A high-efficiency variable spraying method and system for plant protection UAV

By constructing a plant protection drone map model and ant colony algorithm, real-time variable spraying scheme is generated, and the problem that the spray system cannot be adjusted in real time in the existing technology is solved, and the pesticide utilization rate and spray effect are improved to meet the needs of complex plantation areas.

CN118426485BActive Publication Date: 2024-11-15PLANT PROTECTION RES INST OF GUANGDONG ACADEMY OF AGRI SCI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410518062.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-28
Publication Date
2024-11-15
Estimated Expiration
2044-04-28

AI Technical Summary

Technical Problem

The existing plant protection drone spray system cannot adjust the spray parameters in real time according to crop growth conditions and pest conditions, resulting in poor spraying effect, low pesticide utilization rate, and even adversely affecting the environment and crops.

Method used

By obtaining basic information and historical pest monitoring data of the planting area, building a map model, combining ant colony algorithm and shortest path algorithm, generating real-time variable spray schemes, adjusting spray paths and parameters, and ensuring the optimization of spray effect.

Benefits of technology

Improve the utilization rate and spray effect of pesticides, reduce the negative impact on the environment and crops, and adapt to the spray demand of complex plantation areas.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118426485B_ABST
    Figure CN118426485B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for efficient variable spraying of plant protection drones. A plant protection map model is constructed to divide the planting area into minimum areas to form multiple planting sub-areas; pest impact levels are divided for multiple planting sub-areas through pest prediction data; based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model, the starting point and the end point are set, and the planting sub-areas are used as the mobile unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively. Based on the path analysis, the optimal path information and the real-time variable spraying scheme are generated. Through the invention, the pesticide spray can be maximized to meet the spraying needs of a certain area while ensuring the basic pesticide spraying needs, thereby improving the utilization rate of plant protection resources and having a better improvement effect in complex plantation areas.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of plant protection data analysis, and more specifically, to a method and system for high-efficiency variable spraying of a plant protection unmanned aerial vehicle. Background Art

[0002] With the advancement of agricultural modernization, plant protection drones are increasingly used in agricultural production. However, existing plant protection drone spray systems often use fixed parameters for spraying operations, and are unable to adjust spray parameters and real-time path planning analysis in real time according to crop growth conditions, pest and disease occurrence, etc., resulting in poor spraying effects, low pesticide utilization, and even adverse effects on the environment and crops. Therefore, the development of a plant protection drone spray system that can intelligently adjust spray parameters has become an urgent problem to be solved in the current agricultural production field. Summary of the invention

[0003] The present invention overcomes the defects of the prior art and proposes a highly efficient variable spraying method and system for a plant protection UAV.

[0004] The first aspect of the present invention provides a plant protection UAV efficient variable spraying method, comprising:

[0005] Obtain basic information and spray parameter information of the plant protection drone in the preset planting area, build a map model based on the basic information, and divide the planting area into minimum areas through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas;

[0006] Obtain historical pest monitoring data, import the historical pest monitoring data into the LSTM-based prediction model to predict pest trends, obtain pest prediction data, and classify pest impact levels for multiple planting sub-areas based on the pest prediction data;

[0007] Acquire aerial image data of a preset planting area, perform regional density analysis based on the aerial image data, and associate planting density data for multiple planting sub-areas based on the analysis results;

[0008] Based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model. The starting and ending positions are set, and the planting sub-area is used as the moving unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively;

[0009] Based on the map model, the first path information and the second path information are path-fused to generate a fused path area, and based on the shortest path algorithm, path simulation analysis is performed on the fused path area to generate optimal path information;

[0010] Based on the dimensions of pest impact and planting density, the variable parameters of plant protection spraying are analyzed in the fused path area, and a real-time variable spraying plan is generated by combining the optimal path information.

[0011] In this solution, the basic information and the spray parameter information of the plant protection drone in the preset planting area are obtained, a map model is constructed based on the basic information, and the planting area is divided into the smallest area through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas, specifically:

[0012] Obtain basic information in the preset planting area and spray parameter information of the plant protection drone;

[0013] The basic information includes the area of ​​the preset planting area, the map outline, and the planting distribution information; the spray parameter information of the plant protection drone includes the spray range, spray height, and spray volume information;

[0014] Construct a two-dimensional visual map model based on the preset planting area based on basic information;

[0015] Based on the spray range of the plant protection drone, a minimum area range is set, and the planting area is divided into the minimum area through the map model to form multiple planting sub-areas to ensure that the area of ​​each planting sub-area meets the minimum area range.

[0016] In this solution, the historical pest monitoring data is obtained, and the historical pest monitoring data is imported into the LSTM-based prediction model to predict pest trends, and pest prediction data is obtained. The pest impact level of multiple planting sub-areas is divided according to the pest prediction data, specifically:

[0017] Acquire historical pest monitoring data, wherein the historical pest monitoring data includes pest species, quantity, and distribution data in a preset planting area;

[0018] Based on a linear regression algorithm, a linear change analysis of the historical pest monitoring data based on a time dimension is performed, and data with a linear change higher than a preset growth rate is marked to form first monitoring data;

[0019] The first monitoring data is preprocessed by data standardization and time serialization to form sequence data, and the sequence data is imported into a prediction model based on LSTM to perform pest trend prediction to obtain pest prediction data;

[0020] The pest prediction data and historical pest monitoring data are integrated to form monitoring big data. The pest distribution analysis and pest data volume trend analysis are carried out on multiple planting sub-areas based on the monitoring big data. Based on the pest change trend, each planting sub-area is divided into pest impact levels, and each planting sub-area corresponds to a pest impact level.

[0021] In this solution, the aerial image data of the preset planting area is obtained, regional density analysis is performed based on the aerial image data, and planting density data association is performed on multiple planting sub-areas based on the analysis results, specifically:

[0022] Based on the drone camera device, obtain aerial image data of the preset planting area;

[0023] Based on the CNN image recognition model, color feature extraction and contour feature extraction are performed on the aerial image data, and the obtained feature data are used to identify plant species and analyze planting density. The planting density analysis is based on color features and contour features, and the planting sub-area is used as the identification and analysis unit to obtain the planting density data of each planting sub-area.

[0024] In this scheme, the ant colony algorithm is used to simulate the spray path of the plant protection drone through the map model, set the starting and end positions, take the planting sub-area as the mobile unit path, perform two path simulations, one based on the pest impact level as the initial pheromone basis, and the second based on the planting density as the initial pheromone basis, and generate the first path information and the second path information respectively, specifically:

[0025] Based on the ant colony algorithm, the spray path of the plant protection drone is simulated in the map model, the preset starting point and the preset end point are set, and each planting sub-area is used as the moving unit path;

[0026] During the first path simulation, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the pest impact level, and the pheromone is proportional to the pest impact level.

[0027] The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the first path information.

[0028] In the second path simulation process, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the planting density value, and the pheromone is proportional to the planting density value.

[0029] The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the second path information.

[0030] In this solution, the first path information and the second path information are fused based on the map model to generate a fused path area, and the fused path area is simulated and analyzed based on the shortest path algorithm to generate the optimal path information, specifically:

[0031] By using the map model, path fusion is performed based on the first path information and the second path information to form a fused path;

[0032] In the map model, the path area of ​​the fusion path is extracted, and the obtained multiple planting sub-areas are regionally integrated to form a fusion path area;

[0033] In the map model, based on the shortest path algorithm, the shortest path simulation analysis is performed on the fused path area, and it is ensured that the path can pass through all the planting sub-areas under all the fused path areas, and the optimal path information is generated.

[0034] In this scheme, based on the pest impact and planting density dimensions, the variable parameters of plant protection spray are analyzed for the fusion path area, and a real-time variable spray scheme is generated in combination with the optimal path information, specifically:

[0035] Based on the fusion path area, the pest impact level and planting density information of each planting sub-area are obtained and integrated into the planting basic monitoring data;

[0036] Based on the basic planting monitoring data, the spray pesticide ratio and spray variable demand analysis are carried out for each planting sub-area in the fusion path area, and the real-time variable spray plan is generated in combination with the optimal path information;

[0037] The real-time variable spraying scheme and the optimal path information are sent to the plant protection UAV terminal.

[0038] The second aspect of the present invention further provides a plant protection UAV efficient variable spray system, the system comprising: a memory, a processor, the memory comprising a plant protection UAV efficient variable spray program, the plant protection UAV efficient variable spray program when executed by the processor to implement the following steps:

[0039] Obtain basic information and spray parameter information of the plant protection drone in the preset planting area, build a map model based on the basic information, and divide the planting area into minimum areas through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas;

[0040] Obtain historical pest monitoring data, import the historical pest monitoring data into the LSTM-based prediction model to predict pest trends, obtain pest prediction data, and classify pest impact levels for multiple planting sub-areas based on the pest prediction data;

[0041] Acquire aerial image data of a preset planting area, perform regional density analysis based on the aerial image data, and associate planting density data for multiple planting sub-areas based on the analysis results;

[0042] Based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model. The starting and ending positions are set, and the planting sub-area is used as the moving unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively;

[0043] Based on the map model, the first path information and the second path information are path-fused to generate a fused path area, and based on the shortest path algorithm, path simulation analysis is performed on the fused path area to generate optimal path information;

[0044] Based on the dimensions of pest impact and planting density, the variable parameters of plant protection spraying are analyzed in the fused path area, and a real-time variable spraying plan is generated by combining the optimal path information.

[0045] The third aspect of the present invention also provides a computer-readable storage medium, which includes a high-efficiency variable spray program for a plant protection drone. When the high-efficiency variable spray program for a plant protection drone is executed by a processor, the steps of the high-efficiency variable spray method for a plant protection drone as described in any of the above items are implemented.

[0046] The present invention discloses a method and system for efficient variable spraying of plant protection drones. A plant protection map model is constructed to divide the planting area into minimum areas to form multiple planting sub-areas; pest impact levels are divided for multiple planting sub-areas through pest prediction data; based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model, the starting point and the end point are set, and the planting sub-areas are used as the mobile unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively. Based on the path analysis, the optimal path information and the real-time variable spraying scheme are generated. Through the invention, the pesticide spray can be maximized to meet the spraying needs of a certain area while ensuring the basic pesticide spraying needs, thereby improving the utilization rate of plant protection resources and having a better improvement effect in complex plantation areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A flow chart showing a highly efficient variable spraying method of a plant protection UAV according to the present invention is shown;

[0048] Figure 2 A block diagram of a high-efficiency variable spray system of a plant protection UAV of the present invention is shown. DETAILED DESCRIPTION

[0049] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0050] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0051] Figure 1 A flow chart of a highly efficient variable spraying method of a plant protection UAV of the present invention is shown.

[0052] like Figure 1 As shown, the first aspect of the present invention provides a plant protection UAV efficient variable spraying method, comprising:

[0053] S102, obtaining basic information and spray parameter information of the plant protection drone in a preset planting area, constructing a map model based on the basic information, combining the spray parameter information of the plant protection drone, dividing the planting area into minimum areas through the map model, and forming multiple planting sub-areas;

[0054] S104, obtaining historical pest monitoring data, importing the historical pest monitoring data into a prediction model based on LSTM to perform pest trend prediction, obtaining pest prediction data, and classifying pest impact levels for multiple planting sub-areas based on the pest prediction data;

[0055] S106, acquiring aerial image data of a preset planting area, performing regional density analysis based on the aerial image data, and associating planting density data of multiple planting sub-areas based on the analysis results;

[0056] S108, based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model, the starting point and the end point are set, the planting sub-area is used as the moving unit path, and two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively;

[0057] S110, performing path fusion on the first path information and the second path information based on the map model to generate a fused path area, performing path simulation analysis on the fused path area based on a shortest path algorithm, and generating optimal path information;

[0058] S112, based on the dimensions of pest impact and planting density, analyze the variable parameters of plant protection spray in the fused path area, and generate a real-time variable spray plan in combination with the optimal path information.

[0059] According to an embodiment of the present invention, the basic information and the spray parameter information of the plant protection drone in the preset planting area are obtained, a map model is constructed based on the basic information, and the planting area is divided into the smallest area through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas, specifically:

[0060] Obtain basic information in the preset planting area and spray parameter information of the plant protection drone;

[0061] The basic information includes the area of ​​the preset planting area, the map outline, and the planting distribution information; the spray parameter information of the plant protection drone includes the spray range, spray height, and spray volume information;

[0062] Construct a two-dimensional visual map model based on the preset planting area based on basic information;

[0063] Based on the spray range of the plant protection drone, a minimum area range is set, and the planting area is divided into the minimum area through the map model to form multiple planting sub-areas to ensure that the area of ​​each planting sub-area meets the minimum area range.

[0064] It should be noted that the map model is used for drone path simulation and planting visualization analysis. The minimum area range is set based on the spray range, generally 2 to 3 times the spray range area. For example, a certain area value can be set and the preset threshold can be floated up and down. The determined area value is 2 to 3 times the spray range area. The minimum area is set to divide multiple small areas for path analysis to ensure the precision of the path and the efficiency of the path analysis results for plant protection spraying.

[0065] According to an embodiment of the present invention, the historical pest monitoring data is obtained, and the historical pest monitoring data is imported into the LSTM-based prediction model to predict pest trends, thereby obtaining pest prediction data. The pest impact levels of multiple planting sub-areas are classified according to the pest prediction data, specifically:

[0066] Acquire historical pest monitoring data, wherein the historical pest monitoring data includes pest species, quantity, and distribution data in a preset planting area;

[0067] Based on a linear regression algorithm, a linear change analysis of the historical pest monitoring data based on a time dimension is performed, and data with a linear change higher than a preset growth rate is marked to form first monitoring data;

[0068] The first monitoring data is preprocessed by data standardization and time serialization to form sequence data, and the sequence data is imported into a prediction model based on LSTM to perform pest trend prediction to obtain pest prediction data;

[0069] The pest prediction data and historical pest monitoring data are integrated to form monitoring big data. The pest distribution analysis and pest data volume trend analysis are carried out on multiple planting sub-areas based on the monitoring big data. Based on the pest change trend, each planting sub-area is divided into pest impact levels, and each planting sub-area corresponds to a pest impact level.

[0070] It should be noted that the linear change is higher than the preset growth rate, that is, the linear growth rate or the linear decline rate is higher than the preset growth rate. By mining the first monitoring data, data with corresponding changes in the pest data can be extracted, such as pest quantity, trend, distribution and other data. The pest impact level can be divided into N levels, where N is a preset size value. The larger the level, the greater the impact, the more obvious the pest growth trend, and the greater the corresponding plant protection demand.

[0071] According to an embodiment of the present invention, the aerial image data of a preset planting area is obtained, regional density analysis is performed based on the aerial image data, and planting density data association is performed on multiple planting sub-areas based on the analysis results, specifically:

[0072] Based on the drone camera device, obtain aerial image data of the preset planting area;

[0073] Based on the CNN image recognition model, color feature extraction and contour feature extraction are performed on the aerial image data, and the obtained feature data are used to identify plant species and analyze planting density. The planting density analysis is based on color features and contour features, and the planting sub-area is used as the identification and analysis unit to obtain the planting density data of each planting sub-area.

[0074] It should be noted that the CNN image recognition model can be used to perform planting analysis on the plant protection area, and the planting density can be further calculated through color feature analysis. Each planting sub-area includes a planting density value.

[0075] According to an embodiment of the present invention, the ant colony algorithm is used to simulate the spray path of the plant protection drone through a map model, set the starting point and the end point, use the planting sub-area as the mobile unit path, perform two path simulations, one based on the pest impact level as the initial pheromone basis, and the second based on the planting density as the initial pheromone basis, and generate the first path information and the second path information respectively, specifically:

[0076] Based on the ant colony algorithm, the spray path of the plant protection drone is simulated in the map model, the preset starting point and the preset end point are set, and each planting sub-area is used as the moving unit path;

[0077] During the first path simulation, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the pest impact level, and the pheromone is proportional to the pest impact level.

[0078] The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the first path information.

[0079] In the second path simulation process, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the planting density value, and the pheromone is proportional to the planting density value.

[0080] The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the second path information.

[0081] It should be noted that during the first path simulation, the initialization pheromone of each mobile unit path is proportional to the corresponding pest impact level, and the initialization pheromone is equal to the pest impact level multiplied by a preset coefficient. During the second path simulation, the initialization pheromone is equal to the planting density value multiplied by a preset coefficient.

[0082] According to an embodiment of the present invention, the first path information and the second path information are fused based on a map model to generate a fused path area, a path simulation analysis is performed on the fused path area based on a shortest path algorithm, and optimal path information is generated, specifically:

[0083] By using the map model, path fusion is performed based on the first path information and the second path information to form a fused path;

[0084] In the map model, the path area of ​​the fusion path is extracted, and the obtained multiple planting sub-areas are regionally integrated to form a fusion path area;

[0085] In the map model, based on the shortest path algorithm, the shortest path simulation analysis is performed on the fused path area, and it is ensured that the path can pass through all the planting sub-areas under all the fused path areas, and the optimal path information is generated.

[0086] It should be noted that the fused path is a path set of the first path and the second path, and the repeated path segments are removed. The path area is a planted sub-area of ​​the path. The shortest path algorithm includes Floyd, Dijkstra and other algorithms.

[0087] It is worth mentioning that under the complex planting environment and plant protection requirements, the existing technology is difficult to achieve an efficient plant protection drone spraying solution, especially for regional requirements with different spray requirements. The existing technology lacks a targeted route planning solution. The traditional technology is based on a fixed amount of spraying solution, and the same spray form and spray amount are adopted in any area. There is a lack of dynamic planning for plant protection areas, and the adaptability is poor. In the present invention, by dividing the planting area into regions, two spray routes are obtained based on the analysis of pest trends and planting density. The first route is planned and analyzed based on trends, and the second route is planned and analyzed based on planting density. Finally, the shortest path algorithm is used to simulate and analyze the optimal path. Through the present invention, it is possible to maximize the use of pesticide spray to meet the spray requirements of a certain area while ensuring the basic pesticide spray requirements, improve the utilization rate of plant protection resources, and have a better improvement effect in complex planting areas.

[0088] According to an embodiment of the present invention, based on the pest impact and planting density dimensions, the plant protection spray variable parameter analysis is performed on the fusion path area, and a real-time variable spray scheme is generated in combination with the optimal path information, specifically:

[0089] Based on the fusion path area, the pest impact level and planting density information of each planting sub-area are obtained and integrated into the planting basic monitoring data;

[0090] Based on the basic planting monitoring data, the spray pesticide ratio and spray variable demand analysis are carried out for each planting sub-area in the fusion path area, and the real-time variable spray plan is generated in combination with the optimal path information;

[0091] The real-time variable spraying scheme and the optimal path information are sent to the plant protection UAV terminal.

[0092] It should be noted that the spray variable demand analysis includes parameters such as pesticide ratio parameters, pesticide types, spray volume, spray range, spray height, spray duration, etc. The real-time variable spray scheme includes spray variable parameters for each sub-area in the path.

[0093] Figure 2 A block diagram of a high-efficiency variable spray system of a plant protection UAV of the present invention is shown.

[0094] The second aspect of the present invention further provides a plant protection UAV efficient variable spray system 2, the system comprising: a memory 21, a processor 22, the memory comprising a plant protection UAV efficient variable spray program, the plant protection UAV efficient variable spray program is executed by the processor to implement the following steps:

[0095] Obtain basic information and spray parameter information of the plant protection drone in the preset planting area, build a map model based on the basic information, and divide the planting area into minimum areas through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas;

[0096] Obtain historical pest monitoring data, import the historical pest monitoring data into the LSTM-based prediction model to predict pest trends, obtain pest prediction data, and classify pest impact levels for multiple planting sub-areas based on the pest prediction data;

[0097] Acquire aerial image data of a preset planting area, perform regional density analysis based on the aerial image data, and associate planting density data for multiple planting sub-areas based on the analysis results;

[0098] Based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model. The starting and ending positions are set, and the planting sub-area is used as the moving unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively;

[0099] Based on the map model, the first path information and the second path information are path-fused to generate a fused path area, and based on the shortest path algorithm, path simulation analysis is performed on the fused path area to generate optimal path information;

[0100] Based on the dimensions of pest impact and planting density, the variable parameters of plant protection spraying are analyzed in the fused path area, and a real-time variable spraying plan is generated by combining the optimal path information.

[0101] According to an embodiment of the present invention, the basic information and the spray parameter information of the plant protection drone in the preset planting area are obtained, a map model is constructed based on the basic information, and the planting area is divided into the smallest area through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas, specifically:

[0102] Obtain basic information in the preset planting area and spray parameter information of the plant protection drone;

[0103] The basic information includes the area of ​​the preset planting area, the map outline, and the planting distribution information; the spray parameter information of the plant protection drone includes the spray range, spray height, and spray volume information;

[0104] Construct a two-dimensional visual map model based on the preset planting area based on basic information;

[0105] Based on the spray range of the plant protection drone, a minimum area range is set, and the planting area is divided into the minimum area through the map model to form multiple planting sub-areas to ensure that the area of ​​each planting sub-area meets the minimum area range.

[0106] It should be noted that the map model is used for drone path simulation and planting visualization analysis. The minimum area range is set based on the spray range, generally 2 to 3 times the spray range area. For example, a certain area value can be set and the preset threshold can be floated up and down. The determined area value is 2 to 3 times the spray range area. The minimum area is set to divide multiple small areas for path analysis to ensure the precision of the path and the efficiency of the path analysis results for plant protection spraying.

[0107] According to an embodiment of the present invention, the historical pest monitoring data is obtained, and the historical pest monitoring data is imported into the LSTM-based prediction model to predict pest trends, thereby obtaining pest prediction data. The pest impact levels of multiple planting sub-areas are classified according to the pest prediction data, specifically:

[0108] Acquire historical pest monitoring data, wherein the historical pest monitoring data includes pest species, quantity, and distribution data in a preset planting area;

[0109] Based on a linear regression algorithm, a linear change analysis of the historical pest monitoring data based on a time dimension is performed, and data with a linear change higher than a preset growth rate is marked to form first monitoring data;

[0110] The first monitoring data is preprocessed by data standardization and time serialization to form sequence data, and the sequence data is imported into a prediction model based on LSTM to perform pest trend prediction to obtain pest prediction data;

[0111] The pest prediction data and historical pest monitoring data are integrated to form monitoring big data. The pest distribution analysis and pest data volume trend analysis are carried out on multiple planting sub-areas based on the monitoring big data. Based on the pest change trend, each planting sub-area is divided into pest impact levels, and each planting sub-area corresponds to a pest impact level.

[0112] It should be noted that the linear change is higher than the preset growth rate, that is, the linear growth rate or the linear decline rate is higher than the preset growth rate. By mining the first monitoring data, data with corresponding changes in the pest data can be extracted, such as pest quantity, trend, distribution and other data. The pest impact level can be divided into N levels, where N is a preset size value. The larger the level, the greater the impact, the more obvious the pest growth trend, and the greater the corresponding plant protection demand.

[0113] According to an embodiment of the present invention, the aerial image data of a preset planting area is obtained, regional density analysis is performed based on the aerial image data, and planting density data association is performed on multiple planting sub-areas based on the analysis results, specifically:

[0114] Based on the drone camera device, obtain aerial image data of the preset planting area;

[0115] Based on the CNN image recognition model, color feature extraction and contour feature extraction are performed on the aerial image data, and the obtained feature data are used to identify plant species and analyze planting density. The planting density analysis is based on color features and contour features, and the planting sub-area is used as the identification and analysis unit to obtain the planting density data of each planting sub-area.

[0116] It should be noted that the CNN image recognition model can be used to perform planting analysis on the plant protection area, and the planting density can be further calculated through color feature analysis. Each planting sub-area includes a planting density value.

[0117] According to an embodiment of the present invention, the ant colony algorithm is used to simulate the spray path of the plant protection drone through a map model, set the starting point and the end point, use the planting sub-area as the mobile unit path, perform two path simulations, one based on the pest impact level as the initial pheromone basis, and the second based on the planting density as the initial pheromone basis, and generate the first path information and the second path information respectively, specifically:

[0118] Based on the ant colony algorithm, the spray path of the plant protection drone is simulated in the map model, the preset starting point and the preset end point are set, and each planting sub-area is used as the moving unit path;

[0119] During the first path simulation, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the pest impact level, and the pheromone is proportional to the pest impact level.

[0120] The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the first path information.

[0121] In the second path simulation process, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the planting density value, and the pheromone is proportional to the planting density value.

[0122] The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the second path information.

[0123] It should be noted that during the first path simulation, the initialization pheromone of each mobile unit path is proportional to the corresponding pest impact level, and the initialization pheromone is equal to the pest impact level multiplied by a preset coefficient. During the second path simulation, the initialization pheromone is equal to the planting density value multiplied by a preset coefficient.

[0124] According to an embodiment of the present invention, the first path information and the second path information are fused based on a map model to generate a fused path area, a path simulation analysis is performed on the fused path area based on a shortest path algorithm, and optimal path information is generated, specifically:

[0125] By using the map model, path fusion is performed based on the first path information and the second path information to form a fused path;

[0126] In the map model, the path area of ​​the fusion path is extracted, and the obtained multiple planting sub-areas are regionally integrated to form a fusion path area;

[0127] In the map model, based on the shortest path algorithm, the shortest path simulation analysis is performed on the fused path area, and it is ensured that the path can pass through all the planting sub-areas under all the fused path areas, and the optimal path information is generated.

[0128] It should be noted that the fused path is a path set of the first path and the second path, and the repeated path segments are removed. The path area is a planted sub-area of ​​the path. The shortest path algorithm includes Floyd, Dijkstra and other algorithms.

[0129] It is worth mentioning that under the complex planting environment and plant protection requirements, the existing technology is difficult to achieve an efficient plant protection drone spraying solution, especially for regional requirements with different spray requirements. The existing technology lacks a targeted route planning solution. The traditional technology is based on a fixed amount of spraying solution, and the same spray form and spray amount are adopted in any area. There is a lack of dynamic planning for plant protection areas, and the adaptability is poor. In the present invention, by dividing the planting area into regions, two spray routes are obtained based on the analysis of pest trends and planting density. The first route is planned and analyzed based on trends, and the second route is planned and analyzed based on planting density. Finally, the shortest path algorithm is used to simulate and analyze the optimal path. Through the present invention, it is possible to maximize the use of pesticide spray to meet the spray requirements of a certain area while ensuring the basic pesticide spray requirements, improve the utilization rate of plant protection resources, and have a better improvement effect in complex planting areas.

[0130] According to an embodiment of the present invention, based on the pest impact and planting density dimensions, the plant protection spray variable parameter analysis is performed on the fusion path area, and a real-time variable spray scheme is generated in combination with the optimal path information, specifically:

[0131] Based on the fusion path area, the pest impact level and planting density information of each planting sub-area are obtained and integrated into the planting basic monitoring data;

[0132] Based on the basic planting monitoring data, the spray pesticide ratio and spray variable demand analysis are carried out for each planting sub-area in the fusion path area, and the real-time variable spray plan is generated in combination with the optimal path information;

[0133] The real-time variable spraying scheme and the optimal path information are sent to the plant protection UAV terminal.

[0134] It should be noted that the spray variable demand analysis includes parameters such as pesticide ratio parameters, pesticide types, spray volume, spray range, spray height, spray duration, etc. The real-time variable spray scheme includes spray variable parameters for each sub-area in the path.

[0135] The third aspect of the present invention also provides a computer-readable storage medium, which includes a high-efficiency variable spray program for a plant protection drone. When the high-efficiency variable spray program for a plant protection drone is executed by a processor, the steps of the high-efficiency variable spray method for a plant protection drone as described in any of the above items are implemented.

[0136] The present invention discloses a method and system for efficient variable spraying of plant protection drones. A plant protection map model is constructed to divide the planting area into minimum areas to form multiple planting sub-areas; pest impact levels are divided for multiple planting sub-areas through pest prediction data; based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model, the starting point and the end point are set, and the planting sub-areas are used as the mobile unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively. Based on the path analysis, the optimal path information and the real-time variable spraying scheme are generated. Through the invention, the pesticide spray can be maximized to meet the spraying needs of a certain area while ensuring the basic pesticide spraying needs, thereby improving the utilization rate of plant protection resources and having a better improvement effect in complex plantation areas.

[0137] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0138] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0139] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0140] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical disks, and other media that can store program codes.

[0141] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0142] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A highly efficient variable spraying method for plant protection drones, characterized in that: include: Obtain basic information and spray parameter information of the plant protection drone in the preset planting area, build a map model based on the basic information, and divide the planting area into minimum areas through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas; Obtain historical pest monitoring data, import the historical pest monitoring data into the LSTM-based prediction model to predict pest trends, obtain pest prediction data, and classify pest impact levels for multiple planting sub-areas based on the pest prediction data; Acquire aerial image data of a preset planting area, perform regional density analysis based on the aerial image data, and associate planting density data for multiple planting sub-areas based on the analysis results; Based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model. The starting and ending positions are set, and the planting sub-area is used as the moving unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively; Based on the map model, the first path information and the second path information are path-fused to generate a fused path area, and based on the shortest path algorithm, path simulation analysis is performed on the fused path area to generate optimal path information; Based on the dimensions of pest impact and planting density, the variable parameters of plant protection spray are analyzed for the fused path area, and a real-time variable spray plan is generated by combining the optimal path information; The method of acquiring aerial image data of a preset planting area, performing regional density analysis based on the aerial image data, and associating planting density data of multiple planting sub-areas based on the analysis results is specifically as follows: Based on the drone camera device, obtain aerial image data of the preset planting area; Based on the CNN image recognition model, color feature extraction and contour feature extraction are performed on the aerial image data, and the obtained feature data are used to identify plant species and analyze planting density. The planting density analysis is based on color features and contour features, and the planting sub-area is used as the identification and analysis unit to obtain the planting density data of each planting sub-area.

2. The high-efficiency variable spraying method of a crop protection drone according to claim 1, characterized in that: The basic information and the spray parameter information of the plant protection drone in the preset planting area are obtained, a map model is constructed based on the basic information, and the planting area is divided into the smallest area through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas, specifically: Obtain basic information in the preset planting area and spray parameter information of the plant protection drone; The basic information includes the area of ​​the preset planting area, the map outline, and the planting distribution information; the spray parameter information of the plant protection drone includes the spray range, spray height, and spray volume information; Construct a two-dimensional visual map model based on the preset planting area based on basic information; Based on the spray range of the plant protection drone, a minimum area range is set, and the planting area is divided into the minimum area through the map model to form multiple planting sub-areas to ensure that the area of ​​each planting sub-area meets the minimum area range.

3. The high-efficiency variable spraying method of a crop protection drone according to claim 1, characterized in that: The historical pest monitoring data is obtained, and the historical pest monitoring data is imported into the LSTM-based prediction model to predict pest trends, thereby obtaining pest prediction data. The pest impact levels of multiple planting sub-areas are classified according to the pest prediction data, specifically: Acquire historical pest monitoring data, wherein the historical pest monitoring data includes pest species, quantity, and distribution data in a preset planting area; Based on a linear regression algorithm, a linear change analysis of the historical pest monitoring data based on a time dimension is performed, and data with a linear change higher than a preset growth rate is marked to form first monitoring data; The first monitoring data is preprocessed by data standardization and time serialization to form sequence data, and the sequence data is imported into a prediction model based on LSTM to perform pest trend prediction to obtain pest prediction data; The pest prediction data and historical pest monitoring data are integrated to form monitoring big data. The pest distribution analysis and pest data volume trend analysis are carried out on multiple planting sub-areas based on the monitoring big data. Based on the pest change trend, each planting sub-area is divided into pest impact levels, and each planting sub-area corresponds to a pest impact level.

4. The method for high-efficiency variable spraying by a plant protection drone according to claim 1, characterized in that: Based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model, the starting point and the end point are set, the planting sub-area is used as the mobile unit path, and two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively, specifically: Based on the ant colony algorithm, the spray path of the plant protection drone is simulated in the map model, the preset starting point and the preset end point are set, and each planting sub-area is used as the moving unit path; During the first path simulation, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the pest impact level, and the pheromone is proportional to the pest impact level. The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the first path information. In the second path simulation process, the pheromone of the path is initialized for each planting sub-area. The pheromone is calculated based on the planting density value, and the pheromone is proportional to the planting density value. The simulation process sets N ants to perform ant colony movement simulation, and repeats the ant colony movement simulation multiple times. After each ant colony movement is completed, the path pheromone is updated in real time based on the movement result until the final path does not change. The final path is recorded to obtain the second path information.

5. The high-efficiency variable spraying method of a crop protection drone according to claim 1, characterized in that: The first path information and the second path information are fused based on the map model to generate a fused path area, and the fused path area is simulated and analyzed based on the shortest path algorithm to generate optimal path information, specifically: By using the map model, path fusion is performed based on the first path information and the second path information to form a fused path; In the map model, the path area of ​​the fusion path is extracted, and the obtained multiple planting sub-areas are regionally integrated to form a fusion path area; In the map model, based on the shortest path algorithm, the shortest path simulation analysis is performed on the fused path area, and it is ensured that the path can pass through all the planting sub-areas under all the fused path areas, and the optimal path information is generated.

6. The high-efficiency variable spraying method of a crop protection drone according to claim 1, characterized in that: Based on the pest impact and planting density dimensions, the plant protection spray variable parameter analysis is performed on the fusion path area, and a real-time variable spray plan is generated in combination with the optimal path information, specifically: Based on the fusion path area, the pest impact level and planting density information of each planting sub-area are obtained and integrated into the planting basic monitoring data; Based on the basic planting monitoring data, the spray pesticide ratio and spray variable demand analysis are carried out for each planting sub-area in the fusion path area, and the real-time variable spray plan is generated in combination with the optimal path information; The real-time variable spraying scheme and the optimal path information are sent to the plant protection UAV terminal.

7. A high-efficiency variable spray system for crop protection drones, characterized in that: The system includes: a memory and a processor. The memory includes a high-efficiency variable spray program for a plant protection drone. When the high-efficiency variable spray program for a plant protection drone is executed by the processor, the following steps are implemented: Obtain basic information and spray parameter information of the plant protection drone in the preset planting area, build a map model based on the basic information, and divide the planting area into minimum areas through the map model in combination with the spray parameter information of the plant protection drone to form multiple planting sub-areas; Obtain historical pest monitoring data, import the historical pest monitoring data into the LSTM-based prediction model to predict pest trends, obtain pest prediction data, and classify pest impact levels for multiple planting sub-areas based on the pest prediction data; Acquire aerial image data of a preset planting area, perform regional density analysis based on the aerial image data, and associate planting density data for multiple planting sub-areas based on the analysis results; Based on the ant colony algorithm, the spray path of the plant protection drone is simulated through the map model. The starting and ending positions are set, and the planting sub-area is used as the moving unit path. Two path simulations are performed, one using the pest impact level as the initial pheromone basis, and the second using the planting density as the initial pheromone basis, and the first path information and the second path information are generated respectively; Based on the map model, the first path information and the second path information are path-fused to generate a fused path area, and based on the shortest path algorithm, path simulation analysis is performed on the fused path area to generate optimal path information; Based on the dimensions of pest impact and planting density, the variable parameters of plant protection spray are analyzed for the fused path area, and a real-time variable spray plan is generated by combining the optimal path information; The method of acquiring aerial image data of a preset planting area, performing regional density analysis based on the aerial image data, and associating planting density data of multiple planting sub-areas based on the analysis results is specifically as follows: Based on the drone camera device, obtain aerial image data of the preset planting area; Based on the CNN image recognition model, color feature extraction and contour feature extraction are performed on the aerial image data, and the obtained feature data are used to identify plant species and analyze planting density. The planting density analysis is based on color features and contour features, and the planting sub-area is used as the identification and analysis unit to obtain the planting density data of each planting sub-area.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a high-efficiency variable spray program for a plant protection UAV. When the high-efficiency variable spray program for a plant protection UAV is executed by a processor, the steps of the high-efficiency variable spray method for a plant protection UAV as described in any one of claims 1 to 6 are implemented.