Intelligent self-adaptive spraying unmanned aerial vehicle system

Abstract

The invention provides an intelligent self-adaptive spraying unmanned aerial vehicle system, and relates to the field of agricultural intelligent spraying, the system comprises a target area verification module, a monitoring area sensitivity determination module, a spray drift prediction module and a boundary point location updating module, the target area verification module determines whether a spraying unmanned aerial vehicle arrives at a preset target area, and the monitoring area sensitivity determination module determines whether the spraying unmanned aerial vehicle arrives at the boundary point location updating module; determining whether the crop information in the target area is matched with the spraying object information or not; the monitoring area sensitivity judgment module identifies crop information in a monitoring area and determines a predicted sensitive value of crops in the monitoring area to a sprayed object; the spray drift prediction module determines a predicted landing area of spray droplets when the spray unmanned aerial vehicle sprays at a preset boundary point in the target area; and a boundary point location updating module determines entity-free isolation zone information according to the predicted landing area and the predicted sensitive value, and updates an actual boundary point location in the target area. The system dynamically adapts to different sprayed objects and weather, and cross-boundary pollution of the sprayed objects is effectively avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agricultural intelligent spraying, in particular to an intelligent adaptive spraying unmanned aerial vehicle system. BACKGROUND

[0002] At present, unmanned aerial vehicle spraying technology has been widely used in many fields due to its high efficiency and flexibility, mainly including the following directions: agricultural plant protection, forestry protection, urban greening and garden management, etc. In the field of agriculture, agricultural unmanned aerial vehicle spraying technology has been widely used in the plant protection of field crops (such as rice and wheat), and its core functions include: path planning: automatic flight based on GPS or RTK positioning. Variable spraying: adjusting the amount of pesticide according to the preset map. Basic obstacle avoidance: avoiding obstacles through infrared or ultrasonic sensors.

[0003] However, the agricultural unmanned aerial vehicle spraying technology has the following key problems: the traditional technology does not establish an effective monitoring area protection mechanism, and the drift of pesticides may cause damage to adjacent sensitive crops (such as mulberry trees and vegetables). The existing solutions mostly rely on fixed-width physical isolation belts or spraying boundaries that are usually preset fixed values, which not only wastes the effective area of farmland, but also cannot adapt to the changes of different pesticide characteristics and weather conditions, which may cause over-protection (waste of pesticides) or insufficient protection (pesticide damage accidents), and is easy to cause economic losses. Therefore, the present application provides an intelligent adaptive spraying unmanned aerial vehicle system based on multi-modal perception and dynamic risk control. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present application, the embodiments of the present application provide an intelligent adaptive spraying unmanned aerial vehicle system, which comprises: A target area verification module is configured to acquire position information of a spraying unmanned aerial vehicle in real time, determine whether the spraying unmanned aerial vehicle reaches a preset target area, and when the spraying unmanned aerial vehicle reaches the preset target area, collect crop images and spectral data in the preset target area based on a camera, identify crop information in the target area, and determine whether the crop information in the target area matches pesticide information. A monitoring area sensitive determination module is configured to, when the crop information in the target area matches the pesticide information successfully, enter a monitoring area through the spraying unmanned aerial vehicle, identify crop information in the monitoring area, and determine a predicted sensitivity value of crops in the monitoring area to the pesticide according to the pesticide information and the crop information in the monitoring area. The spray drift prediction module is configured to drive the spraying unmanned aerial vehicle to perform pre-spraying at a preset center point in the target area, acquire real-time meteorological data and operation parameters of the spraying unmanned aerial vehicle, input the real-time meteorological data and the operation parameters into a preset spray drift prediction model, and output a prediction result, so as to determine a predicted falling area of spray droplets when the spraying unmanned aerial vehicle sprays at a preset boundary point in the target area. The boundary point updating module is configured to determine no-entity isolation belt information according to the predicted falling area and the predicted sensitive value, the no-entity isolation belt information including a position and a size of a no-entity isolation belt, and update actual boundary points in the target area according to the no-entity isolation belt information.

[0005] Preferably, the camera is mounted on the spraying unmanned aerial vehicle, and the camera includes a multispectral camera and a high-resolution RGB camera. The content of determining whether the spraying unmanned aerial vehicle reaches the preset target area includes: When the position of the spraying unmanned aerial vehicle is in the preset target area, and a stay duration of the spraying unmanned aerial vehicle in the preset target area in a preset interval is within a preset duration threshold, it is determined that the spraying unmanned aerial vehicle reaches the preset target area, the preset interval is a time interval with the current time as an end point and a preset duration as an interval length, and the preset target area refers to a preset planting area that needs to be sprayed.

[0006] Preferably, the spraying material information in the target area verification module includes a spraying material type, a concentration, and a list of applicable crops, the list of applicable crops includes a spraying material name, an applicable crop type, and an applicable growth stage, and the spraying material type includes irrigation water, nutrient solution, and pesticide solution.

[0007] Preferably, the content of determining whether the crop information in the target area matches the spraying material information includes: The crop type and the growth stage in the crop information in the target area are taken as a to-be-detected set. Data in the list of applicable crops are read in sequence, the to-be-detected set is traversed according to the read data, and when matching is successful, corresponding elements in the to-be-detected set are deleted. When the data in the to-be-detected set is empty, the traversal is stopped, and it is determined that the crop information in the target area matches the spraying material information successfully. When each data in the list of applicable crops is read and the to-be-detected set is not empty, it is determined that the crop information in the target area does not match the spraying material information, and an exception prompt information is generated, the exception prompt information carrying the crop information in the target area and the spraying material information.

[0008] Preferably, the monitoring area is an area that is extended outward by a preset distance based on the boundary of the target area, and the outward extension by the preset distance is adjusted according to the spraying material information.

[0009] Preferably, the content of determining the predicted sensitivity value of the crop in the monitoring area to the spray according to the spray information and the crop information in the monitoring area comprises: obtaining a basic sensitivity value based on a preset static matching of a crop-drug interaction database, that is, matching the crop type and growth stage in the crop information in the monitoring area and the spray type with the preset crop-drug interaction database, so as to determine the basic sensitivity value, wherein the crop-drug interaction database includes the spray type, the crop type, the growth stage and the corresponding sensitivity value; obtaining a predicted sensitivity value by dynamically correcting the real-time weather data, wherein the real-time weather data includes wind speed, temperature and humidity.

[0010] Preferably, the content of obtaining a predicted sensitivity value by dynamically correcting the real-time weather data comprises: determining a correction coefficient according to the real-time weather data, correcting the basic sensitivity value based on the correction coefficient to obtain a predicted sensitivity value, wherein the correction coefficient includes a temperature correction coefficient and a wind speed correction coefficient, the wind speed in the real-time weather data and the wind speed correction coefficient are in a positive correlation, and the temperature in the real-time weather data and the temperature correction coefficient are in a positive correlation.

[0011] Preferably, the monitoring area sensitivity determination module further comprises a stop spraying determination unit, and the stop spraying determination unit comprises: a sensitivity value comparison subunit, configured to compare the predicted sensitivity value with a preset high-risk sensitivity value; a prohibition determination subunit, configured to generate a prohibition spraying operation prompt information when the predicted sensitivity value is greater than a preset risk sensitivity value, wherein the prohibition spraying operation prompt information carries the predicted sensitivity value; an allowance determination subunit, configured to generate an allowance spraying operation prompt information when the predicted sensitivity value is not greater than the preset risk sensitivity value.

[0012] Preferably, the preset center point refers to the center point of the target area, the preset spray drift prediction model is an AGDISP model, the AGDISP model is based on a Lagrangian particle tracking method and is used for simulating the motion trajectory of the spray, and the prediction result includes a drift distance range and a drift path.

[0013] Preferably, the preset boundary point is a point whose distance between the spraying unmanned aerial vehicle and the boundary of the target area is within a preset distance threshold, and the preset distance threshold is determined according to the type of the spray head.

[0014] Preferably, the boundary point updating module comprises: An influence area prediction unit is configured to compare the predicted landing area and the target area, and determine a predicted influence area outside the overlap area between the predicted landing area and the target area. A buffer zone determination unit is configured to determine non-entity buffer zone information according to the predicted influence area and the predicted sensitivity value, that is, when the predicted sensitivity value is greater than a preset sensitivity threshold, the predicted influence area is correspondingly expanded outward according to the size of the predicted sensitivity value, when the predicted sensitivity value is not greater than the preset sensitivity threshold, the predicted influence area is correspondingly reduced inward according to the direction of the target area, and the non-entity buffer zone information is determined according to the position and size of the adjusted predicted influence area. A boundary point updating unit is configured to update preset boundary points at the boundary in the direction of the target area according to the non-entity buffer zone information, to obtain actual boundary points in the target area.

[0015] Compared with the prior art, the present application has the following beneficial effects: Multi-dimensional crop-drug matching: combined with spectral imaging and deep learning, the crop type and growth stage are identified, and dynamically matched with the database (containing applicable growth stages) to avoid mis-spraying.

[0016] Dynamic monitoring area adjustment and boundary point updating: the monitoring area range is automatically adjusted according to the spraying information, the actual boundary point position is updated by the sensitivity value driving, the position and size of the non-entity buffer zone are dynamically adjusted based on the crop sensitivity value (0-10 levels), and the high-risk area is ensured to be zero-contact. The present application dynamically adapts to different spraying materials or meteorological conditions, effectively avoiding cross-border pollution of the spraying material.

[0017] Pre-spraying drift prediction: before formal operation, the AGDISP model is used to simulate the droplet landing point through pre-spraying, to find out the potential drift risk in advance, and to optimize the actual spraying path and subsequent operation parameters accordingly. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a structural block diagram of an intelligent self-adaptive spraying unmanned aerial vehicle system provided by an embodiment of the present application.

[0019] Figure 2 is a flowchart of an intelligent self-adaptive spraying unmanned aerial vehicle method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the technical problems to be solved by the present application, the technical solutions and beneficial effects more clear and explicit, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0021] Please refer toFigure 1 The embodiment of the present application provides a kind of intelligent adaptive unmanned plane system of spraying, the system 10 includes: Target area verification module 11, for obtaining the position information of spraying unmanned plane in real time, determine whether the spraying unmanned plane reaches the preset target area, when reaching the preset target area, crop image and spectrum data in the preset target area are collected based on camera, the crop information in the target area is identified, and it is determined whether the crop information in the target area matches the spraying material information; In the embodiment of the present application, the position information of the spraying unmanned plane is obtained in real time based on the position acquisition device built in the spraying unmanned plane, which includes GPS and the like;The position information of the spraying unmanned plane is obtained in real time, and it is determined whether the spraying unmanned plane reaches the preset target area: when the position of the spraying unmanned plane is in the preset target area, and the stay duration of the spraying unmanned plane in the preset target area in the preset interval is within the preset duration threshold, it is determined that the spraying unmanned plane reaches the preset target area.The preset target area refers to the preset planting area that needs to be sprayed, and the preset interval is a time interval with the current time as the end point and the preset duration as the interval length; When the spraying unmanned plane reaches the preset target area, crop image and spectrum data in the preset target area are collected based on camera, and the crop information in the target area is identified according to the crop image and spectrum data in the target area, and it is determined whether the crop information in the target area matches the spraying material information, which includes crop type (such as rice, wheat, etc.), growth stage (tillering stage, heading stage), and the spraying material information refers to the information of the spraying material loaded by the spraying unmanned plane;Spraying material information includes spraying material type (irrigation water, nutrient solution, pesticide solution, etc.), concentration, and applicable crop list, which includes spraying material name, applicable crop type and applicable growth stage; Among them, the camera is carried on the spraying unmanned plane, and the camera includes multispectral camera and high-resolution RGB camera, the multispectral camera collects 5-10 waveband spectrum data (such as red edge, near infrared), which is used to distinguish crop type and health status;High-resolution RGB camera collects crop image (resolution ≥20 million pixels), which is used for morphological feature extraction (such as leaf shape, texture); Crop type identification: based on deep learning model output crop type (such as "rice"), the training data of deep learning model: based on public data set (such as PlantVillage, EuroSAT) and self-built field data set (covering rice, wheat, corn and other main crops); Growth stage determination: a multi-scale feature fusion algorithm is used to output growth stage labels (such as "rice-tillering stage") based on spectral data and crop images; wherein, the tillering stage detection: the occurrence of tillering is determined by a sudden increase in near-infrared band reflectance; the heading stage detection: the inflorescence morphology is segmented based on the crop image (U-Net model); the crop type and growth stage recognition method is the prior art, and the present application will not be described here; The specific content of determining whether the crop information and the spraying information in the target area match includes: taking the crop type and the growth stage in the crop information in the target area as a to-be-detected set, reading data from the applicable crop list in sequence, traversing the to-be-detected set according to the read data, deleting the corresponding element in the to-be-detected set when matching succeeds (here, the matching success means that the similarity between the read data and the corresponding element in the to-be-detected set meets a preset similarity threshold), and stopping traversing and determining that the crop information and the spraying information in the target area successfully match when the data in the to-be-detected set is empty; when each data in the applicable crop list is read and the to-be-detected set is not empty, it is determined that the crop information and the spraying information in the target area fail to match, and abnormal prompt information is generated, wherein the abnormal prompt information carries the crop information and the spraying information in the target area; The present application can effectively avoid the unmanned aerial vehicle from starting spraying in a non-target area due to positioning errors or operation errors, thereby causing pesticide waste or crop damage.

[0022] In the embodiment of the present application, the target area verification module 11 can further include a crop health state evaluation unit, which obtains vegetation indexes such as NDVI (normalized difference vegetation index) and PRI (photochemical reflectance index) through a multispectral camera in real time; compares the NDVI value and the PRI value with corresponding threshold values respectively; when NDVI>0.7: healthy vegetation; 0.4<NDVI≤0.7: mild stress; NDVI≤0.4: severe stress; PRI>0.15: normal photosynthesis; 0.05 On this basis, a double-factor evaluation matrix is established: when the NDVI interval is >0.7 and the PRI interval is >0.15, it is determined that the crop health status is healthy; when the NDVI interval is >0.7 and the PRI interval is 0.05-0.15, it is determined that the crop health status is short-term stress; when the NDVI interval is 0.4-0.7 and the PRI interval is >0.15, it is determined that the crop health status is nutrient deficiency; when the NDVI interval is <0.4 and the PRI interval is any value, it is determined that the crop health status is disease / death; the spraying unmanned aerial vehicle can realize the synchronous evaluation of the crops in the target region based on the crop health status evaluation unit during the spraying process; The monitoring area sensitive determination module 12 is used for entering the monitoring area through the spraying unmanned aerial vehicle when the crop information in the target region is successfully matched with the spraying material information, identifying the crop information in the monitoring area, and determining the predicted sensitivity of the crops in the monitoring area to the spraying material according to the spraying material information and the crop information in the monitoring area. In the embodiment of the application, when the crop information in the target region is successfully matched with the spraying material information, the monitoring area is entered through the spraying unmanned aerial vehicle, and the crop information in the monitoring area is identified. In the embodiment, the method for identifying the crop information in the monitoring area is the same as the method for identifying the crop information in the target region, and will not be described again. The monitoring area is a region that is extended by a preset distance from the boundary of the target region. The preset distance is adjusted according to the spraying material information, and is usually at least greater than 3 meters. The preset distance can be determined according to the type of the spraying material. When the spraying material is irrigation water, nutrient solution or pesticide solution, the corresponding preset distance is gradually increased, that is, the toxicity of the spraying material and the preset distance are in a positive correlation. The predicted sensitivity of the crops in the monitoring area to the spraying material is determined according to the spraying material information and the crop information in the monitoring area. The sensitivity score (SS) reflects the potential damage degree of the crops in the monitoring area after contacting the spraying material, and is usually divided into 0-10 levels: 0-3: low risk (such as irrigation water on wheat), 4-6: medium risk (such as urea nutrient solution on flowering rape), and 7-10: high risk (such as herbicide on mulberry). The method for determining the predicted sensitive value is: obtaining a basic sensitive value based on a preset static matching of a crop-drug interaction database, the crop-drug interaction database including information such as spray type, crop type, growth stage and corresponding sensitive value; for example, when the spray type is urea, the crop type is wheat, and the growth stage is the jointing stage, the corresponding sensitive value is 2; when the spray type is glyphosate, the crop type is rice, and the growth stage is the tillering stage, the corresponding sensitive value is 3, and so on, without listing the contents of the crop-drug interaction database; matching the crop type and growth stage in the crop information in the monitoring area and the spray type with the preset crop-drug interaction database to determine the basic sensitive value, and then dynamically correcting through real-time weather data to obtain the predicted sensitive value; The content corrected by the real-time weather data includes: determining a correction coefficient according to the real-time weather data, correcting the basic sensitive value based on the correction coefficient to obtain the predicted sensitive value, the correction coefficient including a temperature correction coefficient and a wind speed correction coefficient, both of which are greater than 0, that is, each correction coefficient is multiplied by the basic sensitive value to obtain the predicted sensitive value, and generally the wind speed in the real-time weather data is positively correlated with the wind speed correction coefficient; the temperature in the real-time weather data is positively correlated with the temperature correction coefficient; for example, when the temperature is 30°C, the corresponding temperature correction coefficient is usually 1.2, and when the temperature is lower than 15°C, the corresponding temperature correction coefficient is usually 0.8; In this embodiment, the monitoring area is expanded to include crops outside the target area, and the monitoring area range is automatically adjusted according to the spray type, generally the higher the toxicity of the spray, the larger the corresponding monitoring area. After determining the monitoring area, the crops in the monitoring area are identified, and the sensitivity of the crops in the monitoring area to the sprayed drug is determined, for example, when the spray type is irrigation water, the toxicity of the irrigation water is low, and the mist droplets drift to other areas, generally having little effect on the crops in other areas; but for high-toxicity sprays, when the mist droplets drift to other areas, they can have a greater impact on the crops in other areas; in addition, the present application considers the correction coefficient such as temperature when determining the sensitive value, when the temperature is high, the drug activity is enhanced, and the sensitive value is increased; when the temperature is low, the drug activity is reduced, and the sensitive value is decreased; the greater the wind speed, the farther the mist droplets drift, and the greater the impact on the crops.

[0023] The monitoring area sensitive determination module of the present application further comprises a stop spraying determination unit, which comprises: A sensitive value comparison subunit for comparing the predicted sensitive value with a preset high-risk sensitive value; A prohibition determination subunit is configured to generate a prohibition of spraying operation prompt information when the predicted risk-sensitive value is greater than a preset risk-sensitive value, the prohibition of spraying operation prompt information carrying the predicted risk-sensitive value; An allowance determination subunit is configured to generate an allowance of spraying operation prompt information when the predicted risk-sensitive value is not greater than a preset risk-sensitive value.

[0024] The stop spraying determination subunit can timely remind the user of the high risk of the crops in the monitoring area during spraying, effectively prevent loss, and effectively avoid loss for the manager.

[0025] The spray drift prediction module 13 is configured to drive the spraying unmanned aerial vehicle to perform pre-spraying at a preset center point in the target area, acquire real-time meteorological data and operation parameters of the spraying unmanned aerial vehicle, input the real-time meteorological data and the operation parameters into a preset spray drift prediction model, and output a prediction result, so as to determine a predicted falling area of the sprayed mist droplets when the spraying unmanned aerial vehicle sprays at a preset boundary point in the target area according to the prediction result. In this embodiment, the spraying unmanned aerial vehicle is driven to perform pre-spraying at a preset center point in the target area, real-time meteorological data and operation parameters of the spraying unmanned aerial vehicle are acquired, and the real-time meteorological data and the operation parameters are input into a preset spray drift prediction model to output a prediction result. The preset center point usually refers to a center point of the target area. The real-time meteorological data includes wind speed, wind direction, temperature and humidity, and the operation parameters include the type of the spraying unmanned aerial vehicle, the flight height, the speed, the spraying flow, the nozzle parameters (the type of the nozzle and the distribution of the mist droplet particle size), and the like. The distribution of the mist droplet particle size can be measured by a wind tunnel experiment or a laser particle size analyzer. The preset spray drift prediction model can be an AGDISP model. The AGDISP model is based on the Lagrangian particle tracking method and is used to simulate the motion trajectory of the spray. This method predicts the drift of the entire spray group by tracking the motion of each mist droplet in the air. The model considers the influence of air dynamics and meteorological data on the drift of the mist droplets. The core equation of the model is the motion equation of the mist droplets, which describes the changes of the acceleration, speed and position of the mist droplets in the air. These equations are based on Newton's second law and consider the effects of gravity, air resistance and other external forces on the mist droplets. The AGDISP model is a prior art and will not be described here. The prediction result includes a drift distance range, a drift path (including a drift direction), and the like. According to the prediction result, the predicted falling area of the sprayed mist droplets when the spraying unmanned aerial vehicle sprays at a preset boundary point in the target area is determined. The preset boundary point is a point whose distance from the boundary of the spraying unmanned aerial vehicle and the target area is within a preset distance threshold. The preset distance threshold is determined according to the type of the nozzle. Generally, the greater the spraying range of the nozzle, the greater the preset distance threshold. The preset boundary point is usually set in each boundary direction of the target area, and there are generally at least four preset boundary points. The boundary point position updating module 14 is configured to determine no-entity isolation belt information according to the predicted falling area and the predicted sensitive value, the no-entity isolation belt information including the position and size of the no-entity isolation belt; and update the actual boundary point positions in the target area according to the no-entity isolation belt information. The boundary point position updating module 14 specifically includes: The influence area prediction unit is configured to compare the predicted falling area with the target area, and determine the predicted falling area outside the overlapping area between the predicted falling area and the target area as a predicted influence area. The isolation belt determination unit is configured to determine no-entity isolation belt information according to the predicted influence area and the predicted sensitive value, that is, when the predicted sensitive value is greater than a preset sensitive threshold, correspondingly expand the predicted influence area outward according to the size of the predicted sensitive value, when the predicted sensitive value is not greater than the preset sensitive threshold, correspondingly reduce the predicted influence area inward according to the direction of the target area, and determine the no-entity isolation belt information according to the position and size of the adjusted predicted influence area. The boundary point updating unit is configured to update the preset boundary point positions at the boundary in the direction of the target area according to the no-entity isolation belt information, to obtain the actual boundary point positions in the target area. The number of the actual boundary point positions in the target area is generally at least four. After the actual boundary point positions are determined, the actual moving spraying area is formed by sequentially connecting the actual boundary point positions according to the boundary direction of the target area. The spraying unmanned aerial vehicle only performs a spraying task in the actual moving spraying area, dynamically adapts to different spraying materials or meteorological conditions, effectively avoids cross-border pollution of the spraying material, and avoids the generation of economic losses.

[0026] Please refer to Figure 2 Another object of the embodiment of the present application is to provide an intelligent self-adaptive spraying unmanned aerial vehicle method, the method including the following steps: Step S100, real-time acquisition of position information of a spraying unmanned aerial vehicle, determination of whether the spraying unmanned aerial vehicle reaches a preset target area, when reaching the preset target area, based on a camera, collection of crop images and spectral data in the preset target area, identification of crop information in the target area, and determination of whether the crop information in the target area matches spraying material information; Step S200, when the crop information in the target area and the spraying material information are successfully matched, the spraying unmanned aerial vehicle enters a monitoring area, identifies crop information in the monitoring area, and determines a predicted sensitive value of crops in the monitoring area to the spraying material according to the spraying material information and the crop information in the monitoring area. In step S300, the pre-driving spraying unmanned aerial vehicle sprays at a preset center point in the target area, and real-time meteorological data and operation parameters of the spraying unmanned aerial vehicle are acquired and input into a preset spray drift prediction model, a prediction result is output, and the predicted falling area of the spraying material droplets is determined when the spraying unmanned aerial vehicle sprays at a preset boundary point in the target area according to the prediction result. In step S400, the predicted falling area and the predicted sensitive value are used to determine no-entity isolation belt information, and the no-entity isolation belt information includes the position and size of the no-entity isolation belt. The actual boundary point in the target area is updated according to the no-entity isolation belt information.

[0027] It should be noted that, in this document, the terms such as first and second are used only 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. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device.

[0028] Although the embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

[0029] Those skilled in the art will also appreciate that the various example logical blocks, modules, circuits, and algorithm steps described in connection with the present application herein can be implemented as electronic hardware, computer software, or a combination of the two.

[0030] The flowcharts and block diagrams in the drawings show the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment, or a portion of code that comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that shown in the figures. For example, two blocks shown in succession can in fact be executed substantially concurrently, or the blocks can sometimes be executed in reverse order, depending on the functionality involved.

[0031] It is also important to note that each block of the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, can be implemented by dedicated hardware-based systems which perform the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0032] Embodiments of the application have been described above with the understanding that these descriptions are exemplary only, and are not intended to be exhaustive or to limit the embodiments disclosed to the precise forms disclosed. Many modifications and variations are possible in light of the above teachings. The embodiments described are meant to be illustrative only and are not intended to limit the scope of the disclosure. It is therefore to be understood that within the scope of the appended claims and their equivalents, modifications and variations of the embodiments described herein can be made by those skilled in the art. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

Claims

1. An intelligent adaptive spraying drone system, characterized in that: The system comprises: A target area verification module is used to obtain the position information of the spraying drone in real time, determine whether the spraying drone has reached the preset target area, and when it reaches the preset target area, collect crop images and spectral data within the preset target area based on the camera, identify the crop information within the target area, and determine whether the crop information within the target area matches the spraying material information; a monitoring area sensitivity determination module, configured to, when the crop information in the target area successfully matches the spraying material information, have the spraying drone enter the monitoring area, identify the crop information in the monitoring area, and determine a predicted sensitivity value of the crops in the monitoring area to the spraying material based on the spraying material information and the crop information in the monitoring area; A spray drift prediction module is used to pre-drive the spraying drone to perform pre-spraying at a preset center point within the target area, obtain real-time meteorological data and operating parameters of the spraying drone, input them into a preset spray drift prediction model, output prediction results, and determine the predicted landing area of ​​the spray droplets when the spraying drone sprays at preset boundary points within the target area based on the prediction results; A boundary point updating module is used to determine the non-physical isolation zone information according to the predicted landing area and the predicted sensitivity value, and update the actual boundary points in the target area according to the non-physical isolation zone information.

2. The intelligent adaptive spraying drone system according to claim 1 is characterized in that: The camera is mounted on the spraying drone and includes a multispectral camera and a high-resolution RGB camera. The content of determining whether the spraying drone reaches the preset target area includes: When the position of the spraying drone is within the preset target area and the residence time of the spraying drone in the preset target area within the preset interval is within the preset time threshold, it is determined that the spraying drone has reached the preset target area. The preset interval is a time interval with the current moment as the end point and the preset time as the interval length; the preset target area refers to the preset planting area that needs to be sprayed.

3. The intelligent adaptive spraying drone system according to claim 1, characterized in that: The spraying material information in the target area verification module includes the spraying material type, concentration, and applicable crop list, wherein the applicable crop list includes the spraying material name, applicable crop type, and applicable growth stage; the spraying material type includes irrigation water, nutrient solution, and liquid medicine; The crop information includes crop type and growth stage; and the content of determining whether the crop information in the target area matches the spraying material information includes: The crop types and growth stages in the crop information within the target area are used as a to-be-tested set; Reading data from the applicable crop list in sequence, traversing the set to be checked according to the read data, and deleting the corresponding element in the set to be checked when a match is successful; When the data in the to-be-checked set is empty, the traversal is stopped and it is determined that the crop information in the target area is successfully matched with the spraying material information; When each data in the applicable crop list is read and the to-be-checked set is not empty, it is determined that the crop information in the target area fails to match the spraying information, and an abnormal prompt message is generated, which carries the crop information and spraying information in the target area.

4. The intelligent adaptive spraying drone system according to claim 1, characterized in that: The monitoring area is an area extending outwards by a preset distance based on the boundary of the target area, and the preset outward extension distance is adjusted according to the spraying object information.

5. The intelligent adaptive spraying drone system according to claim 1, characterized in that: The content of determining the predicted sensitivity value of crops in the monitoring area to the sprayed object based on the sprayed object information and the crop information in the monitoring area includes: Obtaining a basic sensitivity value based on static matching of a preset crop-drug interaction database, that is, matching the crop type, growth stage, and spray type in the crop information within the monitoring area with the preset crop-drug interaction database to determine the basic sensitivity value. The crop-drug interaction database includes spray type, crop type, growth stage, and their corresponding sensitivity values; The predicted sensitivity value is obtained by dynamic correction of real-time meteorological data, wherein the real-time meteorological data includes wind speed, temperature and humidity.

6. The intelligent adaptive spraying drone system according to claim 5, characterized in that: The content of the predicted sensitivity value obtained by dynamic correction of real-time meteorological data includes: determining a correction factor based on the real-time meteorological data; The basic sensitive value is corrected based on the correction coefficient to obtain a predicted sensitive value, the correction coefficient includes a temperature correction coefficient and a wind speed correction coefficient, the wind speed in the real-time meteorological data is positively correlated with the wind speed correction coefficient, and the temperature in the real-time meteorological data is positively correlated with the temperature correction coefficient.

7. The intelligent adaptive spraying drone system according to claim 1, characterized in that: The monitoring area sensitivity determination module further includes a spray stop determination unit, and the spray stop determination unit includes: A sensitivity value comparison subunit, configured to compare the predicted sensitivity value with a preset high-risk sensitivity value; a prohibition determination subunit, configured to generate a spraying operation prohibition prompt message when the predicted sensitivity value is greater than a preset risk sensitivity value, wherein the spraying operation prohibition prompt message carries the predicted sensitivity value; The permission determination subunit is used to generate a prompt message for allowing the spraying operation when the predicted sensitivity value is not greater than the preset risk sensitivity value.

8. The intelligent adaptive spraying drone system according to claim 1, characterized in that: The preset center point refers to the center point of the target area. The preset spray drift prediction model is the AGDISP model, which is based on the Lagrangian particle tracking method and is used to simulate the motion trajectory of the spray. The prediction results include the drift distance range and the drift path. The preset boundary point is a point where the distance between the spraying drone and the boundary of the target area is within a preset distance threshold, which is determined by the type of sprinkler head.

9. The intelligent adaptive spraying drone system according to claim 1, characterized in that: The boundary point updating module includes: an impact area prediction unit, configured to compare the predicted landing area with the target area, and determine that the predicted landing area outside the overlapping area between the predicted landing area and the target area is the predicted impact area; an isolation zone determining unit, configured to determine information of a non-physical isolation zone according to the predicted impact area and the predicted sensitivity value; The boundary point updating unit is used to update the preset boundary point position at the boundary in the direction of the target area according to the non-physical isolation zone information to obtain the actual boundary point position in the target area.

10. The intelligent adaptive spraying drone system according to claim 9, characterized in that: The content of determining the non-physical isolation zone information according to the predicted impact area and the predicted sensitivity value includes: When the predicted sensitivity value is greater than a preset sensitivity threshold, the predicted impact area is expanded outwards according to the magnitude of the predicted sensitivity value; When the predicted sensitivity value is not greater than the preset sensitivity threshold, the predicted impact area is reduced inwardly according to the predicted sensitivity value in the direction of the target area; The information of the non-physical isolation zone is determined based on the location and size of the adjusted predicted impact area.

Images