Crop nutrient real-time inversion and precise variable fertilization method based on air-ground cooperation of unmanned aerial vehicle

Abstract

The invention provides a crop nutrient real-time inversion and precise variable fertilization method based on air-ground cooperation of an unmanned aerial vehicle, and relates to the field of agricultural precise management. The method comprises the steps of collecting and integrating multi-source data to obtain an original data matrix; preprocessing and fusing the original data matrix to obtain a fused data matrix; based on the fused data matrix, calculating a predicted demand value of the crop for nutrients, and further, calculating time dynamic characteristics of the crop nutrient demand; calculating fertilization amount distribution based on the predicted demand value of crops for nutrients; based on the fertilization amount distribution, optimizing operation paths of the unmanned aerial vehicle and the ground water and fertilizer equipment to obtain an optimal path of air and ground collaborative fertilization; and through a feedback adjustment mechanism, the fertilization amount distribution is dynamically optimized. The problems that an existing crop nutrient management and fertilization technology cannot reflect the dynamic growth condition and environmental condition changes of crops in real time, nutrient requirements of the crops are difficult to accurately calculate, and the fertilization amount and the fertilization position are difficult to adjust in real time are solved.

Description

Technical Field

[0001] The present invention relates to the field of agricultural precision management, and in particular to a method for real-time inversion of crop nutrients and precise variable fertilization based on unmanned aerial vehicle (UAV) air-ground collaboration. Background Art

[0002] With the continuous growth of the global population and the increasing demand for agriculture, improving crop yields and the efficiency of agricultural resource utilization has become an important goal of current agricultural development. Traditional agriculture relies on large-scale fertilization and irrigation. However, with the excessive use of chemical fertilizers, problems such as soil quality degradation, environmental pollution, and unbalanced crop nutrition have become more prominent. Therefore, the research on precise crop nutrient management and precision fertilization technology has become a hot topic in the field of agricultural science and technology.

[0003] In recent years, the rapid development of drone technology, sensor technology and big data analysis technology has provided new opportunities for precision agriculture. Drones can cover large areas of farmland in a short period of time and collect data on crop growth, soil nutrients, meteorological environment, etc. in real time by carrying a variety of sensors. The acquisition of these data can provide precise guidance for the nutrient needs of crops, thereby achieving quantitative, timed and precise fertilization, significantly improving the efficiency of fertilization and crop yields, while reducing resource waste and environmental pollution.

[0004] However, the existing crop nutrient management and fertilization technologies cannot reflect the dynamic growth of crops and changes in environmental conditions in real time, and lack an efficient air-ground collaborative operation mechanism. Therefore, a real-time inversion of crop nutrients and precise variable fertilization method based on UAV air-ground collaboration came into being. Summary of the invention

[0005] The present invention provides a method for real-time inversion of crop nutrients and precise variable fertilization based on air-ground collaboration of unmanned aerial vehicles, so as to solve the problems that the existing crop nutrient management and fertilization technologies cannot reflect the dynamic growth conditions of crops and changes in environmental conditions in real time, resulting in fertilization strategies being difficult to adapt to the rapidly changing farmland environment, thereby causing low fertilization efficiency, and even the possibility of over- or under-fertilization; it is difficult to accurately calculate the nutrient requirements of crops, there is a lack of efficient air-ground collaborative operation mechanism, the operation efficiency is low and the resource utilization rate is not high; and it is difficult to adjust the amount and position of fertilizer in real time to cope with the dynamic changes and uncertainties in the crop growth process, resulting in a large gap between fertilization accuracy and crop response.

[0006] The present invention provides a method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration, comprising the following steps:

[0007] S1. Collect multi-source data through UAV and ground sensor network, and integrate multi-source data synchronously to obtain original data matrix; perform preprocessing operation on original data matrix to obtain preprocessed data; fuse preprocessed data to obtain fused data matrix; calculate predicted nutrient demand value of crops based on fused data matrix through multi-dimensional space and time coupling mechanism and recursive weight mechanism; calculate temporal dynamic characteristics of nutrient demand of crops based on predicted nutrient demand value of crops;

[0008] S2. Based on the predicted nutrient demand of crops, a precise variable fertilization algorithm is used to calculate the fertilizer amount distribution. Based on the fertilizer amount distribution, the operation paths of drones and ground water and fertilizer equipment are optimized to obtain the optimal path for aerial and ground coordinated fertilization. The fertilizer amount distribution is dynamically optimized through a feedback adjustment mechanism.

[0009] Preferably, the S1 specifically includes:

[0010] The time integration operation is used to fuse the preprocessed data into a high-dimensional matrix. The formula is as follows:

[0011]

[0012] Where D is the fused data matrix; t0 and are the start time of data collection and the current time; R vi is the red edge band reflectance; G is the green band reflectance; B is the blue band reflectance; NDVI is the normalized difference vegetation index, which indicates the crop health index; E is the environmental factor; R vi ,G, B, NDVI, and E come from the preprocessed data.

[0013] Preferably, the S1 specifically includes:

[0014] According to the spatial variation characteristics of the fused data matrix and crop health indicators, the contribution of crop spatial distribution dynamics to nutrient inversion is comprehensively analyzed through gradient and curvature analysis combined with a recursive weight mechanism.

[0015] Preferably, the S1 specifically includes:

[0016] Through the multi-dimensional space and time coupling mechanism, the dynamic distribution characteristics and spatial nonlinear change laws of crop nutrients are captured, and the predicted nutrient demand value of crops is calculated through the crop nutrient demand prediction formula; the crop nutrient demand prediction formula is:

[0017]

[0018] in, is the k-th recursive inversion crop nutrient requirement value, which represents the predicted nutrient requirement value of the crop at the current moment; Represents the first-order gradient change of the fused data matrix in the horizontal spatial dimension; is the second-order derivative of the fused data matrix in the vertical spatial dimension, indicating the curvature of nutrient distribution or the severity of change; (x, y) is the spatial coordinate; ω is the time recursive weight factor; It is the inverted crop nutrient requirement value of the k-1th recursion, indicating the historical inversion result.

[0019] Preferably, the S1 specifically includes:

[0020] When the set convergence conditions are met, the final inversion results are output; based on the final inversion results, the instantaneous change rate of nutrient demand is calculated using the target nutrient demand, the predicted nutrient demand value of the crop and the dynamic changes in environmental conditions.

[0021] Preferably, the S2 specifically includes:

[0022] The precise variable fertilization algorithm uses a dynamic nonlinear distribution model to calculate the amount of fertilizer for each plot using the fertilizer amount distribution formula. The fertilizer amount distribution formula is:

[0023]

[0024] Where F(x, y, t) is the fertilizer distribution at spatial position (x, y) and time t; C is the fertilizer concentration factor; N opt Indicates target nutrient requirement; N pred is the predicted nutrient requirement of crops at the current moment; E is the environmental factor; NDVI is the Normalized Difference Vegetation Index, which indicates the crop health index; is the rate of change of the k-th recursive inversion crop nutrient requirement at time t.

[0025] Preferably, the S2 specifically includes:

[0026] UAVs and ground-based fertilization equipment perform aerial and ground operations respectively according to the spatial and temporal distribution of fertilization. The operation paths of UAVs and ground-based fertilization equipment are optimized by balancing the difference in fertilization between UAVs and ground-based fertilization equipment while minimizing the redundancy of their operation paths.

[0027] Preferably, the S2 specifically includes:

[0028] The operation path optimization formula is:

[0029]

[0030] Among them, P opt (t) is the optimal path for aerial and ground fertilization; F uav and Fground are the distribution of aerial and ground fertilizer application respectively; t1 and t2 are the starting and ending time points of the aerial and ground coordinated fertilization tasks respectively; λ is the air-ground coordinated balance factor.

[0031] Preferably, the S2 specifically includes:

[0032] The feedback adjustment mechanism optimizes and adjusts the fertilizer application distribution in real time by combining the optimal path of aerial and ground coordinated fertilization with the dynamic changing trend of crop nutrient demand.

[0033] The beneficial effects of the technical solution of the present invention are:

[0034] 1. Through the collaborative work of drones and ground sensor networks, all-round real-time monitoring of farmland environment and crop status is achieved, providing accurate crop nutrient demand inversion results. The inversion results combined with spatiotemporal distribution characteristics enable fertilization strategies to more accurately match the actual needs of crops, reduce fertilizer waste, and improve resource utilization; by synchronously integrating, preprocessing and fusing multi-source data from drones and ground sensors, the dynamic coupling characteristics of crops and the environment can be captured; through multi-layer recursion and spatiotemporal gradient analysis, not only can the nutrient needs of crops be accurately predicted, but also their dynamic changes can be tracked in real time, so as to adjust the fertilization strategy in time and achieve accurate nutrient replenishment.

[0035] 2. The precise variable fertilization algorithm calculates the distribution of fertilizer application through a dynamic nonlinear distribution model. By combining the dynamic changes in environmental conditions and crop nutrient requirements, it ensures that the fertilization strategy is more efficient and accurate, minimizes the impact of environmental fluctuations on fertilization effects, and thus improves crop yields and fertilizer utilization efficiency. Through the efficient collaborative operation path optimization of drones and ground water and fertilizer equipment, it can not only optimize the distribution of fertilizer application, but also reduce the redundancy of equipment operation paths, reduce operation time, and improve the overall efficiency of fertilization operations.

[0036] 3. The feedback adjustment mechanism can dynamically adjust the distribution of fertilizer application by real-time monitoring of crop status and operation path changes, ensuring that the fertilization strategy can flexibly respond to fluctuations in crop demand and further improve the accuracy and coordination of fertilization operations; through precise variable fertilization, crops can obtain the required nutrients at different growth stages, thereby improving crop yield and quality. At the same time, the optimization and precise control of fertilizer application distribution greatly reduces environmental pollution and fertilizer waste, and improves the sustainability of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration described in the present invention. DETAILED DESCRIPTION

[0038] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0040] The following is a detailed description of a specific scheme of a method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration provided by the present invention in conjunction with the accompanying drawings.

[0041] See attached Figure 1 , which shows a flow chart of a method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration provided by an embodiment of the present invention, the method comprising the following steps:

[0042] S1. Collect multi-source data collaboratively through UAVs and ground sensor networks, and synchronously integrate multi-source data to obtain the original data matrix; preprocess the original data matrix to obtain preprocessed data; fuse the preprocessed data to obtain a fused data matrix; based on the fused data matrix, calculate the predicted nutrient demand value of crops through a multi-dimensional space and time coupling mechanism combined with a recursive weight mechanism; based on the predicted nutrient demand value of crops, deduce the temporal dynamic characteristics of crop nutrient demand.

[0043] Relying on drones and ground sensor networks, we can achieve comprehensive coverage of farmland environment and crop status to collect multi-source data. Drones are equipped with hyperspectral sensors, multispectral cameras and RGB cameras. They scan above farmland according to the preset flight path to collect crop reflectance spectral data, i.e. remote sensing data. Hyperspectral sensors are used to capture subtle spectral changes to quantify leaf health, multispectral cameras are used to obtain crop growth status indicators, and RGB cameras provide support for identifying crop coverage and texture features. At the same time, the ground sensor network will be distributed in key areas of farmland, and the collected ground sensor data will be uploaded in real time through the Internet of Things technology. The ground sensor data covers the moisture and nutrient concentration in the soil and the temperature and humidity changes in the air. Key areas refer to representative and influential areas selected in farmland according to crop growth needs, soil characteristics and fertilization needs, such as areas with different crop growth conditions and soil property differences. The ground sensor network includes equipment such as soil moisture sensors, temperature and humidity sensors and soil nutrient sensors.

[0044] The collected multi-source data are synchronously integrated with time and space coordinates as indexes to form a preliminary raw data matrix; the resolution of the multi-source data is dynamically adjusted according to the flight altitude of the UAV, the sampling frequency of each sensor in the ground sensor network, and the distribution density, ensuring that high-precision data collection requirements are met while satisfying global coverage.

[0045] The raw data matrix is ​​preprocessed by performing spatiotemporal alignment, data cleaning, and format unification to obtain preprocessed data. Specifically, the hyperspectral data from the hyperspectral sensor is processed by dimensionality reduction to extract key spectral features; the multispectral data from the multispectral camera is corrected to eliminate the impact of changes in sunlight intensity; the ground sensor data fills the spatial sampling gaps through the interpolation algorithm, and is finally matched and fused with the remote sensing data, providing a high-quality data input basis for subsequent real-time inversion and precise variable fertilization. The above-mentioned dimensionality reduction, correction, and interpolation algorithms can all use existing technologies.

[0046] The multi-source data collected by drones and ground sensor networks need to simultaneously reflect the temporal dynamic characteristics, spatial distribution patterns and environmental impacts of crops in order to show the dynamic distribution characteristics of crop growth status and environmental conditions. Therefore, the time integration operation is used to fuse the preprocessed data into a continuous high-dimensional matrix. The formula is as follows:

[0047]

[0048] Where D is the fusion data matrix, which represents the combination of multi-source data in time t and spatial distribution; t0 and are the start time of data collection and the current time, indicating the range of the time dimension; Rvi is the red edge band reflectance, a hyperspectral characteristic used to quantify the chlorophyll content of crops, and its temporal variation rate Indicates the dynamic trend of crop photosynthesis capacity; G is the reflectance of the green light band; B is the reflectance of the blue light band; It is the ratio of the reflectance of the green and blue bands. It uses a logarithmic function to smooth the spectral fluctuations and is used to capture the color change characteristics of crop leaves and quantify the chlorophyll content and growth status of crops. NDVI is the normalized vegetation index, which is a crop health index and is used to quantify the health of crop growth. E is an environmental factor, including soil moisture, temperature, etc., and the negative impact of high dynamic environmental conditions is weakened through fractional operations. Among them, R vi , G, B, NDVI, and E all come from preprocessed data. The time integration operation fuses multi-temporal data to ensure that the fused data matrix reflects the dynamic coupling characteristics of crops and the environment.

[0049] According to the spatial variation characteristics of the fused data matrix and crop health indicators, the predicted nutrient demand value of crops at the current time step is calculated. Through multi-layer gradient and curvature analysis and combined with the recursive weight mechanism, the contribution of the spatial distribution dynamics of crops to nutrient inversion is comprehensively considered; through the multi-dimensional space and time coupling mechanism, the formula for predicting crop nutrient demand can capture the dynamic distribution characteristics and spatial nonlinear variation laws of crop nutrients, providing high-precision data support for further precise variable fertilization. The formula for predicting crop nutrient demand is:

[0050]

[0051] in, is the k-th recursive inversion crop nutrient requirement value, that is, the predicted requirement value of the crop for nutrients (such as nitrogen, phosphorus or potassium and other major nutrients) at the current moment; It represents the first-order gradient change of the fused data matrix in the horizontal spatial dimension, which is used to capture the horizontal distribution law of crop status, such as spectral reflectance, vegetation index, etc. is the second-order derivative of the fused data matrix in the vertical spatial dimension, indicating the curvature of nutrient distribution or the severity of change, reflecting the spatial curvature of crop growth status; (x, y) is the spatial coordinate; the trigonometric function sin(πNDVI) is used to introduce nonlinear weights and strengthen the nonlinear contribution modeling of the normalized difference vegetation index NDVI to crop photosynthesis; the exponential function e -E It is used to attenuate the impact of environmental factor E on nutrient prediction to enhance the stability of prediction; ω is the time recursive weight factor, which is used to balance the weight of current input data and historical inversion results to ensure that the recursive process retains the characteristics of current data and has temporal consistency; It is the inverted crop nutrient requirement value of the k-1th recursion, indicating the historical inversion result.

[0052] When the convergence condition is met When the final inversion result N is output pred , represents the predicted nutrient demand of crops at the current moment. ∈ represents the demand change threshold, which is set according to the expert experience method.

[0053] The time dynamic characteristics of crop nutrient demand are further calculated based on the final inversion results, capturing the predicted nutrient demand value of crops, that is, the nonlinear change trend between the predicted nutrient demand and the target nutrient demand. The target nutrient demand, predicted nutrient demand and the dynamic changes of environmental conditions are used to deduce the instantaneous change rate of nutrient demand, thus providing a basis for subsequent fertilization strategies. The calculation formula for time dynamic characteristics is:

[0054]

[0055] in, is the rate of change of the crop nutrient demand of the kth recursive inversion at time t, reflecting the dynamic change of crop nutrient demand; N opt represents the target nutrient requirement, which is artificially determined by the crop type and target yield; N pred is the predicted nutrient requirement of the crop at the current moment; logarithmic function The nonlinear adjustment characteristics of nutrient demand are introduced to avoid the errors that may be caused by direct linear fitting; trigonometric function The weight of the Normalized Difference Vegetation Index (NDVI) on nutrient demand was adjusted to capture the nonlinear effects of crop photosynthesis dynamics; The influence of environmental conditions is weakened into a smooth curve; α is a dynamic adjustment coefficient obtained through experiments and used to balance the weights of input parameters, usually in the range of [0.1, 1].

[0056] S2. Based on the predicted nutrient demand of crops, a precise variable fertilization algorithm is used to calculate the fertilizer amount distribution. Based on the fertilizer amount distribution, the operation paths of drones and ground water and fertilizer equipment are optimized to obtain the optimal path for aerial and ground coordinated fertilization. The fertilizer amount distribution is dynamically optimized through a feedback adjustment mechanism.

[0057] The precise variable fertilization algorithm uses a dynamic nonlinear distribution model to calculate the amount of fertilizer for each plot using the fertilizer amount distribution formula, taking into account the dynamic changes in crop nutrient requirements and environmental conditions to ensure the efficiency and accuracy of fertilization. The fertilizer amount distribution formula is:

[0058]

[0059] Where F(x, y, t) is the fertilizer amount distribution at spatial position (x, y) and time t; C is the fertilizer concentration factor, which is calibrated by the physical properties of the fertilizer and the performance of the operating equipment, and the value is calibrated by experiments; The difference between the target nutrient demand and the predicted nutrient demand is combined, and the impact of environmental fluctuations is weakened in the form of a score. By integrating the above multiple factors, a multi-dimensional coupled dynamic fertilization strategy is realized, providing efficient nutrient supply for crops while avoiding resource waste.

[0060] UAVs and ground-based fertilization equipment perform aerial and ground operations respectively according to the distribution of fertilizer application in space and time. To achieve efficient collaborative operation of UAVs and ground-based fertilization equipment, it is necessary to optimize the operation paths of the two to ensure accurate execution of fertilizer application distribution and maximize operation efficiency. The core of operation path optimization is to balance the difference in fertilizer application between UAVs and ground-based fertilization equipment, while minimizing the redundancy of operation paths of UAVs and ground-based fertilization equipment. The goal of operation path optimization is to minimize the comprehensive cost of the fertilizer change rate and the fertilizer application difference between UAVs and ground-based fertilization equipment during the entire operation time, thereby achieving optimal path while dynamically adjusting the fertilization strategy. The operation path optimization formula is:

[0061]

[0062] Among them, P opt (t) is the optimal path for aerial and ground coordinated fertilization, and the output is a spatial path sequence at time t; F uav and F ground are the distribution of aerial and ground fertilizer application respectively; t1 and t2 are the starting and ending time points of the aerial and ground collaborative fertilization tasks respectively; λ is the air-ground collaborative balance factor, which is determined by experimental fitting and is used to coordinate the aerial and ground collaborative operation effects, thereby minimizing the operation path deviation and fertilization difference.

[0063] In order to further improve the accuracy and coordination of fertilization operations, a feedback adjustment mechanism is proposed to optimize and adjust the fertilizer amount distribution in real time by dynamically monitoring the changes in crop status and operation paths; the core of the feedback adjustment is to combine the operation path optimization results, that is, the optimal path for aerial and ground coordinated fertilization, with the dynamic trend of crop nutrient demand, to dynamically optimize the fertilizer amount distribution and ensure that the fertilization strategy can adapt to the dynamic fluctuations of crop demand in real time. The formula for the optimized and adjusted fertilizer amount distribution is as follows:

[0064]

[0065] Among them, F adj(x, y, t) is the adjusted fertilizer application amount distribution, which represents the optimization result of fertilizer application amount at spatial position (x, y) and time t; δ is the feedback adjustment coefficient, which is obtained through experiments and usually ranges from [0.1, 1.0]; is the time acceleration of crop nutrient demand during the inversion process, indicating the trend of crop nutrient demand change and the degree of rate fluctuation; It is the time change rate of the operation path optimization result, which reflects the adjustment range of the optimal path of aerial and ground coordinated fertilization in the time dimension.

[0066] In summary, a method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration was completed.

[0067] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0068] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0069] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration, characterized in that: The following steps are involved: S1. Collect multi-source data through UAV and ground sensor network, and integrate multi-source data synchronously to obtain original data matrix; perform preprocessing operation on original data matrix to obtain preprocessed data; Fuse the preprocessed data to obtain a fused data matrix; Based on the fusion data matrix, through the multi-dimensional space and time coupling mechanism, combined with the recursive weight mechanism, the predicted nutrient demand value of crops is calculated; based on the predicted nutrient demand value of crops, the temporal dynamic characteristics of crop nutrient demand are calculated; S2. Based on the predicted nutrient demand of crops, a precise variable fertilization algorithm is used to calculate the fertilizer amount distribution. Based on the fertilizer amount distribution, the operation paths of drones and ground water and fertilizer equipment are optimized to obtain the optimal path for aerial and ground coordinated fertilization. The fertilizer amount distribution is dynamically optimized through a feedback adjustment mechanism.

2. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 1 is characterized in that: The S1 specifically includes: The time integration operation is used to fuse the preprocessed data into a high-dimensional matrix. The formula is as follows: Where D is the fused data matrix; t0 and are the start time of data collection and the current time; R vi is the red edge band reflectance; G is the green band reflectance; B is the blue band reflectance; NDVI is the normalized difference vegetation index, which indicates the crop health index; E is the environmental factor; R vi ,G, B, NDVI, and E come from the preprocessed data.

3. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 2 is characterized in that: The S1 specifically includes: According to the spatial variation characteristics of the fused data matrix and crop health indicators, the contribution of crop spatial distribution dynamics to nutrient inversion is comprehensively analyzed through gradient and curvature analysis combined with a recursive weight mechanism.

4. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 3 is characterized in that: The S1 specifically includes: Through the multi-dimensional space and time coupling mechanism, the dynamic distribution characteristics and spatial nonlinear change laws of crop nutrients are captured, and the predicted nutrient demand value of crops is calculated through the crop nutrient demand prediction formula; the crop nutrient demand prediction formula is: in, is the k-th recursive inversion crop nutrient requirement value, which represents the predicted nutrient requirement value of the crop at the current moment; Represents the first-order gradient change of the fused data matrix in the horizontal spatial dimension; is the second-order derivative of the fused data matrix in the vertical spatial dimension, indicating the curvature of nutrient distribution or the severity of change; (x, y) is the spatial coordinate; ω is the time recursive weight factor; It is the inverted crop nutrient requirement value of the k-1th recursion, indicating the historical inversion result.

5. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 4 is characterized in that: The S1 specifically includes: When the set convergence conditions are met, the final inversion results are output; based on the final inversion results, the instantaneous change rate of nutrient demand is calculated using the target nutrient demand, the predicted nutrient demand value of the crop and the dynamic changes of environmental conditions.

6. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 1 is characterized in that: The S2 specifically includes: The precise variable fertilization algorithm uses a dynamic nonlinear distribution model to calculate the amount of fertilizer for each plot using the fertilizer amount distribution formula. The fertilizer amount distribution formula is: Where F(x, y, t) is the fertilizer distribution at spatial position (x, y) and time t; C is the fertilizer concentration factor; N opt Indicates target nutrient requirement; N pred is the predicted nutrient requirement of crops at the current moment; E is the environmental factor; NDVI is the Normalized Difference Vegetation Index, which indicates the crop health index; is the rate of change of the k-th recursive inversion crop nutrient requirement at time t.

7. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 6 is characterized in that: The S2 specifically includes: UAVs and ground-based fertilization and irrigation equipment perform aerial and ground operations respectively according to the spatial and temporal distribution of fertilization. The operation paths of UAVs and ground-based fertilization and irrigation equipment are optimized by balancing the difference in fertilization between UAVs and ground-based fertilization and irrigation equipment while minimizing the redundancy of their operation paths.

8. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 7 is characterized in that: The S2 specifically includes: The operation path optimization formula is: Among them, P opt (t) is the optimal path for aerial and ground fertilization; F uav and F ground are the distribution of aerial and ground fertilizer application respectively; t1 and t2 are the starting and ending time points of the aerial and ground coordinated fertilization tasks respectively; λ is the air-ground coordinated balance factor.

9. The method for real-time inversion of crop nutrients and precise variable fertilization based on UAV air-ground collaboration according to claim 8 is characterized in that: The S2 specifically includes: The feedback adjustment mechanism optimizes and adjusts the fertilizer application distribution in real time by combining the optimal path of aerial and ground coordinated fertilization with the dynamic changing trend of crop nutrient demand.

Images