Intelligent field patrol method, device, equipment, system, medium and product

Abstract

The invention discloses an intelligent field patrol method, device, equipment, system, medium and product, which are applied to the technical field of intelligent agriculture, and the method comprises the following steps: obtaining real-time attribute data of each preset area in a to-be-patrolled farmland from a pre-constructed digital real-time map of the to-be-patrolled farmland; inputting the real-time attribute data into a trained prediction model, and outputting patrol urgency scores of the preset areas by adopting the trained prediction model; determining a to-be-inspected area set according to each inspection urgency score, and generating a field inspection path covering the to-be-inspected area set according to a preset field inspection strategy by adopting a preset path planning algorithm; the preset field patrol strategy comprises preferentially patrolling a to-be-patrolled area with a higher urgency score; and controlling the field patrol equipment to carry out field patrol operation according to the field patrol path. The method can respond to the sudden event in time, inspect the area where the sudden event occurs, improve the field inspection efficiency, guarantee the field inspection effect, and avoid the expansion of a small-range problem.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent agriculture, and particularly relates to an intelligent field patrolling method, device, equipment, system, medium and product. BACKGROUND

[0002] With the rapid development of precision agriculture and intelligent agriculture, using automatic equipment to patrol farmland has become an important means to improve the efficiency of agricultural operations and ensure the healthy growth of crops.

[0003] At present, the method of using automatic equipment to patrol the field needs to determine the patrol point first, and then plan the path with the shortest total path or the shortest time as the goal. When determining the patrol point, the farmland is either directly fully covered or the problem points in the farmland are determined through artificial reconnaissance or remote sensing images, and then the problem points are determined as the patrol points. However, the farmland environment is in dynamic change, and when a sudden event occurs in a certain area of the farmland, the above two field patrol methods cannot patrol the area where the sudden event occurs in time, which is easy to cause the expansion of small-scale problems. The first kind of full coverage of the farmland only considers the path or time consumption, and the efficiency is low, which cannot patrol the area where the sudden event occurs in time. The second method determines the problem points in the farmland through artificial reconnaissance or remote sensing images, and the sudden event may have spread rapidly within the artificial reconnaissance period or the interval between two shootings, which cannot patrol the area where the sudden event occurs in time.

[0004] In summary, the field patrol method in the prior art has the problem of being unable to respond to sudden events in the farmland environment in time. SUMMARY

[0005] The embodiments of the present application provide an intelligent field patrolling method, device, equipment and computer storage medium, which can respond to sudden events in time, patrol the area where the sudden event occurs, improve the field patrolling efficiency, ensure the field patrolling effect, and avoid the expansion of small-scale problems.

[0006] In a first aspect, the embodiments of the present application provide an intelligent field patrolling method applied to a field patrolling management device, which comprises: acquiring real-time attribute data of each preset area in a to-be-patrolled farmland from a pre-constructed digital real-time map of the to-be-patrolled farmland; the to-be-patrolled farmland in the digital real-time map is divided into a plurality of preset areas, and each preset area has real-time attribute information; the real-time attribute information includes crop variety, crop growth stage and multi-source real-time environmental data; the multi-source real-time environmental data is collected by a sensor network arranged in the preset area;

[0007] inputting each of the real-time attribute data into a trained prediction model, and outputting a patrol urgency score of each preset region by using the trained prediction model; the patrol urgency score is used to represent an abnormal risk degree of a crop growth state; the trained prediction model is obtained by training a preset prediction model by using a plurality of sample data; the sample data includes multi-source environment sample data and patrol urgency sample scores of different crop varieties in different growth environments and different growth stages;

[0008] determining a set of to-be-patrolled regions according to each of the patrol urgency scores, and generating a field patrol path covering the set of to-be-patrolled regions according to a preset field patrol strategy by using a preset path planning algorithm; the preset field patrol strategy includes preferentially patrolling a to-be-patrolled region with a higher patrol urgency score;

[0009] controlling a field patrol device to perform field patrol work according to the field patrol path.

[0010] In some possible implementation manners, the controlling the field patrol device to perform field patrol work according to the field patrol path includes:

[0011] monitoring whether a sudden event occurs in each preset region during the field patrol work; the sudden event includes an outbreak of a pest, a sudden change in weather, and a failure of an irrigation system;

[0012] in response to a sudden event occurring in any preset region, updating the set of to-be-patrolled regions and the field patrol path according to a current position of the field patrol device, and controlling the field patrol device to perform a field patrol task according to the updated field patrol path; the updated set of to-be-patrolled regions includes the preset region in which the sudden event occurs, and the updated field patrol path covers the updated set of to-be-patrolled regions.

[0013] In some possible implementation manners, the monitoring whether a sudden event occurs in each preset region includes:

[0014] updating the patrol urgency score according to a preset time interval;

[0015] in response to a change value of the patrol urgency score of a preset region in adjacent preset time intervals reaching a preset change threshold, determining that a sudden event occurs in the preset region.

[0016] In some possible implementation manners, the updating the set of to-be-patrolled regions and the field patrol path according to the current position of the field patrol device in response to a sudden event occurring in any preset region includes:

[0017] in response to a sudden event occurring in any preset region, determining the preset region in which the sudden event occurs as a sudden event region;

[0018] determining an unpatrolled region in the field patrol path before the update according to the current position of the field patrol device.

[0019] The emergency incident area and the unpatroled area are combined into a set, and duplicates are removed to obtain the updated area to be patrolled;

[0020] Using the current position of the field patrol equipment as the starting point for path planning, the updated field patrol path is generated based on the preset field patrol strategy using a preset path planning algorithm.

[0021] In some possible implementations, before obtaining real-time attribute data of each preset area in the farmland to be inspected from a pre-constructed digital real-time map of the farmland to be inspected, the method further includes:

[0022] Obtain the geographical environment information of the farmland to be inspected; the geographical environment information includes the water network information, slope information, and shape, size and boundary of the farmland to be inspected on a two-dimensional map;

[0023] Based on the preset grid size corresponding to the crop varieties planted in the farmland to be inspected, the farmland to be inspected is divided into multiple initial grids on the two-dimensional map;

[0024] Based on the water network information and the slope information, determine whether each initial grid has water network partitions and slope partitions;

[0025] If it is determined that the initial grid does not have water network partitions or slope partitions, then the area corresponding to the initial grid is determined as the preset area;

[0026] If it is determined that the initial network has either water network partitions or slope partitions, then the areas corresponding to each water network partition and each slope partition are determined as preset areas;

[0027] The crop varieties and growth stages of each preset area are set, and the multi-source real-time environmental information of each preset area is updated in real time through a sensor network set in each preset area to obtain a digital real-time map of the farmland to be inspected.

[0028] In some possible implementations, determining the set of areas to be patrolled based on each of the patrol urgency scores includes:

[0029] Obtain the dynamic threshold of each preset area; the dynamic threshold is determined based on the frequency of inspection and the inspection results of each preset area;

[0030] The preset areas with patrol urgency scores greater than the dynamic threshold are grouped into a set to determine the set of areas to be patrolled.

[0031] Secondly, embodiments of this application provide an intelligent field patrol device, applied to field patrol management equipment, the device comprising:

[0032] The acquisition module is used to acquire real-time attribute data of each preset area in the farmland to be inspected from a pre-constructed digital real-time map of the farmland to be inspected; the farmland to be inspected in the digital real-time map is divided into multiple preset areas, and each preset area has real-time attribute information; the real-time attribute information includes crop variety, crop growth stage and multi-source real-time environmental data; the multi-source real-time environmental data is collected by a sensor network set in the preset areas;

[0033] The prediction module is used to input the real-time attribute data into the trained prediction model and output the inspection urgency score for each preset area using the trained prediction model. The inspection urgency score is used to characterize the degree of abnormal risk of crop growth status. The trained prediction model is obtained by training the preset prediction model with multiple sample data. The sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages, as well as inspection urgency sample scores.

[0034] The planning module is used to determine the set of areas to be inspected based on the inspection urgency scores, and to generate an inspection path covering the set of areas to be inspected using a preset path planning algorithm and a preset inspection strategy; the preset inspection strategy includes prioritizing the inspection of the areas to be inspected with higher urgency scores.

[0035] The control module is used to control the field patrol equipment to perform field patrol operations according to the field patrol path.

[0036] Thirdly, embodiments of this application provide a field patrol and management device, which includes:

[0037] A processor and a memory storing computer program instructions; an intelligent field patrol method that implements any of the above when the processor executes the computer program instructions.

[0038] Fourthly, embodiments of this application provide an intelligent field patrol system, the system including a field patrol management device, a field patrol device, and a sensor network; the field patrol management device is communicatively connected to both the field patrol device and the sensor network.

[0039] The field patrol management equipment is used to execute any of the above-mentioned intelligent field patrol methods;

[0040] The sensor network is set in a preset area of ​​the farmland to be inspected, and is used to collect multi-source real-time environmental data of the preset area;

[0041] Field patrol equipment is used to conduct field patrol operations according to the stated field patrol route.

[0042] Fifthly, embodiments of this application provide a computer storage medium storing computer program instructions, which, when executed by a processor, implement the intelligent field patrol method described above.

[0043] Sixthly, embodiments of this application provide a computer program product, wherein the instructions in the computer program product, when executed by the processor of an electronic device, enable the electronic device to perform any of the above-mentioned intelligent field patrol methods.

[0044] The intelligent field patrol method, apparatus, device, computer storage medium, and computer program product of this application embodiment acquire real-time attribute data of each preset area in the farmland to be patrolled from a pre-constructed digital real-time map of the farmland to be patrolled. The digital real-time map divides the farmland to be patrolled into multiple preset areas, and each preset area has real-time attribute information. The real-time attribute information includes crop variety, crop growth stage, and multi-source real-time environmental data. The multi-source real-time environmental data is collected by a sensor network set in the preset areas. The real-time attribute data is input into a trained prediction model, and the trained prediction model outputs a patrol urgency score for each preset area. The patrol urgency score is used to characterize the degree of abnormal risk in crop growth status. The trained prediction model is obtained by training the preset prediction model with multiple sample data. The sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages, and patrol urgency sample scores. A set of areas to be patrolled is determined based on each patrol urgency score, and a preset path planning algorithm is used to perform the patrol according to the preset field patrol plan. The strategy generates patrol routes covering the set of areas to be patrolled. The preset patrol strategy includes prioritizing patrols of areas with higher urgency scores. Since environmental data in the digital real-time map is up-to-date and directly linked to the area, obtaining real-time attribute data of each preset area in the farmland to be patrolled from the digital real-time map can efficiently and quickly obtain the latest environmental data. Inputting the real-time attribute data into a trained prediction model and using the trained prediction model to output the patrol urgency score of each preset area can determine the risk level of crop growth in each preset area, promptly detect emergencies in the farmland to be patrolled, and then determine the set of areas to be patrolled based on the patrol urgency scores. A preset path planning algorithm is used to generate patrol routes covering the set of areas to be patrolled according to the preset patrol strategy. Among the generated patrol routes, routes with higher urgency scores can be patrolled first, enabling timely response to emergencies, patrolling areas where emergencies occur, improving patrol efficiency, ensuring patrol effectiveness, and preventing small-scale problems from escalating. Attached Figure Description

[0045] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram illustrating an application scenario of the intelligent field patrol method provided in one embodiment of this application;

[0047] Figure 2 This is a flowchart illustrating an intelligent field patrol method provided in another embodiment of this application;

[0048] Figure 3 This is a schematic diagram of the structure of an intelligent field patrol device provided in another embodiment of this application;

[0049] Figure 4 This is a schematic diagram of the structure of a field patrol management device provided in another embodiment of this application. Detailed Implementation

[0050] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0051] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0052] It should be noted that the acquisition, storage, use, and processing of data in this application embodiment all comply with the relevant provisions of national laws and regulations.

[0053] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0054] First, the prior art involved in this application will be introduced.

[0055] Currently, methods for field inspection using automated equipment require first determining inspection points, and then planning routes with the goal of minimizing the total path or time. When determining inspection points, either the entire farmland is directly covered, or problem areas are first identified through manual surveys or remote sensing images, and then these problem areas are designated as inspection points. However, the farmland environment is constantly changing. When a sudden event occurs in a certain area of ​​the farmland, such as an outbreak of pests or diseases, a sudden change in weather, or a failure of the irrigation system, neither of the above two methods can promptly inspect the area where the sudden event occurred, easily leading to the escalation of small-scale problems.

[0056] To address the problems of the prior art, embodiments of this application provide an intelligent field patrol method, apparatus, device, computer storage medium, and computer program product.

[0057] First, the application scenarios of the intelligent field patrol method provided in the embodiments of this application will be introduced.

[0058] Figure 1 This is a schematic diagram illustrating an application scenario of the intelligent field patrol method provided in one embodiment of this application. For example... Figure 1 As shown, in one application scenario of the intelligent field patrol method provided in this application, there are farmland to be patrolled 10, field patrol management equipment 11, sensor network 12, and field patrol equipment 14. The farmland to be patrolled 10 is divided into multiple preset areas 15. Only some of the preset areas and sensor network are shown in the figure.

[0059] Sensor network 12 is set in each preset area 15 of the farmland 10 to be inspected, and is used to collect multi-source real-time environmental data of each preset area and synchronize the multi-source real-time environmental data of each preset area to the digital real-time map. The digital real-time map can be configured in the field inspection management device 11.

[0060] The field patrol management device 11 obtains real-time attribute data of each preset area from the pre-built digital real-time map of the farmland 10 to be patrolled, including crop variety, crop growth stage and multi-source real-time environmental data.

[0061] The field patrol management device 11 inputs real-time attribute data into the trained prediction model and outputs patrol urgency scores for each preset area.

[0062] The field patrol management device 11 determines the set of areas to be patrolled based on the urgency scores of each patrol, and uses a preset path planning algorithm to generate a patrol path covering the set of areas to be patrolled according to a preset field patrol strategy.

[0063] The field patrol management equipment 11 controls the field patrol equipment 14 to carry out field patrol operations according to the field patrol path 16.

[0064] The intelligent field patrol method provided in the embodiments of this application will be introduced first below.

[0065] Figure 2 This is a flowchart illustrating an intelligent field patrol method according to an embodiment of this application. The intelligent field patrol method provided in one embodiment of this application is applied to field patrol management equipment, such as... Figure 1 As shown, the method may include steps 101 to 104.

[0066] Step 101: Obtain real-time attribute data of each preset area in the farmland to be inspected from the pre-constructed digital real-time map of the farmland to be inspected; the farmland to be inspected in the digital real-time map is divided into multiple preset areas, and each preset area has real-time attribute information; the real-time attribute information includes crop variety, crop growth stage and multi-source real-time environmental data; the multi-source real-time environmental data is collected by a sensor network set in the preset areas.

[0067] In some embodiments, the digital real-time map is an electronic map that is dynamically updated based on multi-source real-time environmental data collected by a Geographic Information System (GIS) and sensor networks. In the digital real-time map, the farmland to be inspected is divided into multiple preset areas, and each preset area has real-time attribute information. In the digital real-time map, each preset area can be associated with a unique identifier to index its real-time attribute information.

[0068] In some embodiments, the sensor network is a network of multiple sensors distributed throughout the farmland to be inspected, used to collect multi-source real-time environmental data for each preset area and update it to a digital real-time map. For example, the sensor network can collect multi-source real-time environmental data every 5 minutes and update it to the digital real-time map.

[0069] In some embodiments, multi-source real-time environmental data may include real-time soil data and real-time meteorological data. Real-time soil data may include soil moisture, pH value, nitrogen, phosphorus, and potassium content, etc.; real-time meteorological data may include temperature, humidity, light intensity, wind speed, etc.

[0070] In some embodiments, sensors can be set at key locations in the farmland to be inspected, such as crop planting points, irrigation points, farmland boundaries, and critical paths, and can cover the farmland to be inspected by point coverage and fence coverage methods.

[0071] In some embodiments, a digital real-time map of the farmland to be inspected can be configured in the farmland inspection management device, which can acquire multi-source real-time environmental data of each preset area in the farmland to be inspected from itself. Alternatively, the digital real-time map of the farmland to be inspected can be configured in a cloud data management device, which is communicatively connected to the farmland inspection management device. The farmland inspection management device can acquire multi-source real-time environmental data of each preset area in the farmland to be inspected from the cloud data management device through communication with it.

[0072] Step 102: Input the real-time attribute data into the trained prediction model and use the trained prediction model to output the inspection urgency score for each preset area; the inspection urgency score is used to characterize the degree of abnormal risk of crop growth status; the trained prediction model is obtained by training the preset prediction model with multiple sample data; the sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages and inspection urgency sample scores.

[0073] In some embodiments, the trained prediction model can be a linear regression model, and the inspection urgency score is a quantitative risk assessment value that represents the possibility of abnormal crop growth, such as pests and diseases, excessive watering, excessive dryness, insufficient sunlight, insufficient fertility, etc. The higher the inspection urgency score, the more priority the area needs to be inspected.

[0074] In some embodiments, before inputting each real-time attribute data into the trained prediction model, the method further includes training a preset prediction model using multiple sample data to obtain the trained prediction model.

[0075] The trained prediction model can be obtained by training a preset prediction model using multiple sample data, as shown below.

[0076] Multiple sample data were acquired, including multi-source environmental sample data of different crop varieties under different growth environments and stages, as well as patrol urgency sample scores. The sample data was obtained by collecting multi-source environmental data of different crop varieties under different growth environments and stages, and then performing data cleaning and normalization on the collected multi-source environmental data. Data cleaning included removing missing values, outliers, and duplicate data. Formula X was used. n =(XX) min ) / (X max -X min Perform data normalization processing, where X is the original data, X... min X is the minimum value of this type of data in the dataset. max X is the maximum value of this type of data in the dataset. n It is the normalized data.

[0077] Multiple sample data are input into a preset prediction model, and the preset prediction model outputs a patrol urgency prediction score for each sample data. The preset prediction model can be a linear regression model. Outputting the patrol urgency prediction score for each sample data using the preset prediction model can include: using preset regression coefficients corresponding to various types of environmental data to perform weighted summation on various types of environmental data in each sample data to obtain a weighted sum value for each sample data; and using preset error correction values ​​to correct each weighted sum value to obtain the patrol urgency prediction score for each sample data.

[0078] Based on the actual patrol urgency score and the predicted patrol urgency score corresponding to each sample data, calculate the loss value of the preset prediction model.

[0079] Adjust the parameters of the preset prediction model based on the loss value of the preset prediction model.

[0080] Repeat the steps from inputting multiple sample data into the preset prediction model to adjusting the parameters of the preset prediction model according to the loss value of the preset prediction model, until the loss value of the preset prediction model is less than the preset loss threshold, or until the number of repetitions reaches the preset number of repetitions, to obtain the trained prediction model.

[0081] Step 103: Determine the set of areas to be inspected based on the urgency scores of each inspection, and use a preset path planning algorithm to generate an inspection path covering the set of areas to be inspected according to a preset inspection strategy; the preset inspection strategy includes prioritizing the inspection of areas with higher urgency scores.

[0082] In some embodiments, a set of preset areas with urgency scores greater than or equal to a preset score threshold can be defined as the set of areas to be patrolled. That is, only the preset areas with urgency scores greater than or equal to the preset score threshold are patrolled, while preset areas with urgency scores less than the preset score threshold may not be patrolled.

[0083] In some embodiments, a preset path planning algorithm is used to calculate the optimal movement path of the field patrol equipment, which may be Dijkstra's algorithm, A* algorithm, genetic algorithm, etc. The preset field patrol strategy may be a shortest path strategy, a minimum energy consumption strategy, a minimum time strategy, etc. The shortest path strategy may prioritize patrol areas with higher urgency scores on the path as much as possible, and minimize the total path length as much as possible. The minimum energy consumption strategy may prioritize patrol areas with higher urgency scores on the path as much as possible, and minimize the energy consumption of the patrol equipment performing field patrol operations according to the patrol path as much as possible. The minimum time strategy may prioritize patrol areas with higher urgency scores on the path as much as possible, and minimize the time taken for the patrol equipment to perform field patrol operations according to the patrol path as much as possible.

[0084] Step 104: Control the field patrol equipment to perform field patrol operations according to the field patrol path.

[0085] In some embodiments, the field patrol management device can send a patrol path to the field patrol equipment and control the patrol equipment to perform patrol operations according to the patrol path. The field patrol equipment can move automatically or semi-automatically in the farmland to be patrolled and perform patrol tasks. The field patrol equipment can be, for example, agricultural drones, field patrol robots, etc.

[0086] The intelligent field patrol method provided in this application obtains real-time attribute data of each preset area in the farmland to be patrolled from a pre-constructed digital real-time map of the farmland to be patrolled. The farmland to be patrolled in the digital real-time map is divided into multiple preset areas, and each preset area has real-time attribute information. The real-time attribute information includes crop variety, crop growth stage, and multi-source real-time environmental data. The multi-source real-time environmental data is collected by a sensor network set in the preset areas. The real-time attribute data is input into a trained prediction model, and the trained prediction model outputs a patrol urgency score for each preset area. The patrol urgency score is used to characterize the degree of abnormal risk of crop growth status. The trained prediction model is obtained by training the preset prediction model with multiple sample data. The sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages, and patrol urgency sample scores. The set of areas to be patrolled is determined according to each patrol urgency score, and a preset path planning algorithm is used to generate a set covering the set of areas to be patrolled according to a preset field patrol strategy. The system provides a pre-defined patrol route and a pre-set patrol strategy that prioritizes patrolling areas with higher urgency scores. Since environmental data in the digital real-time map is up-to-date and directly linked to regions, obtaining real-time attribute data for each pre-defined area in the farmland to be patrolled from the digital real-time map allows for efficient and rapid acquisition of the latest environmental data. Inputting this real-time attribute data into a trained prediction model and using the model's output patrol urgency score for each pre-defined area determines the risk level of crop growth in each area, enabling timely detection of emergencies. Furthermore, based on the patrol urgency scores, a set of areas to be patrolled is determined, and a pre-defined path planning algorithm generates patrol routes covering this set according to the pre-defined patrol strategy. Paths with higher urgency scores are prioritized for patrol, enabling timely response to sudden events and patrolling areas where such events occur, thus improving patrol efficiency, ensuring patrol effectiveness, and preventing small-scale problems from escalating.

[0087] In some embodiments, in order to conduct more timely inspections of areas where sudden events occur and further improve the response speed to emergencies, step 104, "controlling the field inspection equipment to perform field inspection operations according to the field inspection path," includes steps 201 to 202.

[0088] Step 201: During the field inspection, monitor whether any sudden events occur in each preset area; sudden events include outbreaks of pests and diseases, sudden changes in weather, and irrigation system failures.

[0089] In some embodiments, during field inspection operations, the field inspection management device can acquire real-time attribute data of each preset area in the field to be inspected from a pre-built digital real-time map of the field to be inspected at preset time intervals, and monitor whether any sudden changes occur in each preset area based on the changes in the real-time attribute data of the preset areas. For example, if the humidity of a preset area changes significantly between two consecutive outputs, such as the humidity being 30% in the previous acquisition and 80% in the current acquisition, an increase of 50%, it can be considered that a sudden event has occurred in the preset area, which may be rainfall.

[0090] Step 202: In response to a sudden event occurring in any preset area, update the set of areas to be inspected and the inspection path according to the current location of the inspection equipment, and control the inspection equipment to perform the inspection task according to the updated inspection path; the updated set of areas to be inspected includes the preset area where the sudden event occurred, and the updated inspection path covers the updated set of areas to be inspected.

[0091] In some embodiments, if a sudden event is detected in a preset area, in order to patrol the area where the sudden event occurred, the patrol management device can update the set of areas to be patrolled and the patrol path according to the current location of the patrol device. The updated patrol path can be generated by using a preset path planning algorithm based on the current location of the patrol device to cover the updated areas to be patrolled, according to a preset patrol strategy.

[0092] The intelligent field patrol method provided in this embodiment monitors whether a sudden event occurs in each preset area, and in response to a sudden event in any preset area, updates the set of areas to be patrolled and the patrol path according to the current location of the patrol equipment, and controls the patrol equipment to perform the patrol task according to the updated patrol path. This allows for a more timely and faster response to the occurrence of sudden events, patrolling the areas where sudden events have occurred, and handling the sudden events.

[0093] In some embodiments, in order to more accurately monitor whether a sudden event occurs in a preset area, step 201, "monitoring whether a sudden event occurs in each preset area", includes steps 301 to 302.

[0094] Step 301: Update the patrol urgency score according to the preset time interval.

[0095] In some embodiments, the field patrol management device can obtain real-time attribute data of each preset area in the field to be patrolled from a pre-built digital real-time map of the field to be patrolled at preset time intervals, input each real-time attribute data into a trained prediction model, and use the trained prediction model to output the patrol urgency score of each preset area, thereby updating the patrol urgency score of each preset area.

[0096] Step 302: In response to the change value of the patrol urgency score of the preset area in adjacent preset time intervals reaching the preset change threshold, it is determined that a sudden event has occurred in the preset area.

[0097] In some embodiments, if the patrol urgency score of a preset area changes significantly between two consecutive outputs, for example, if the increase in the patrol urgency score reaches a preset increase threshold, then the preset area is determined to be the site of a sudden event.

[0098] The intelligent field patrol method provided in this embodiment determines whether a sudden event has occurred in a preset area by checking the change value of the patrol urgency score. This avoids false alarms of sudden events caused by changes in a single indicator, such as when clouds block the sun and reduce sunlight, but the temperature and humidity do not change much. Therefore, it can more accurately monitor whether a sudden event has occurred in each preset area.

[0099] In some embodiments, in order to patrol the area where a sudden event occurs more promptly and further improve the response speed to the sudden event, step 202, "in response to a sudden event in any preset area, update the set of areas to be patrolled and the patrol path according to the current location of the patrol equipment", includes steps 401 to 403.

[0100] Step 401: In response to a sudden event occurring in any preset area, the preset area where the sudden event occurred is determined as the emergency event area.

[0101] Step 402: Determine the unpatched areas in the patrol path before the update based on the current location of the patrol equipment.

[0102] In some embodiments, the patrol route includes identifiers for the areas to be patrolled and a patrol order, where the patrol order refers to the sequential order in which the areas to be patrolled appear on the patrol route. During the patrol operation, the patrol equipment patrols each area in the patrol order. The patrol equipment can also maintain communication with the patrol management equipment and send its real-time location to the management equipment. Based on the continuously received real-time location data, the patrol management equipment can determine the movement trajectory of the patrol equipment, and thus determine which areas on the patrol route have been patrolled and which have not. The patrol management equipment can then identify areas on the previous patrol route that were not yet patrolled as unpatrolled areas.

[0103] In some embodiments, if a sudden event occurs in an area during the process of the field patrol device performing the field patrol task, the area where the sudden event occurs may be an area that the field patrol device has not inspected yet, or an area that the field patrol device has already inspected. Therefore, in order to respond to sudden events in a timely manner, the field patrol management device can form a set of the sudden event area and the uninspected area, and remove duplicates to obtain the updated area to be inspected, so as to update the field patrol path, so that the updated field patrol path can cover the area where the sudden event occurs.

[0104] Step 403: Starting from the current position of the field patrol device, use a preset path planning algorithm to generate an updated field patrol path according to the preset field patrol strategy.

[0105] In some embodiments, a sudden event occurs during the process of the field patrol device performing the field patrol operation. The position of the field patrol device may not be at the starting position of the field patrol operation. Therefore, when generating the updated field patrol path, it is necessary to use the current position of the field patrol device as the starting point of path planning and the ending position of the field patrol operation as the ending point of path planning for field patrol path planning. Among them, the ending point of the field patrol operation can be an agricultural monitoring station.

[0106] In some embodiments, the preset path planning algorithm can be a genetic algorithm. The method of generating the field patrol path is described below by way of an example.

[0107] Randomly generate multiple initial paths. The starting point of the initial path is the starting point of path planning, the ending point is the ending point of path planning, and each area to be inspected in the set of areas to be inspected is covered.

[0108] Calculate the fitness value of each initial path according to the preset fitness function corresponding to the preset field patrol strategy. Exemplarily, when the preset field patrol strategy is the shortest path strategy, the fitness function F can be expressed as F = Σ(w i *P i ) - β * L, where P i is the inspection urgency score of the i-th area passed by the field patrol path, w i is the weight coefficient corresponding to the i-th position in the field patrol path, and it satisfies that when i < j, w i > w j , that is, the weight of the area inspected first is greater than that of the area visited later. Σ(w i *P i ) is the total inspection benefit after weighting, L is the total length of the field patrol path, and β is the weight coefficient for balancing the inspection benefit and the path cost.

[0109] Select at least one preferred path from the multiple initial paths in descending order of the fitness value.

[0110] Crossover and mutation operations are performed on the preferred path to obtain multiple offspring paths.

[0111] The initial path is updated using child paths. That is, multiple child paths are used as new initial paths, and the steps of calculating the fitness of each initial path according to the preset fitness function and updating the initial path using child paths are repeated until the preset number of iterations or the fitness value converges. The initial path with the highest fitness value is determined as the target path, and the target path is determined as the field patrol path.

[0112] In some embodiments, constraints can be added during the path planning process using a genetic algorithm. These constraints can be the driving range constraints of the field patrol equipment to ensure that the planned path is feasible.

[0113] The intelligent field patrol method provided in this embodiment ensures that areas where sudden events occur can be included in the field patrol path by first updating the area to be patrolled and then updating the field patrol path. Furthermore, it can update the field patrol path in a timely manner according to the sudden event during the field patrol operation, ensuring timely response to the sudden event.

[0114] In some embodiments, in order to obtain more accurate multi-source real-time environmental data of the farmland to be inspected, steps 501 to 506 are included before step 101 "obtaining real-time attribute data of each preset area in the farmland to be inspected from the pre-built digital real-time map of the farmland to be inspected".

[0115] Step 501: Obtain the geographical environment information of the farmland to be inspected; the geographical environment information includes the water network information, slope information, and shape, size and boundary of the farmland to be inspected on a two-dimensional map.

[0116] In some embodiments, geographic environmental information of the farmland to be inspected can be obtained from a geographic information system database, or orthophotos can be taken using an optical camera mounted on a drone, and the generated images can be identified to obtain the geographic environmental information of the farmland to be inspected. Geographic environmental information of the farmland to be inspected can also be obtained from remote sensing satellite images and field mapping data.

[0117] In some embodiments, the size of the farmland to be inspected on a two-dimensional map refers to the area of ​​the farmland. Water network information refers to the distribution, form, and network relationship of all water bodies within and around the farmland. This includes, but is not limited to, irrigation canals, drainage ditches, rivers, streams, ponds, and wellheads. Slope information refers to the degree and direction of the farmland's surface elevation, which may include the steepness of the slope and the direction the slope faces, such as a shady slope or a sunny slope.

[0118] Step 502: Based on the preset grid size corresponding to the crop varieties planted in the farmland to be inspected, divide the farmland to be inspected into multiple initial grids on the two-dimensional map.

[0119] In some embodiments, the crop varieties planted in the farmland to be inspected refer to the varieties of crops mainly planted in the farmland. Since different crops may have different management requirements, the area of ​​each preset zone in the farmland to be inspected may differ when the crop varieties planted are different. For example, rice needs to be managed according to a 5*5 meter grid size, and corn needs to be managed according to a 15*15 meter grid size. Therefore, the preset grid size corresponding to the crop varieties planted in the farmland to be inspected can be determined based on the preset mapping relationship between crop varieties and grid sizes.

[0120] In some embodiments, a spatial gridding tool can be used to divide the farmland to be inspected into multiple initial grids on a two-dimensional map. The initial grids are typically squares or rectangles.

[0121] Step 503: Determine whether each initial grid has water network partitions and slope partitions based on water network information and slope information.

[0122] Step 504: If it is determined that the initial grid does not have water network partitions or slope partitions, then the area corresponding to the initial grid is determined as the preset area.

[0123] Step 505: If it is determined that the initial network has either water network partitions or slope partitions, then the areas corresponding to each water network partition and each slope partition are determined as preset areas.

[0124] In some embodiments, water network zoning refers to plots of land separated by water networks within the initial grid. For example, a river may divide the land corresponding to the initial grid into two plots. Slope zoning refers to plots of land within the initial grid that have significant differences in slope or orientation. For example, if the land corresponding to the initial grid is a small hillside, then the initial grid will contain both shady and sunny slopes. If water network zoning exists within the initial grid, and a river or wide canal divides the land corresponding to the initial grid into two plots, the patrol equipment may be unable to cross the river or wide canal. If slope zoning exists within the initial grid, such as low-lying areas and steep slopes, sunny slopes and shady slopes, the environmental differences are significant because low-lying areas are prone to waterlogging, steep slopes are prone to drought, and sunny slopes receive more sunlight than shady slopes. Therefore, in order to more comprehensively monitor the environmental conditions and crop growth status of the farmland to be patrolled, and for more reasonable path planning, it is necessary to define each water network zoning and slope zoning as a preset area.

[0125] In some embodiments, the initial grid map can be overlaid with the water network map to determine whether a water network line passes through each initial grid. The initial grid map can be overlaid with the slope zoning map to determine whether each initial grid is completely located in the same slope range, such as all within 0-5°, or is divided by different slope ranges, such as one part within 5-10° and another part within 10-15°.

[0126] In some embodiments, if an initial grid has no watercourse passing through it and contains only one slope level, the area corresponding to the initial grid can be directly defined as a preset area. If an initial grid is traversed by a waterway or contains two different slopes, the initial grid is divided into multiple new, smaller preset areas based on the watercourse or the boundary line between the different slopes.

[0127] Precise division of preset areas provides an ideal operational unit for subsequent precision fertilization, variable irrigation, and farmland inspection, avoiding situations where different treatments are needed within the same management unit.

[0128] Step 506: Set the crop varieties and crop growth stages for each preset area, and update the multi-source real-time environmental information of each preset area in real time through the sensor network set in each preset area to obtain a digital real-time map of the farmland to be inspected.

[0129] In some embodiments, after the digital real-time map is constructed, real-time attribute information is defined for each preset area. This real-time attribute information may include crop type, crop growth cycle, and multi-source real-time environmental information. This allows for the direct acquisition of attribute data for each preset area in the farmland to be inspected from the digital real-time map when planning subsequent field inspection routes, thereby improving inspection efficiency.

[0130] The intelligent field patrol method provided in this embodiment can divide the pre-set areas with consistent conditions and slopes, ensuring that the multi-source real-time environmental information acquired by the sensor network set in each pre-set area can accurately reflect the actual environmental conditions in each pre-set area. This allows the multi-source real-time environmental data of each pre-set area in the farmland to be patrolled to accurately reflect the actual environmental conditions in each pre-set area when planning the patrol route in the future. This ensures the timely detection of pre-set areas where sudden events have occurred, enabling precise management and efficient operation of farmland, improving crop yield and quality, and reducing agricultural production costs and environmental impact.

[0131] In some embodiments, in order to respond to sudden events more promptly, step 103, “determining the set of areas to be patrolled based on each patrol urgency score”, includes steps 601 to 602.

[0132] Step 601: Obtain the dynamic threshold of each preset area; the dynamic threshold is determined based on the patrol frequency and patrol results of each preset area.

[0133] In some embodiments, the dynamic threshold of each preset area may have the same initial value. The dynamic threshold may decrease as the patrol frequency increases. When the patrol result of the preset area is normal, the dynamic threshold may increase as the patrol result is normal and decrease as the patrol result is abnormal.

[0134] If an area is added to the inspection set, it indicates that the crop growth status in that area is at high risk or that a sudden event has occurred in the pre-defined area. Therefore, it is necessary to pay close attention to that area and lower its dynamic threshold. This will ensure that the area can be added to the inspection set when planning the next field inspection route, ensuring a rapid response to areas of focus and preventing small problems from escalating into big problems.

[0135] If an area is found to be normal after inspection, its dynamic threshold is increased. In the next assessment, this area will require a higher urgency score to be added to the inspection list, thus avoiding over-inspection of low-risk areas and saving inspection resources.

[0136] If problems such as pests and diseases are found after inspecting an area, then that area needs to be given special attention. Lowering the dynamic threshold of that area can ensure that it can be added to the inspection set when planning the next field inspection route, so as to ensure a rapid response to key areas and prevent small problems from escalating into big problems.

[0137] Step 602: Form a set of preset areas in each preset area whose patrol urgency score is greater than the dynamic threshold, and determine the set of areas to be patrolled.

[0138] In some embodiments, if the patrol urgency score of a preset area is greater than its corresponding dynamic threshold, the preset area is determined as a patrol area, and the patrol urgency score of each preset area is determined sequentially to be greater than its corresponding dynamic threshold to obtain a set of patrol areas.

[0139] The intelligent field patrol method provided in this embodiment obtains the dynamic threshold of each preset area, and forms a set of preset areas whose patrol urgency scores are greater than the dynamic threshold, thereby determining the set of areas to be patrolled. This method can maintain high vigilance over high-risk areas, ensuring that problems and emergencies can be dealt with in a timely manner, and can also improve field patrol efficiency and avoid patrolling areas without problems.

[0140] Based on the intelligent field patrol method provided in the above embodiments, this application also provides specific implementation methods of the intelligent field patrol device. Please refer to the following embodiments.

[0141] Figure 3 This is a schematic diagram of the structure of an intelligent field patrol device provided in another embodiment of this application. See also... Figure 3 An embodiment of this application provides an intelligent field patrol device 30, which includes: an acquisition module 31, a prediction module 32, a planning module 33, and a control module 34.

[0142] The acquisition module 31 is used to acquire real-time attribute data of each preset area in the farmland to be inspected from a pre-constructed digital real-time map of the farmland to be inspected; the farmland to be inspected in the digital real-time map is divided into multiple preset areas, and each preset area has real-time attribute information; the real-time attribute information includes crop variety, crop growth stage and multi-source real-time environmental data; the multi-source real-time environmental data is collected by a sensor network set in the preset areas.

[0143] The prediction module 32 is used to input the real-time attribute data into the trained prediction model and output the inspection urgency score for each preset area using the trained prediction model; the inspection urgency score is used to characterize the degree of abnormal risk of crop growth status; the trained prediction model is obtained by training the preset prediction model with multiple sample data; the sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages and inspection urgency sample scores.

[0144] The planning module 33 is used to determine the set of areas to be inspected based on the inspection urgency scores, and to generate an inspection path covering the set of areas to be inspected using a preset path planning algorithm and a preset inspection strategy; the preset inspection strategy includes prioritizing the inspection of the areas to be inspected with higher urgency scores.

[0145] The control module 34 is used to control the field patrol equipment to perform field patrol operations according to the field patrol path.

[0146] In some possible implementations, the control module can also be used for:

[0147] During field patrol operations, monitor whether any sudden events occur in each preset area; such sudden events include outbreaks of pests and diseases, sudden changes in weather, and irrigation system malfunctions.

[0148] In response to a sudden event occurring in any preset area, the set of areas to be inspected and the inspection path are updated according to the current location of the inspection equipment, and the inspection equipment is controlled to perform the inspection task according to the updated inspection path; the updated set of areas to be inspected includes the preset area where the sudden event occurred, and the updated inspection path covers the updated set of areas to be inspected.

[0149] In some possible implementations, the control module can also be used for:

[0150] The patrol urgency score is updated at preset time intervals;

[0151] When the patrol urgency score of a preset area changes by a preset threshold within an adjacent preset time interval, a sudden event is determined to have occurred in the preset area.

[0152] In some possible implementations, the control module can also be used for:

[0153] In response to a sudden event occurring in any preset area, the preset area where the sudden event occurred is designated as the emergency event area;

[0154] Determine the unpatched areas in the original patrol route based on the current location of the patrol equipment;

[0155] The emergency incident area and the unpatroled area are combined into a set, and duplicates are removed to obtain the updated area to be patrolled;

[0156] Using the current position of the field patrol equipment as the starting point for path planning, the updated field patrol path is generated based on the preset field patrol strategy using a preset path planning algorithm.

[0157] In some possible implementations, the intelligent field inspection device also includes a building module, which is used for:

[0158] Obtain the geographical environment information of the farmland to be inspected; the geographical environment information includes the water network information, slope information, and shape, size and boundary of the farmland to be inspected on a two-dimensional map;

[0159] Based on the preset grid size corresponding to the crop varieties planted in the farmland to be inspected, the farmland to be inspected is divided into multiple initial grids on the two-dimensional map;

[0160] Based on the water network information and the slope information, determine whether each initial grid has water network partitions and slope partitions;

[0161] If it is determined that the initial grid does not have water network partitions or slope partitions, then the area corresponding to the initial grid is determined as the preset area;

[0162] If it is determined that the initial network has either water network partitions or slope partitions, then the areas corresponding to each water network partition and each slope partition are determined as preset areas;

[0163] The crop varieties and growth stages of each preset area are set, and the multi-source real-time environmental information of each preset area is updated in real time through a sensor network set in each preset area to obtain a digital real-time map of the farmland to be inspected.

[0164] In some possible implementations, the planning module is specifically used for:

[0165] The dynamic threshold of each preset area is obtained; the dynamic threshold is determined based on the patrol frequency and patrol results of each preset area.

[0166] The preset areas with patrol urgency scores greater than the dynamic threshold are grouped into a set to determine the set of areas to be patrolled.

[0167] The various modules of the intelligent field patrol device provided in this application embodiment can achieve Figure 2 It provides the functions for each step of the intelligent field inspection method and can achieve its corresponding technical effects. For the sake of brevity, it will not be described in detail here.

[0168] Figure 4 This is a schematic diagram of the structure of a field patrol management device provided in another embodiment of this application. See also... Figure 4 In order to implement the intelligent field patrol method in the above embodiments, this application embodiment can provide a field patrol management device 40, which includes: a processor 41 and a memory 42 storing computer program instructions; when the processor 41 executes the computer program instructions, it implements any one of the intelligent field patrol methods in the above embodiments.

[0169] The intelligent field patrol method described in the above embodiments can be implemented using an intelligent field patrol system provided in this application. This system includes a field patrol management device, a field patrol device, and a sensor network; the field patrol management device is communicatively connected to both the field patrol device and the sensor network.

[0170] The field patrol management equipment is used to execute any one of the intelligent field patrol methods described in the above embodiments;

[0171] The sensor network is set in a preset area of ​​the farmland to be inspected, and is used to collect multi-source real-time environmental data of the preset area;

[0172] Field patrol equipment is used to conduct field patrol operations according to the stated field patrol route.

[0173] The intelligent field patrol method in the above embodiments can be implemented using a computer storage medium. This computer storage medium stores computer program instructions; when these instructions are executed by a processor, they implement any of the intelligent field patrol methods described in the above embodiments.

[0174] This application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the intelligent field patrol methods described in the above embodiments.

[0175] It should be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0176] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0177] The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A smart field inspection method, characterized in that, The method, applied to field patrol and management equipment, includes: Real-time attribute data of each preset area in the farmland to be inspected is obtained from a pre-constructed digital real-time map of the farmland to be inspected. The farmland to be inspected in the digital real-time map is divided into multiple preset areas, and each preset area has real-time attribute information. The real-time attribute information includes crop variety, crop growth stage and multi-source real-time environmental data. The multi-source real-time environmental data is collected by a sensor network set in the preset areas. The real-time attribute data are input into the trained prediction model, and the trained prediction model is used to output the inspection urgency score for each preset area. The inspection urgency score is used to characterize the degree of abnormal risk of crop growth status. The trained prediction model is obtained by training the preset prediction model with multiple sample data. The sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages, as well as inspection urgency sample scores. The set of areas to be inspected is determined based on the urgency scores of each inspection, and a pre-defined path planning algorithm is used to generate an inspection path covering the set of areas to be inspected according to a pre-defined inspection strategy; the pre-defined inspection strategy includes prioritizing the inspection of the areas to be inspected with higher urgency scores. Control the field patrol equipment to perform field patrol operations according to the stated field patrol path.

2. The intelligent field inspection method according to claim 1, characterized in that, The controlled field patrol equipment performs field patrol operations according to the patrol path, including: During field patrol operations, monitor whether any sudden events occur in each preset area; such sudden events include outbreaks of pests and diseases, sudden changes in weather, and irrigation system malfunctions. In response to a sudden event occurring in any preset area, the set of areas to be inspected and the inspection path are updated according to the current location of the inspection equipment, and the inspection equipment is controlled to perform the inspection task according to the updated inspection path; the updated set of areas to be inspected includes the preset area where the sudden event occurred, and the updated inspection path covers the updated set of areas to be inspected.

3. The intelligent field inspection method according to claim 2, characterized in that, The monitoring of whether a sudden event occurs in each preset area includes: The patrol urgency score is updated at preset time intervals; When the patrol urgency score of a preset area changes by a preset threshold within an adjacent preset time interval, a sudden event is determined to have occurred in the preset area.

4. The intelligent field inspection method according to claim 2, characterized in that, The response to a sudden event occurring in any preset area, updating the set of areas to be inspected and the inspection path based on the current location of the inspection equipment, includes: In response to a sudden event occurring in any preset area, the preset area where the sudden event occurred is designated as the emergency event area; Determine the unpatched areas in the original patrol route based on the current location of the patrol equipment; The emergency incident area and the unpatroled area are combined into a set, and duplicates are removed to obtain the updated area to be patrolled; Using the current position of the field patrol equipment as the starting point for path planning, an updated field patrol path is generated based on a preset path planning algorithm and a preset field patrol strategy.

5. The intelligent field inspection method according to claim 4, characterized in that, Before obtaining real-time attribute data of each preset area in the farmland to be inspected from the pre-constructed digital real-time map of the farmland to be inspected, the process also includes: Obtain the geographical environment information of the farmland to be inspected; the geographical environment information includes the water network information, slope information, and shape, size and boundary of the farmland to be inspected on a two-dimensional map; Based on the preset grid size corresponding to the crop varieties planted in the farmland to be inspected, the farmland to be inspected is divided into multiple initial grids on the two-dimensional map; Based on the water network information and the slope information, determine whether each initial grid has water network partitions and slope partitions; If it is determined that the initial grid does not have water network partitions or slope partitions, then the area corresponding to the initial grid is determined as the preset area; If it is determined that the initial network has either water network partitions or slope partitions, then the areas corresponding to each water network partition and each slope partition are determined as preset areas; The crop varieties and growth stages of each preset area are set, and the multi-source real-time environmental information of each preset area is updated in real time through a sensor network set in each preset area to obtain a digital real-time map of the farmland to be inspected.

6. The intelligent field inspection method according to claim 1, characterized in that, The process of determining the set of areas to be patrolled based on the patrol urgency scores includes: Obtain the dynamic threshold of each preset area; the dynamic threshold is determined based on the patrol frequency and patrol results of each preset area; The preset areas with patrol urgency scores greater than the dynamic threshold are grouped into a set to determine the set of areas to be patrolled.

7. An intelligent field patrol device, characterized in that, Applied to field patrol and management equipment, the device includes: The acquisition module is used to acquire real-time attribute data of each preset area in the farmland to be inspected from a pre-constructed digital real-time map of the farmland to be inspected; the farmland to be inspected in the digital real-time map is divided into multiple preset areas, and each preset area has real-time attribute information; the real-time attribute information includes crop variety, crop growth stage and multi-source real-time environmental data; the multi-source real-time environmental data is collected by a sensor network set in the preset areas; The prediction module is used to input the real-time attribute data into the trained prediction model and output the inspection urgency score for each preset area using the trained prediction model. The inspection urgency score is used to characterize the degree of abnormal risk of crop growth status. The trained prediction model is obtained by training the preset prediction model with multiple sample data. The sample data includes multi-source environmental sample data of different crop varieties in different growth environments and different growth stages, as well as inspection urgency sample scores. The planning module is used to determine the set of areas to be inspected based on the inspection urgency scores, and to generate an inspection path covering the set of areas to be inspected using a preset path planning algorithm and a preset inspection strategy; the preset inspection strategy includes prioritizing the inspection of the areas to be inspected with higher urgency scores. The control module is used to control the field patrol equipment to perform field patrol operations according to the field patrol path.

8. A field patrol and management device, characterized in that, The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the intelligent field patrol method as described in any one of claims 1-6.

9. An intelligent field patrol system, characterized in that, The system includes field patrol management equipment, field patrol equipment, and a sensor network; the field patrol management equipment is communicatively connected to both the field patrol equipment and the sensor network. The field patrol management equipment is used to perform the intelligent field patrol method as described in any one of claims 1-6; The sensor network is set in a preset area of ​​the farmland to be inspected, and is used to collect multi-source real-time environmental data of the preset area; Field patrol equipment is used to conduct field patrol operations according to the stated field patrol route.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the intelligent field patrol method as described in any one of claims 1-6.

11. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device is able to perform the intelligent field patrol method as described in any one of claims 1-6.

Images