, acquire the environmental information set and crop information set of the target area farmland, and use the edge processor to identify the crop damage status score of the crop information set to obtain the damage status matrix, and acquire the historical damage status matrix within the preset pest and disease development time period to obtain the damage matrix change sequence.

[0079] The sensor cluster refers to a collection of sensors responsible for collecting various crop growth information in the farm, such as sensors that collect soil moisture, air humidity, temperature, and crop images.

[0080] The target area farmland refers to farmland within the management range of the agricultural robot base station.

[0081] The environmental information set refers to a collection of information unrelated to crops, such as humidity, temperature, and soil material content.

[0082] The crop information set refers to a collection of information related to crops, such as crop type, use, quality, growth stage, and pest status.

[0083] The crop pest status score identification operation refers to the process of determining whether crops are affected by pests or diseases by using artificial intelligence to recognize images. The pest status matrix refers to the crop pest status score of crops at various locations in the target area of ​​farmland.

[0084] The preset pest and disease development time period can be configured to 20 days, which represents the time interval between the occurrence of crop abnormalities and the formation of agricultural damage.

[0085] The agricultural pest matrix change sequence refers to the set of agricultural pest state matrices for each day during the pest development period.

[0086] Specifically, in this embodiment of the invention, to ensure the scientific use of the agricultural robot base station and achieve the goal of early detection and early treatment, the invention employs sensor acquisition and timed patrol photography to obtain environmental and crop information sets. Then, an edge processor identifies the crop damage status score at each location, obtaining a damage status matrix. Furthermore, the historical damage status matrix within a preset pest and disease development time period is recorded to obtain a damage matrix change sequence.

[0087] In detail, in this embodiment of the invention, the step of using the edge processor to perform crop pest status score identification on the crop information set to obtain a pest status matrix includes:

[0088] Using the edge processor, growth nodes are identified in the crop information set to obtain crop growth nodes;

[0089] Based on the crop growth nodes, abnormal growth progress is identified to obtain an abnormal growth progress score matrix;

[0090] Insect identification is performed on the crop information set to obtain an insect quantity set and an insect type set. Based on the pre-configured insect type weight table, the influence coefficient corresponding to each insect type in the insect type set is queried to obtain the influence parameter set.

[0091] The set of influencing parameters and the set of insect numbers are weighted according to insect type to obtain the insect interference score matrix;

[0092] The agricultural damage state matrix is ​​obtained by summing the abnormal growth progress score matrix and the insect disturbance score matrix.

[0093] The growth node identification refers to the process of identifying crop growth stages through key crop features, such as leaf number, flowering, and ear formation.

[0094] The crop growth node refers to the result of the growth node identification operation, such as the seedling stage, jointing stage, and pollination stage.

[0095] The abnormal growth progress identification refers to the process of comparing the crop growth nodes with the standard growth nodes of normal growth to understand whether the crop is growing slowly.

[0096] The growth progress anomaly score matrix refers to the identification result of the growth progress anomaly identification operation.

[0097] The insect identification refers to the process of periodically attracting and collecting insects using a collection device, and then recording the insect type and quantity. The insect quantity set refers to the set of quantities of each insect species in a pre-defined list of common pests (such as fall armyworm, locust, and rice planthopper). The insect type set refers to the set of types or names of various insects in the list of common pests.

[0098] The insect type weight table refers to a pre-configured table by the enterprise, used to describe the degree of influence of various insects on various crops. The influence coefficient refers to the degree of influence of various insects on various crops. The set of influence parameters refers to the set of influence coefficients corresponding to each insect type in the insect type set.

[0099] The weighted calculation operation based on insect type refers to the process of multiplying the influence parameters and the number of insects for the same insect. The insect interference score matrix refers to the result of the weighted calculation operation based on insect type on the set of influence parameters and the set of insect numbers.

[0100] The matrix summation refers to the process of simply adding the values ​​of the corresponding elements in the two matrices: the growth progress anomaly score matrix and the insect disturbance score matrix.

[0101] Specifically, when identifying agricultural damage, this invention comprehensively identifies the damage from two aspects: the agricultural plant's own growth and external pests.

[0102] First, this invention obtains the current crop growth node by identifying the growth node. Then, through the abnormal growth progress identification operation, it compares the normal crop node with the current crop growth node to understand the abnormalities in factors such as crop growth rate, color, and lesions, and obtains a growth progress abnormality score matrix.

[0103] Furthermore, this invention performs insect identification on the crop information set to obtain a set of insect quantities and insect types. Then, it queries an insect type weight table to obtain a set of influencing parameters. Subsequently, it performs a weighted calculation operation based on the insect type to obtain an insect interference score matrix, thereby achieving an understanding of crop pests.

[0104] Finally, this invention comprehensively considers the abnormal growth progress score matrix and the insect interference score matrix to obtain the agricultural damage state matrix.

[0105] S3. Based on the preset agricultural damage threshold, identify the set of risk areas in the agricultural damage matrix change sequence, and obtain the total agricultural damage intensity of the set of risk areas.

[0106] The agricultural damage threshold refers to a threshold value assessed by the enterprise to determine the impact on crop yield, pesticide properties, and other values. For example, if the crop agricultural damage status score exceeds this value, there is a risk of crop death or reduced yield.

[0107] The risk area set refers to the set of areas that require treatment intervention.

[0108] The total agricultural damage intensity refers to the sum of crop agricultural damage status scores contained in the risk area set.

[0109] In detail, in this embodiment of the invention, the step of identifying a set of risk regions in the agricultural damage matrix change sequence according to a preset agricultural damage threshold, and obtaining the total agricultural damage intensity of the risk region set, includes:

[0110] Identify regions in the crop damage status matrix whose crop damage status scores are greater than a preset damage threshold to obtain a set of risk regions;

[0111] The sum of crop damage status scores in each risk region of the risk region set is calculated to obtain the damage intensity set.

[0112] The total agricultural damage intensity is obtained by summing the intensity of each agricultural damage in the set of agricultural damage intensities.

[0113] The agricultural damage intensity set refers to the sum of crop damage status scores in each independent risk zone.

[0114] Specifically, in this embodiment of the invention, the area requiring crop damage treatment is first identified by comparing the crop damage status score with a preset damage threshold, thus obtaining a set of risk areas.

[0115] Specifically, this invention could directly obtain the total agricultural damage intensity. However, considering that the subsequent pesticide application process requires non-uniform application based on the severity of the agricultural damage, the sum of the agricultural damage status scores of each crop in each risk area is calculated independently to obtain a set of agricultural damage intensities, which are then summed to obtain the total agricultural damage intensity.

[0116] S4. Determine whether the total agricultural pest intensity is greater than the preset pesticide application threshold. When the total agricultural pest intensity is greater than the pesticide application threshold, identify the spraying suitability of the environmental information set.

[0117] The pesticide application threshold is configured as 10%, meaning that pesticide application can begin when 10% of the crops in the target area's farmland show signs of crop damage.

[0118] The spray suitability score refers to the score indicating whether the current environment is suitable for pesticide application.

[0119] Specifically, in this embodiment of the invention, considering the significant impact of environmental factors on the difficulty and effectiveness of pesticide application, it is necessary to identify the appropriateness of spraying as a prerequisite for initiating pesticide application.

[0120] In detail, in this embodiment of the invention, identifying the spraying suitability of the environmental information set includes:

[0121] Using pre-constructed hierarchical rules, each piece of environmental information in the environmental information set is evaluated based on a good-medium-bad level, resulting in a set of environmental assessment variables.

[0122] Based on the pre-built set of membership functions, the pre-built Mamdani inference method is used to calculate the membership values ​​of the environmental assessment variable set based on the pre-built fuzzy rule base, so as to obtain the rule membership set.

[0123] Identify the fuzzy rule with the highest membership value in the rule membership set to obtain the target fuzzy rule, and output the pre-constructed output decision in the target fuzzy rule to obtain the spraying suitability.

[0124] The grading rule refers to the rule of dividing each environmental information into three levels: good, medium, and bad. For example, the temperature range of 20~25℃ is configured as good, the temperature ranges of 15~20℃ and 25~30℃ are configured as medium, and the temperature ranges of less than 15℃ and greater than 30℃ are configured as bad.

[0125] The aforementioned good-medium-bad rating assessment refers to the process of transforming each piece of environmental information in the environmental information set into a vector of good, bad, or medium according to a grading rule. The environmental assessment variable set refers to the assessment result of the rating assessment operation on the environmental information set.

[0126] The membership function set refers to functions preset by the enterprise, used to transform the "good / neutral / bad" status of each input variable into a membership function (commonly triangular or trapezoidal functions) to quantify the degree of fuzziness. For example, the membership function for "good" temperature is a trapezoidal function, with a membership degree of 0.5 for 20℃, 1 for 22-23℃, and then dropping back to 0.5 for 25℃.

[0127] The Mamdani inference method refers to taking the minimum membership degree of the input variable as the activation strength for each rule; and for each category of the output variable, the membership degrees of all activated rules are superimposed. For example, when a rule 1 is preset based on agricultural experience as [(T=Medium)∧(H=Good)∧(W=Medium), then the output is "Apply pesticides with caution"], where T represents temperature, H represents humidity, and W represents wind speed. When the input environmental information set [T=Good (membership degree 0.8), H=Medium (0.6), W=Good (0.9)] is activated for rule 1, then min(0.8, 0.6, 0.9) = 0.6 is obtained, indicating that the output "Apply pesticides with caution" has a membership degree of 0.6.

[0128] When rule 3 is defined as [(T=good)∧(H=good)∧(W=good), then output "Appropriate to administer medicine"], activating rule 3 results in min(0.8, 0.6, 0.9) = 0.6, indicating that the membership degree of "Appropriate to administer medicine" is 0.6. Therefore, according to Mamdani's inference method, the final output is [Appropriate to administer medicine (0.6), Cautious to administer medicine (0.6), Inappropriate to administer medicine (0)].

[0129] The fuzzy rule base refers to rules summarized from agricultural machinery experience regarding suitable pesticide spraying, such as "good temperature, good humidity, and good wind speed indicate suitable pesticide application." This is expressed as "(T=good) ∧ (H=good) ∧ (W=good), then the output is 'suitable for pesticide application'." The fuzzy rule base iterates through various environmental conditions (good, bad, neutral) to achieve multiple rule combinations.

[0130] The rule membership set refers to the membership degree of a subset of rules calculated using the Mamdani inference method after substituting the environmental information set into each rule in the preset fuzzy rule base.

[0131] The target fuzzy rule refers to the fuzzy rule with the highest membership value in the rule membership set, such as "Rule 3".

[0132] The output decision pre-constructed in the target fuzzy rule, as can be seen from the fuzzy rule base, is "suitable for pesticide application," and the spraying suitability is "spraying suitability."

[0133] Specifically, in this embodiment of the invention, to ensure that the machine has a drug administration assessment capability similar to human experience, a membership function set and a fuzzy rule base are constructed. Then, through preset hierarchical rules, various environmental information is transformed into a set of environmental assessment variables.

[0134] Then, based on the Mamdani inference method, the membership values ​​of the set of environmental assessment variables with respect to various rules are calculated to obtain the rule membership set. Finally, the fuzzy rule with the highest membership value is selected to obtain the target fuzzy rule, and the pre-constructed output decision in the target fuzzy rule is output to obtain the spraying suitability.

[0135] S5. When the spraying suitability is greater than the preset implementation condition threshold, the crop information set is used to identify the pesticide type to obtain the first pesticide configuration information and the second pesticide configuration information.

[0136] The implementation condition threshold can be configured as "cautious application", which means that the application process can be carried out if the spraying suitability is in the state of "suitable application" or "cautious application".

[0137] The drug type identification refers to the process of selecting the appropriate drug based on the type of agricultural pest.

[0138] The first pesticide formulation information refers to the pesticide ratio for treating conditions in risk areas. The second pesticide formulation information refers to the pesticide ratio for areas that may develop into risk areas but have not yet reached those areas.

[0139] In detail, in this embodiment of the invention, the step of identifying the pesticide type of the crop information set to obtain first pesticide configuration information and second pesticide configuration information includes:

[0140] The crop information set is subjected to crop pest type identification to obtain crop pest type labels;

[0141] Using a pre-constructed pesticide configuration table, a query operation is performed based on the agricultural pest type label to obtain the first pesticide configuration information;

[0142] Based on the preset preventive pesticide application strategy, the first pesticide application information is fine-tuned to obtain the second pesticide application information.

[0143] The agricultural damage type identification refers to the process of using a neural network model to check whether the current agricultural damage situation has occurred historically. If it has occurred historically, the output is an agricultural damage type label.

[0144] The drug formulation table refers to a table that records how pesticides should be mixed for various types of agricultural pests.

[0145] The first pesticide configuration information refers to the query results of the pesticide configuration table based on the pesticide type label.

[0146] The preventive pesticide formulation strategy refers to the process of how different types of agricultural pests should be controlled by pesticide formulation. It usually involves selectively modifying the content in the first pesticide formulation information. Each company can adaptively construct the system based on its region, variety, and agricultural technology experience.

[0147] The second pesticide configuration information refers to the first pesticide configuration information adjusted according to the preventive pesticide application strategy.

[0148] Specifically, in this embodiment of the invention, machine learning can be used to learn from historical experience in dealing with agricultural pests, thereby identifying pest types and obtaining pest type labels. Then, a pesticide configuration table is queried to obtain the corresponding first pesticide configuration information. Finally, to prevent surrounding crops from being affected, this invention adopts a preventative approach, using a preset preventative pesticide application strategy to fine-tune the first pesticide configuration information, thereby obtaining second pesticide configuration information.

[0149] In detail, in this embodiment of the invention, the step of identifying crop pest types from the crop information set to obtain crop pest type labels includes:

[0150] Obtain historical records of agricultural damage scenarios, wherein the historical records of agricultural damage scenarios include historical crop information and historical assessment tags;

[0151] Using a pre-built regression network model, machine learning is performed on the historical agricultural damage scene records to obtain an agricultural damage type identification network;

[0152] Using the aforementioned agricultural pest type identification network, the crop information set is subjected to text quantization processing to obtain a crop information vector set;

[0153] The crop information vector set is subjected to a clustering operation based on the similarity of each crop pest type pre-learned in the crop pest type identification network to obtain the crop pest similarity score corresponding to each crop pest type.

[0154] Extract the crop type corresponding to the crop type with the highest crop similarity score from the crop similarity scores of each crop type to obtain the crop type label.

[0155] The historical crop damage scene record refers to data that records historical crop damage scenes and historical assessment levels. The historical crop information is the historical crop damage scene, and the historical assessment tag is the historical assessment level.

[0156] The regression network model refers to a model that can learn the relationships between multiple vectors. The crop damage type identification network refers to a model that has learned the mapping relationship between crop information and evaluation labels.

[0157] The machine learning mentioned here refers to the process of training a model using the cross-entropy loss algorithm, which will not be elaborated here.

[0158] The text quantization processing refers to the process of segmenting and quantizing text. The crop information vector set refers to the text quantization processing result of the crop information set.

[0159] The agricultural pest types pre-learned in the agricultural pest type identification network refer to the agricultural pest types that appear or are summarized in the historical agricultural pest scene records.

[0160] The similarity clustering operation refers to the process of calculating the similarity between the crop information vector set and historical crop information of various types of agricultural pests through the cosine similarity clustering algorithm.

[0161] The agricultural pest similarity score corresponding to each type of agricultural pest refers to the result of similarity clustering operation of crop information vector set.

[0162] Specifically, in this embodiment of the invention, a regression network model is trained to quickly identify agricultural pest type labels.

[0163] Specifically, in this embodiment of the invention, historical records of agricultural damage scenarios are first acquired. Then, a regression network model is trained using these records to obtain an agricultural damage type identification network. This network is then used to quantize the crop information vector set, resulting in a crop information vector set. Finally, a similarity clustering operation is performed to obtain agricultural damage similarity scores, which improves the efficiency and accuracy of agricultural damage type identification. Ultimately, agricultural damage type labels are obtained.

[0164] In detail, in this embodiment of the invention, after the total agricultural pest intensity exceeds the application threshold, the method further includes:

[0165] Obtain the application time interval;

[0166] When the drug administration time interval is greater than the preset treatment period, the drug administration time interval is weighted according to the preset delay coefficient to obtain the time influence bias;

[0167] Based on the time influence bias, time spraying rules are constructed and filled into the fuzzy rule base to obtain an updated fuzzy rule base.

[0168] The application time interval refers to the time between the planned application and the current time.

[0169] The treatment period is configured to be 3 days.

[0170] The delay coefficient is a coefficient used to adjust the time length index so that the influence of the application time interval on the spraying suitability decision is comparable to other factors such as temperature and wind speed.

[0171] The time-effect bias refers to the product of the delay coefficient and the application time interval.

[0172] The time-based spraying rule states that the longer the spraying time is delayed, the higher the probability of suitable application.

[0173] The phrase "filling the time-spraying rule into the fuzzy rule base" refers to adding the time-spraying rule to the basic fuzzy rules. For example, the above "Rule 3" [(T=good)∧(H=good)∧(W=good), then output "suitable for application"] can be replaced with "Rule 3 (change)" [(T=good)∧(H=good)∧(W=good)∧(D=good), then output "suitable for application"], where D represents the time-spraying rule.

[0174] The updated fuzzy rule base refers to the fuzzy rule base that incorporates time-based spraying rules.

[0175] Specifically, in this embodiment of the invention, if continuous unsuitable weather conditions occur, leading to a prolonged failure to effectively address agricultural pests, the conditions for suitable pesticide application are relaxed through a time-based spraying rule, thereby accelerating the application process. For example, if unsuitable weather conditions persist for five days, a decision should be made to apply pesticides cautiously, even if the weather is unsuitable. This can be used to prevent agricultural pests from being prolonged and causing the situation to worsen.

[0176] S6. Perform agricultural damage development trend identification operation on the agricultural damage matrix change sequence to obtain the agricultural damage trend range.

[0177] The agricultural pest development trend identification operation refers to the process of identifying whether changes in crop information of crops that have not yet become risk areas are consistent with changes in crop information of crops that were previously in risk areas.

[0178] The agricultural damage trend range refers to the range in which the characteristics of crop information changes in the previous risk area have appeared in the crop information changes of the crop.

[0179] Specifically, in this embodiment of the invention, a similarity algorithm can be used to calculate whether the crop damage status matrix of each region is trending towards becoming a risk region. If there is a trend towards becoming a risk region, it indicates that the crop region may be infected by pests and diseases and needs to be prevented and treated in advance.

[0180] S7. Using the drug storage device, perform drug mixing operation according to the first pesticide configuration information and risk area set to obtain the first agent, and perform drug mixing operation according to the second pesticide configuration information and agricultural pest trend range to obtain the second agent.

[0181] The process of performing drug ratio operation based on the first pesticide configuration information and the risk area set refers to the process of calculating the drug quality based on the types of pesticides in the first pesticide configuration information and the application area of ​​the risk area set.

[0182] The first agent refers to a drug applicable to a set of risk areas.

[0183] The process of mixing pesticides based on the second pesticide formulation information and the range of agricultural pest trends is the same as described above. The second pesticide refers to a pesticide suitable for the range of agricultural pest trends.

[0184] Specifically, in this embodiment of the invention, the drug storage device has an unlimited supply of various drugs by default and an automatic dispensing function. Therefore, it can dispense the first agent and an appropriate amount according to the first pesticide configuration information and the risk area set. Similarly, the second agent can be dispensed and an appropriate amount can be dispensed.

[0185] S8. Send the first and second agents to the application robot, and use the application robot to apply the agents to the risk area set and the range of agricultural pest trends.

[0186] The process of applying pesticides to the risk area set and the range of agricultural pest trends refers to the process of spraying a first pesticide when the pesticide application robot is in the risk area set, and spraying a second pesticide when it is in the range of agricultural pest trends.

[0187] In detail, in this embodiment of the invention, the step of using the spraying robot to perform spraying operations on the risk area set and the range of agricultural pest trends includes:

[0188] Using the spraying robot, a cruise operation is carried out on the target area farmland according to a preset flight route, and the location information of the spraying robot is obtained during the cruise operation.

[0189] Based on the set of risk areas and the range of agricultural pest trends, identify the reagent type corresponding to the location information;

[0190] The location information is obtained in the crop damage status matrix, and the target status score is obtained. Based on the target status score, the spraying speed of the pre-constructed spraying valve is adjusted to obtain the target spraying speed.

[0191] Based on the target spraying speed and reagent type, the first or second agent is selectively sprayed onto the location information.

[0192] The flight route refers to the working route configured by the company that enables the spraying robot to cover the target area of ​​farmland.

[0193] The location information refers to the real-time location of the drug delivery robot.

[0194] The reagent type refers to either the first reagent or the second reagent.

[0195] The target state score refers to the crop damage state score corresponding to the location information in the damage state matrix.

[0196] The spray valve refers to the outlet from which the drug-spraying robot sprays the drug.

[0197] The spraying speed adjustment refers to a method where the higher the target state score, the faster the spraying speed. The target spraying speed refers to the adjusted spraying speed of the application robot.

[0198] Specifically, in this embodiment of the invention, not only is a strategy of spraying pesticides in different areas adopted, but also an adaptive spraying method based on the degree of agricultural damage is adopted to further improve the accuracy of pesticide application and reduce the amount of pesticides used.

[0199] Specifically, the present invention first configures the flight path, and then when the spraying robot starts working along the flight path, it can change the type and speed of the sprayed drug through the location information, thereby achieving adaptive spraying.

[0200] In detail, in this embodiment of the invention, after using the spraying robot to perform spraying operations on the risk area set and the agricultural pest trend range, the method further includes:

[0201] Obtain the pesticide application records within the preset planting time period to obtain a set of pesticide application records;

[0202] The set of risk areas and the range of agricultural pest trends corresponding to each application are obtained from the application record set to obtain the applied areas;

[0203] Obtain the crop types in the farmland of the target area;

[0204] When the crop type is a preset cultivation type, superior plants in the target area farmland are obtained according to the area where the pesticide has been applied;

[0205] When the crop type is a preset commodity type, pesticide-free products in the target area farmland are obtained based on the area where pesticides have been applied.

[0206] The planting time period can be configured according to the type of crop. For example, the planting time for herbs can be configured as 3 years, while the planting time for other crops can be 3 months, 4 months, etc.

[0207] The set of pesticide application records refers to pesticide application records within a preset planting time period.

[0208] The "pesticide-treated area" refers to the area in the target farmland that is not "covered" by the pesticide-treated layer by mapping the various risk areas and the range of agricultural pest trends onto a single layer.

[0209] The crop types mentioned refer to scientific research or cultivation types, as well as economic crops and commodity types.

[0210] The superior plants refer to crops that have not been treated with pesticides. The pesticide-free products refer to crops that have not been treated with pesticides.

[0211] Specifically, in this embodiment of the invention, the application status of pesticides in the target area of ​​farmland can be understood through this solution, which plays an important role in scientific research and improving yields. For example, when the crop type is a cultivation type, finding untreated areas based on the treated areas indicates that the crop in that area has a high resistance to pests and diseases. When the crop type is a commercial type, finding untreated areas allows for independent harvesting, improving economic benefits.

[0212] To address the problems described in the background section, this invention first uses artificial intelligence to identify crop information sets, understanding basic information such as crop pests and growth status, and obtaining a pest matrix change sequence. Then, it identifies areas where pests occur using thresholds, obtaining a set of risk areas. This invention can control the application conditions of pesticides using fuzzy algorithms, thereby determining the appropriate timing for application. Furthermore, it transforms traditional large-scale pesticide application into three levels: therapeutic application, preventative application, and no application, thus significantly reducing the amount of pesticides used and achieving automated management for early detection and treatment. Therefore, this invention can improve the accuracy of pesticide application for crops.

[0213] like Figure 2 The diagram shown is a functional block diagram of an agricultural robot precision spraying system based on adaptive control, provided in an embodiment of the present invention.

[0214] The adaptive control-based precision pesticide application system 100 for agricultural robots described in this invention can be installed in an electronic device. Depending on the functions implemented, the adaptive control-based precision pesticide application system 100 may include a device acquisition module 101, a crop damage status identification module 102, a pesticide application suitability identification module 103, and an adaptive pesticide mixing and application module 104. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0215] The device acquisition module 101 is used to acquire an agricultural robot base station, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a drug application robot.

[0216] The crop damage status identification module 102 is used to acquire environmental information set and crop information set of farmland in the target area using a pre-constructed sensor cluster, and to use the edge processor to identify crop damage status score of crop information set to obtain crop damage status matrix, and to acquire historical crop damage status matrix within a preset pest and disease development time period to obtain crop damage matrix change sequence, and to identify risk area set in the crop damage matrix change sequence according to preset crop damage threshold, and to acquire the total crop damage intensity of the risk area set.

[0217] The pesticide application suitability identification module 103 is used to determine whether the total pesticide intensity is greater than a preset pesticide application threshold. When the total pesticide intensity is greater than the pesticide application threshold, the spraying suitability of the environmental information set is identified.

[0218] The adaptive pesticide application module 104 is used to identify the pesticide type of the crop information set when the spraying suitability is greater than a preset implementation condition threshold, obtain first pesticide configuration information and second pesticide configuration information, identify the agricultural pest development trend of the agricultural pest matrix change sequence, obtain the agricultural pest trend range, and use the pesticide storage device to perform pesticide ratio operation according to the first pesticide configuration information and the risk area set to obtain a first agent, and perform pesticide ratio operation according to the second pesticide configuration information and the agricultural pest trend range to obtain a second agent, and send the first agent and the second agent to the application robot, and use the application robot to perform pesticide application operation on the risk area set and the agricultural pest trend range.

[0219] In detail, the modules in the agricultural robot precision spraying system 100 based on adaptive control described in this embodiment of the invention employ the same methods as described above during use. Figure 1 The method used is the same as the precision pesticide application method for agricultural robots based on adaptive control described in the article, and can produce the same technical effect, so it will not be repeated here.

[0220] like Figure 3 The diagram shown is a structural schematic of an electronic device for implementing a precision pesticide application method for agricultural robots based on adaptive control, according to an embodiment of the present invention.

[0221] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a program for a precise pesticide application method for an agricultural robot based on adaptive control.

[0222] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a precision pesticide application method program for an agricultural robot based on adaptive control, but also to temporarily store data that has been output or will be output.

[0223] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a precision pesticide application method program for agricultural robots based on adaptive control) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0224] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0225] Figure 3 Only electronic devices with components are shown; those skilled in the art will understand that... Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0226] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0227] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0228] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0229] The program for a precision pesticide application method using an agricultural robot based on adaptive control, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0230] An agricultural robot base station is obtained, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a drug application robot;

[0231] Using a pre-built sensor cluster, environmental information sets and crop information sets of farmland in the target area are acquired. The edge processor is then used to identify crop damage status scores in the crop information set to obtain a damage status matrix. The historical damage status matrix within a preset pest and disease development time period is also acquired to obtain a damage matrix change sequence.

[0232] Based on a preset agricultural damage threshold, identify the set of risk regions in the agricultural damage matrix change sequence, and obtain the total agricultural damage intensity of the set of risk regions;

[0233] Determine whether the total agricultural pest intensity is greater than a preset pesticide application threshold. When the total agricultural pest intensity is greater than the pesticide application threshold, identify the spraying suitability of the environmental information set.

[0234] When the spraying suitability is greater than the preset implementation condition threshold, the crop information set is used to identify the pesticide type to obtain the first pesticide configuration information and the second pesticide configuration information.

[0235] Perform a pest development trend identification operation on the pest matrix change sequence to obtain the pest trend range;

[0236] Using the drug storage device, a drug ratio operation is performed based on the first pesticide configuration information and the risk area set to obtain a first agent, and a drug ratio operation is performed based on the second pesticide configuration information and the range of agricultural pest trends to obtain a second agent;

[0237] The first and second pesticides are sent to the application robot, and the application robot is used to apply pesticides to the risk area set and the range of agricultural pest trends.

[0238] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0239] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0240] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0241] An agricultural robot base station is obtained, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a drug application robot;

[0242] Using a pre-built sensor cluster, environmental information sets and crop information sets of farmland in the target area are acquired. The edge processor is then used to identify crop damage status scores in the crop information set to obtain a damage status matrix. The historical damage status matrix within a preset pest and disease development time period is also acquired to obtain a damage matrix change sequence.

[0243] Based on a preset agricultural damage threshold, identify the set of risk regions in the agricultural damage matrix change sequence, and obtain the total agricultural damage intensity of the set of risk regions;

[0244] Determine whether the total agricultural pest intensity is greater than a preset pesticide application threshold. When the total agricultural pest intensity is greater than the pesticide application threshold, identify the spraying suitability of the environmental information set.

[0245] When the spraying suitability is greater than the preset implementation condition threshold, the crop information set is used to identify the pesticide type to obtain the first pesticide configuration information and the second pesticide configuration information.

[0246] Perform a pest development trend identification operation on the pest matrix change sequence to obtain the pest trend range;

[0247] Using the drug storage device, a drug ratio operation is performed based on the first pesticide configuration information and the risk area set to obtain a first agent, and a drug ratio operation is performed based on the second pesticide configuration information and the range of agricultural pest trends to obtain a second agent;

[0248] The first and second pesticides are sent to the application robot, and the application robot is used to apply pesticides to the risk area set and the range of agricultural pest trends.

[0249] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0250] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0251] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, 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 modules.

[0252] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0253] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for precise pesticide application using an agricultural robot based on adaptive control, characterized in that, The method includes: An agricultural robot base station is obtained, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a drug application robot; Using a pre-built sensor cluster, environmental information sets and crop information sets of farmland in the target area are acquired. The edge processor is then used to identify crop damage status scores in the crop information set to obtain a damage status matrix. The historical damage status matrix within a preset pest and disease development time period is also acquired to obtain a damage matrix change sequence. Based on a preset agricultural damage threshold, identify the set of risk regions in the agricultural damage matrix change sequence, and obtain the total agricultural damage intensity of the set of risk regions; Determine whether the total agricultural pest intensity is greater than a preset pesticide application threshold. When the total agricultural pest intensity is greater than the pesticide application threshold, identify the spraying suitability of the environmental information set. When the spraying suitability is greater than the preset implementation condition threshold, the crop information set is used to identify the pesticide type to obtain the first pesticide configuration information and the second pesticide configuration information. Perform a pest development trend identification operation on the pest matrix change sequence to obtain the pest trend range; Using the drug storage device, a drug ratio operation is performed based on the first pesticide configuration information and the risk area set to obtain a first agent, and a drug ratio operation is performed based on the second pesticide configuration information and the range of agricultural pest trends to obtain a second agent; The first and second pesticides are sent to the application robot, and the application robot is used to apply pesticides to the risk area set and the range of agricultural pest trends.

2. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 1, characterized in that, The step of using the edge processor to perform crop damage status score identification on the crop information set to obtain a crop damage status matrix includes: Using the edge processor, growth nodes are identified in the crop information set to obtain crop growth nodes; Based on the crop growth nodes, abnormal growth progress is identified to obtain an abnormal growth progress score matrix; Insect identification is performed on the crop information set to obtain an insect quantity set and an insect type set. Based on the pre-configured insect type weight table, the influence coefficient corresponding to each insect type in the insect type set is queried to obtain the influence parameter set. The set of influencing parameters and the set of insect numbers are weighted according to insect type to obtain the insect interference score matrix; The agricultural damage state matrix is ​​obtained by summing the abnormal growth progress score matrix and the insect disturbance score matrix.

3. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 2, characterized in that, The step of identifying a set of risk regions in the agricultural damage matrix change sequence based on a preset agricultural damage threshold, and obtaining the total agricultural damage intensity of the risk region set, includes: Identify regions in the crop damage status matrix whose crop damage status scores are greater than a preset damage threshold to obtain a set of risk regions; The sum of crop damage status scores in each risk region of the risk region set is calculated to obtain the damage intensity set. The total agricultural damage intensity is obtained by summing the intensity of each agricultural damage in the set of agricultural damage intensities.

4. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 3, characterized in that, The identification of the spraying suitability of the environmental information set includes: Using pre-constructed hierarchical rules, each piece of environmental information in the environmental information set is evaluated based on a good-medium-bad level, resulting in a set of environmental assessment variables. Based on the pre-built set of membership functions, the pre-built Mamdani inference method is used to calculate the membership values ​​of the environmental assessment variable set based on the pre-built fuzzy rule base, so as to obtain the rule membership set. Identify the fuzzy rule with the highest membership value in the rule membership set to obtain the target fuzzy rule, and output the pre-constructed output decision in the target fuzzy rule to obtain the spraying suitability.

5. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 4, characterized in that, When the total pesticide damage intensity exceeds the pesticide application threshold, the method further includes: Obtain the application time interval; When the drug administration time interval is greater than the preset treatment period, the drug administration time interval is weighted according to the preset delay coefficient to obtain the time influence bias; Based on the time influence bias, time spraying rules are constructed and filled into the fuzzy rule base to obtain an updated fuzzy rule base.

6. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 5, characterized in that, The step of identifying pesticide types in the crop information set to obtain first pesticide configuration information and second pesticide configuration information includes: The crop information set is subjected to crop pest type identification to obtain crop pest type labels; Using a pre-constructed pesticide configuration table, a query operation is performed based on the agricultural pest type label to obtain the first pesticide configuration information; Based on the preset preventive pesticide application strategy, the first pesticide application information is fine-tuned to obtain the second pesticide application information.

7. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 6, characterized in that, The step of identifying crop pest types from the crop information set to obtain crop pest type labels includes: Obtain historical records of agricultural damage scenarios, wherein the historical records of agricultural damage scenarios include historical crop information and historical assessment tags; Using a pre-built regression network model, machine learning is performed on the historical agricultural damage scene records to obtain an agricultural damage type identification network; Using the aforementioned agricultural pest type identification network, the crop information set is subjected to text quantization processing to obtain a crop information vector set; The crop information vector set is subjected to a clustering operation based on the similarity of each crop pest type pre-learned in the crop pest type identification network to obtain the crop pest similarity score corresponding to each crop pest type. Extract the crop type corresponding to the crop type with the highest crop similarity score from the crop similarity scores of each crop type to obtain the crop type label.

8. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 7, characterized in that, The process of using the spraying robot to apply pesticides to the set of risk areas and the range of agricultural pest trends includes: Using the spraying robot, a cruise operation is carried out on the target area farmland according to a preset flight route, and the location information of the spraying robot is obtained during the cruise operation. Based on the set of risk areas and the range of agricultural pest trends, identify the reagent type corresponding to the location information; The location information is obtained in the crop damage status matrix, and the target status score is obtained. Based on the target status score, the spraying speed of the pre-constructed spraying valve is adjusted to obtain the target spraying speed. Based on the target spraying speed and reagent type, the first or second agent is selectively sprayed onto the location information.

9. The method for precise pesticide application using an agricultural robot based on adaptive control as described in claim 8, characterized in that, After using the spraying robot to apply pesticides to the set of risk areas and the range of agricultural pest trends, the method further includes: Obtain the pesticide application records within the preset planting time period to obtain a set of pesticide application records; The set of risk areas and the range of agricultural pest trends corresponding to each application are obtained from the application record set to obtain the applied areas; Obtain the crop types in the farmland of the target area; When the crop type is a preset cultivation type, superior plants in the target area farmland are obtained according to the area where the pesticide has been applied; When the crop type is a preset commodity type, pesticide-free products in the target area farmland are obtained based on the area where pesticides have been applied.

10. A precision pesticide application system for agricultural robots based on adaptive control, characterized in that, The system includes: The device acquisition module is used to acquire agricultural robot base stations, wherein the agricultural robot base station includes an edge processor, a drug storage device, and a spraying robot; The crop damage status identification module is used to acquire environmental information set and crop information set of farmland in the target area using a pre-constructed sensor cluster, and to use the edge processor to identify crop damage status score of crop information set to obtain crop damage status matrix, and to acquire historical crop damage status matrix within a preset pest and disease development time period to obtain crop damage matrix change sequence, and to identify risk area set in the crop damage matrix change sequence according to preset crop damage threshold, and to obtain the total crop damage intensity of the risk area set; The pesticide application suitability identification module is used to determine whether the total pesticide intensity is greater than a preset pesticide application threshold. When the total pesticide intensity is greater than the pesticide application threshold, the spraying suitability of the environmental information set is identified. An adaptive pesticide application module is used to identify pesticide types in the crop information set when the spraying suitability exceeds a preset implementation condition threshold, obtain first pesticide configuration information and second pesticide configuration information, identify the agricultural pest development trend in the agricultural pest matrix change sequence to obtain the agricultural pest trend range, and use the pesticide storage device to perform pesticide ratio operation based on the first pesticide configuration information and risk area set to obtain a first agent, and perform pesticide ratio operation based on the second pesticide configuration information and agricultural pest trend range to obtain a second agent, and send the first agent and second agent to the application robot, and use the application robot to perform pesticide application operation on the risk area set and agricultural pest trend range.

Images