RU2537912C2 - Method of automatic control of crop condition - Google Patents

Abstract

FIELD: agriculture.

SUBSTANCE: invention relates to the field of agriculture. The method comprises the procedures to obtain information about the physical properties, chemical composition of the soil and weather conditions on the agricultural field, as well as information about the actual yield of the previous year on each part of the agricultural field, compared to the signals of the system of determining the spatial coordinates at the time of harvesting, the use of mathematical models of the influence of soil and climatic factors on the final yield, production of calculations on the parameters of main technologies before sowing plants and carrying out of technological impacts in real time in accordance with these calculations for each part of the agricultural field. Before the start of the vegetation period the optimal program of changes of the plant development indicators average in the field and the parameters of the soil environment is determined by finding the maximum of the parameters of technological operations of the optimality criterion, taking into account the difference between the cost of the crop and the cost of its preparation. In real time at the working pass of the agricultural machine with tools its spatial coordinates are measured, the signals from the meteorological station on the ambient temperature, the level of solar radiation, the precipitation intensity is periodically recorded. According to the measured information the parameters of models of plants and the soil environment are precised, the measured values of the parameters of the plant development and the parameters of soil environment are compared for each part of the field to their optimal average values, according to the results of comparing the corrections to the average optimal values of parameters of technological impacts are formed. For each fragment of the field the size of the overall technological impact is determined, which is created from the optimal average and local correction, which is transmitted by the modem connection in the form of the task to the on-board controller of the machine tool, which carries out the technological impact. The information about the physical properties, chemical composition of the soil and plants is received by periodic sampling on test sites located next to the main field, on which the same culture is cultivated as on the main field, and which differ from each other by different fixed levels of irrigation and doses of mineral fertilizer applications and regulators of growth and development of plants. Simultaneously with the sampling on the test sites by means of aircraft remote sensing the multispectral images of the test sites and the main field are formed, according to the resulting spectral information and the selected samples the mathematical model of the optical measurements is precised, which reflect the connection of condition of the crops and the soil environment on the test sites with the parameters of reflection in all the used spectra, on the spectral information obtained over the entire area of the main field, the condition of the crops and the soil environment on the main field is assessed using the mathematical model of the optical measurements for each time of measurement, according to the obtained estimates and signals from the meteorological station on the ambient temperature, the level of solar radiation and the precipitation intensity the parameters of mathematical models of conditions of crop and soil environment are precised, on which the optimal program of changes of the plant development indicators and soil environment parameters mean in the field are then precised in real time. When working passes of the technological machines simultaneously with the measurement of the spatial coordinates the multispectral pattern of the entire area of the main field is repeatedly formed, on which with the predetermined pitch the condition of the crops and the soil environment is assessed, the resulting estimates on individual parts of the field are compared to their optimal average values obtained during the formation of an optimal program of change of the plant development indicators and soil environment parameters mean in the field. By comparing the results the corrections are formed to the average optimal values of parameters of technological impacts, and for each part of the field the amount of the overall technological impact is determined, created from the optimal average and local corrections in a given spatial coordinate.

EFFECT: method enables to increase the amount and reliability of the process of yield formation while substantial reducing the amount of manual labour.

2 dwg

Description

The invention relates to the field of agriculture and can be used in automated control systems for agricultural technologies using information from optical sounding devices and systems.

A known method of controlling the condition of crops based on information from optical sensors of remote sensing, placed directly on technological machines (see. In the book "Precision agriculture", edited by D. Shpaar, A.V. Zakharenko, V.P. Yakusheva, p. 144, 145. St. Petersburg-Pushkin, 2009). In this method, based on information about the reflection of light in the near infrared and red regions of the spectrum formed by the optical sensors, an NDVI (Normalized Difference Vegetation Index) index is obtained by dividing the difference in the reflection parameters in the near infrared and red ranges by the sum by which they are regulated doses of nitrogen fertilizers, for which first determine the chlorophyll content in plants, on the basis of which the absorption of nitrogen by crops is estimated, and then the doses of nitrogen fertilizers are calculated.

The disadvantage of this method is that the NDVI index, as well as the chlorophyll content in plants, are not indicators of plant growth and development, by which it is possible to obtain a given yield at the end of the growing season. In addition, these methods do not use information about the state of the soil environment, in particular, the content of mineral fertilizers in it and the moisture supply that affects the absorption of nutrients from the soil by crops. Failure to take this information into account leads to the fact that fertilizers that can be sown from the soil with the nearest rainfall are introduced into the plant-soil system, which in turn causes costly fertilizers to be overspended and violates the required optimal course of crop development. In addition, the content of other chemical elements, which lead to additional errors in determining the content of chlorophyll itself, also affects the color of the crop. In these methods, the operation of optimizing the doses of fertilizers was not disclosed, since mathematical models and optimization algorithms are not used here, which does not allow predicting the expected result from the technological operation.

Partially the above disadvantages are eliminated in a method in which the dose of nitrogen fertilizers is determined by comparing the reflection parameters of crops on the main field and on test sites located near the field, and on which the same crop is cultivated as on the main field, and the same technology is applied . Moreover, each test site has different known doses of nitrogen fertilizer application, and the dose of application on the main field is determined by the difference in reflection parameters between crops on the main field and test sites (see "Materials of the All-Russian Conference" Mathematical Models and Information Technologies in Agricultural Biology: Results and Prospects ”, October 14–15, 2010, Agrophysical Research Institute, St. Petersburg, pp. 72-75).

The disadvantage of this method is that the reflection parameters of crops on the main field and test sites are not indicators of plant growth and development, therefore, the doses of fertilizers, determined by the difference of these parameters on the main field and test sites, are not optimal from the point of view of obtaining a given crop productivity . In addition, the presence of fertilizers in the soil, the availability of which depends on soil moisture, is also not taken into account, which leads to overuse of fertilizers and additional crop losses due to lodging of crops.

The prototype of the invention is a method for automated control of crop formation, which includes operations to obtain information about the physical properties, chemical composition of the soil and weather conditions on the agricultural field, as well as information about the actual crop for the previous year on each fragment of the agricultural field, compared with the signals of the system determination of spatial coordinates during harvesting, the use of mathematical models of the influence of soil and climatic factors on final crop, making calculations according to the parameters of the main technologies before planting and real-time technological impacts in accordance with these calculations for each fragment of the agricultural field, and before the start of the growing season, determine the optimal program for changing average field development indicators of plants and soil parameters by searching for a maximum in the parameters of technological operations, an optimality criterion that takes into account the difference between the cost of the crop and At the time of receiving it, in real time, during the working pass of an agricultural machine with implements, its spatial coordinates are measured, the signals from the weather station about the ambient temperature, solar radiation level, precipitation intensity are periodically recorded, the parameters of plant and soil conditions models are specified using the measured information, for of each field fragment, the measured values of the indicators of plant development and soil parameters are compared with their optimal average values, according to the result comparisons form corrections to the average optimal values of the parameters of technological impacts, for each fragment of the field determine the size of the total technological impact, consisting of the optimal average and local corrections, which are transmitted via modem communication in the form of a task to the onboard controller of the gun of the machine carrying out the technological impact (RU 2264703 C2, A01B 79/02, 11/27/2005 bull. No. 33).

The main disadvantage of this method is that in order to clarify the parameters of mathematical models of the state of crops and soil environment, in addition to television imaging of plants, spectral photoelectronic imaging of the soil surface, measurements of soil moisture and density over the entire area of the field are necessary at least once every three days, which under vegetative conditions plants requires a lot of manual labor. In addition, to improve the accuracy and reliability of managing the mineral nutrition of crops, a mathematical model of the chemical state of the soil environment is needed, to clarify the parameters of which requires information on the actual content of available forms of nitrogen N, potassium K and phosphorus P in the soil, which significantly increases the cost of manual labor for selection samples

The invention solves the problem of increasing the size and reliability of the process of crop formation with a significant decrease in the amount of manual labor spent on measurements by increasing the degree of automation at the stage preceding the period of plant vegetation, and in the process of their growth and development.

The inventive method of automated control of the state of crops, as well as the prototype, includes operations to obtain information about the physical properties, chemical composition of the soil and weather conditions on the agricultural field, as well as information about the actual crop for the previous year on each fragment of the agricultural field, compared with signals of the system for determining spatial coordinates during harvesting, the use of mathematical models of the influence of soil and climatic factors on the final crop ay, making calculations according to the parameters of the main technologies before planting and real-time technological impacts in accordance with these calculations for each fragment of the agricultural field, and before the start of the growing season, determine the optimal program for changing the average field development indicators of plants and soil parameters by search for a maximum in the parameters of technological operations of the optimality criterion, taking into account the difference between the cost of the crop and the cost of e about receiving, in real time during the working passage of an agricultural machine with implements, its spatial coordinates are measured, signals from the weather station about the ambient temperature, solar radiation level, precipitation intensity are periodically recorded, the parameters of plant models and soil environment are specified from the measured information, for each fragment of the field compare the measured values of the indicators of plant development and soil parameters with their optimal average values, according to the results of the comparison form corrections to the average optimal values of the parameters of technological impacts, for each fragment of the field determine the size of the total technological impact, consisting of the optimal average and local corrections, which are transmitted via modem communication in the form of a task to the onboard controller of the gun of the machine carrying out the technological impact.

The inventive method differs from the prototype in that information on the physical properties, chemical composition of soil and plants is obtained by periodic sampling at test sites located next to the main field, on which the same crop is cultivated as on the main field, and which differ from each other by different fixed levels of irrigation and doses of mineral fertilizers and plant growth and development regulators, simultaneously with sampling at test sites by means of aviation remote sensing tests generate multispectral images of the test sites and the main field, using the obtained spectral information and selected samples, we refine the mathematical model of optical measurements, which reflects the relationship between the state of crops and the soil environment on test sites with reflection parameters in all used spectra, from spectral information obtained over the entire area of the main fields, using the mathematical model of optical measurements, assess the condition of crops and soil in the main field for each moment the measurement time, according to the estimates and signals from the weather station about the ambient temperature, the level of solar radiation and the intensity of precipitation, specify the parameters of mathematical models of the state of crops and soil environment, which then specify the optimal program for changing the average field development indicators of plants and soil parameters, in real time during the working passes of technological machines simultaneously with the measurement of spatial coordinates, they re-form the multispectral picture all th area of the main field, according to which the state of crops and soil environment is estimated with a given step, the estimates obtained on individual fragments of the field are compared with their optimal average values obtained during the formation of the optimal program for changing the average field development indicators of plants and soil parameters, according to the comparison results form amendments to the average optimal values of the parameters of technological impacts and for each fragment of the field determine the size of the total technological impact, with stacked from the optimal average and local corrections in a given spatial coordinate.

The increase in the value and reliability of the crop while reducing the amount of manual labor spent on the measurements is achieved by preliminary identification of the mathematical model of optical measurement of the state of crops and soil environment on test sites and the use of this model to assess the condition of crops and identification of mathematical models of the state of crops and soil environment on the main field, on the basis of which the optimal program of technological impacts on the main field is formed, to Thoraya adjusted in real time for all the spatial coordinates.

In FIG. 1 presents a general flow chart of the implementation of the proposed method, and in FIG. 2 is a flowchart of the steps necessary to implement the method.

Near the main field 1 (Fig. 1) with the sowing of agricultural crops, test sites 2 with the same agricultural crop are placed. The area of each test site is 20-30 m ² , and their number is 6-10 units. On the main field 1 and on the test sites 2, the same technology is applied, including the same set of technological influences. At the same time, regardless of the state of crops at each test site, they implement their own set of fixed technological impacts, such as irrigation, the application of mineral fertilizers and growth and development regulators. To obtain spectral optical information about the state of crops and soil environment, an unmanned small aircraft 3 is used, flying over the main field 1 and test sites 2 with a given frequency. Information about the state of crops and soil environment on test sites and spectral information from the main field and test sites are stored in a database 4 connected to a control device 5, the output of which is connected to the executive controller of the technological machine 6.

The control device 5 (Fig. 2) contains an identification block for the mathematical model of optical measurements 8, to which is connected an optical spectral information source located on a small unmanned aerial vehicle 3, and an input unit for information on the status of crops and soil medium 7 obtained when sampling with test sites 2. The output of the identification block of the mathematical model of optical measurements 8 is connected to the unit for the operational assessment of the state of crops and the soil environment of the main field 9, to which a source with spectral information located on the unmanned small aircraft 3 and the spatial coordinate determination unit 10. The output of the operational estimation unit 9 is connected to the identification unit for mathematical models of the state of crops and soil environment of the main field 11, to which the weather station 12 is also connected, which contains information about air temperature , solar radiation and precipitation intensity. The output of the identification block 11 is connected to the block for the formation of optimal programs of technological impacts on the main field 13. Such technological impacts are watering and applying mineral fertilizers, plant growth and development regulators. To correct the control in space and time, the control device 5 comprises a unit for dynamically assessing the state of crops and the soil environment of the main field 14, to which a weather station 12 is connected, a source of optical spectral information located on an unmanned small aircraft 3, and a unit for determining spatial coordinates 10. Exit dynamic evaluation unit 14 is connected to the unit for forming a common technological control 15, the output of which, in turn, is connected to the executive controller of those nological machine 6.

The method of automated control of the state of crops on the example of perennial grasses is as follows. After sowing the crops on the main field 1 and test sites 2, in the spring, before the growing season on the main field 1, the main dressing of mineral fertilizers is applied over the entire area, and different amounts of mineral fertilizers from the entire range of possible dose values are applied at test sites 2, which creates different growing conditions of plants. Periodically, at least 5 days later, a small aircraft 3 receives spectral optical images of test sites 2, which are related to the state of the soil environment and are different for individual sites due to differences in the level of technological impacts. At the same time, soil samples are taken at test sites 2 to determine the real state. According to the spectral information and the selected samples in block 8, the mathematical model of the optical measurement of the state of the soil medium is identified as follows:

where Y _G is the vector of spectral parameters of the soil medium containing at least 4 different spectral regions; X _p - the state vector of the soil environment dimension [4 × 1], the components of which are: x _1n - the moisture content of the root layer of the soil, kg / kg; x _2n is the content of available nitrogen, g / kg; x _3n is the content of available potassium, g / kg; x _4n is the content of available phosphorus, g / kg; all states are combined into a vector; H - matrix of parameters estimated by spectral data and analyzes of the samples taken by the least squares method; E _p is the vector of random errors in measuring optical indicators with zero mathematical expectation and the covariance matrix R.

The model of optical measurements of the state of the soil medium (1) is identified before emergence, when their projective area is not more than 25% of the total area of the main field. From this moment, in addition to soil sampling, plant samples are taken, and in addition to model (1), in block 8, identification and models of optical measurement of the state of crops of the following type are carried out:

where Y _x is the vector of spectral indicators of crops, containing at least 3 different spectral regions; p is the vector of scale parameters, P is the matrix of communication parameters between the conditions of sowing and spectral indicators; X _m - the state vector of crops with dimension [3 × 1], for perennial grasses having the following components: x _1m - total mass per unit area, kg / m ² ; x _2m - dry weight per unit area, kg / m ² ; x _3m - nitrogen content in plants, mg / kg; E _m is the vector of random modeling errors. The parameters of model (2) are estimated from the selected plant samples and spectral characteristics by the least square method.

After receiving 5-6 estimates of the parameters of the models (1), (2) for each next multispectral picture of the main field in block 9, the current state of the crops and soil environment of the main field 1 is carried out. For this, optical measurement models (1), (2 ), the current optical spectral measurements and spatial coordinates generated by block 10. Moreover, estimates of the state of the soil environment based on model (1) for the current spectral measurements, estimates Y _G and spatial coordinates (z, h) are constructed as follows:

- оценка вектора состояния для малого участка поля с координатами (z, h), а все другие обозначения соответствуют обозначениям модели (1).Where

is an estimate of the state vector for a small field section with coordinates (z, h), and all other notations correspond to the notation of model (1).

Estimates of the state of crops based on model (2), by the measured vector of optical indicators Y _X and spatial coordinates (z, h) are determined by solving the following optimization problem by the gradient method:

- оценка состояния посева для малого участка поля с координатами (z, h), а все другие обозначения соответствуют обозначениям модели (2).Where $${\stackrel{⌢}{X}}_i(z,h)$$

- assessment of the state of sowing for a small plot of the field with coordinates (z, h), and all other designations correspond to the designations of the model (2).

и почвенной среды $${\stackrel{⌢}{X}}_{п}(z,h)$$

вместе с информацией о метеорологических параметрах, формируемой метеостанцией 12, поступают на блок идентификации 11, в котором оценивают параметры математической модели состояния посевовCurrent crop status assessments $${\stackrel{⌢}{X}}_m(z,h)$$

and soil $${\stackrel{⌢}{X}}_P(z,h)$$

together with the information about the meteorological parameters generated by the weather station 12, go to the identification unit 11, in which the parameters of the mathematical model of the state of crops are evaluated

- вектор состояния модели размерностью [3×1], компонентами которого для многолетних трав являются: х_1м - общая масса на единице площади, кг/м²; х_2м - сухая масса на единице площади, кг/м²; х_3м - содержание азота в растениях, мг/кг; $${\dot{X}}_i(z,h)$$

- производная вектора состояния по времени; (z,h) - пространственные координаты малого участка поля; F - вектор климатических факторов размерностью [3×1], компонентами которого являются: f₁ - среднесуточная температура окружающей среды, °C; f₂ - суточный уровень осадков, мм; f₃ - среднесуточный уровень радиации, Вт/м²; ξ - [3×1] вектор случайных ошибок моделирования, учитывающих все источники неопределенности в модели;Where $$X_m^T=\left[x_{\mathrm{one}m}{x}_{2m}{x}_{3m}\right]$$

- the state vector of the model of dimension [3 × 1], the components of which for perennial grasses are: x _1m - total mass per unit area, kg / m ² ; x _2m - dry weight per unit area, kg / m ² ; x _3m - nitrogen content in plants, mg / kg; $${\dot{X}}_i(z,h)$$

- time state derivative of the state vector; (z, h) - spatial coordinates of a small field section; F is the vector of climatic factors of dimension [3 × 1], the components of which are: f ₁ is the average daily ambient temperature, ° C; f ₂ - daily precipitation, mm; f ₃ - the average daily radiation level, W / m ² ; ξ - [3 × 1] vector of random modeling errors taking into account all sources of uncertainty in the model;

- динамическая матрица, параметры которой отражают инерционность процесса изменения состояния посевов;

- матрица возмущений, отражающая чувствительность вектора состояния посевов к климатическим факторам; $$B_i={\left[\begin{array}{cccc}{b}_{11}& b_{12}& b_{13}& b_{14}\\ 0& 0& b_{23}& b_{24}\\ b_{31}& b_{32}& 0& 0\end{array}\right]}_i$$

- матрица управлений, отражающая чувствительность вектора состояния к компонентам вектора состояния почвенной среды; z, h - пространственные переменные, t∈(t₀,T) - время начала и окончания периода вегетации, X_м(t₀,z,h)=X_м,0(z,h) - начальные условия отрастания многолетних трав;

- a dynamic matrix, the parameters of which reflect the inertia of the process of changing the state of crops;

- a perturbation matrix reflecting the sensitivity of the state vector of crops to climatic factors; $$B_i={\left[\begin{array}{cccc}{b}_{\mathrm{eleven}}& b_{12}& b_{13}& b_{\mathrm{fourteen}}\\ 0& 0& b_{23}& b_{24}\\ b_{31}& b_{32}& 0& 0\end{array}\right]}_i$$

- matrix of controls, reflecting the sensitivity of the state vector to the components of the state vector of the soil environment; z, h are spatial variables, t∈ (t ₀ , T) is the start and end time of the growing season, X _m (t ₀ , z, h) = X _{m, 0} (z, h) are the initial conditions for the growth of perennial grasses;

as well as a mathematical model of the state of soil

- вектор состояния модели размерностью [4×1], компонентами которого являются: х_1n - влажность корнеобитаемого слоя почвы, кг/кг; х_2n - содержание доступного азота, г/кг; х_3n - содержание доступного калия, г/кг; х_4n - содержание доступного фосфора, г/кг;

- производная по времени вектора состояния почвенной среды; U - вектор технологического управления размерностью [4×1], компонентами которого являются: u₁ - расход воды при поливе, кг/м²; u₂ - расход азотных удобрений, г/м²; u₃ - расход калийных удобрений, г/м²; u₄ - расход фосфорных удобрений, г/м²; X_п(t₀,z,h)=X_n0(z,h) - состояние почвенной среды в начале периода вегетации; ζ - [4×1] вектор случайных помех в модели;

- динамическая матрица, отражающая инерционность процесса изменения состояния почвенной среды; $${С}_{п}={\left[\begin{array}{ccc}{c}_{11}& c_{12}& c_{13}\\ c_{21}& c_{22}& c_{23}\\ c_{31}& c_{32}& c_{33}\\ c_{41}& c_{42}& c_{43}\end{array}\right]}_n$$

- матрица управлений, отражающая чувствительность вектора состояния почвенной среды к климатическим факторам; $$B_{п}={\left[\begin{array}{cccc}{b}_{11}& b_{12}& b_{13}& b_{14}\\ b_{21}& b_{22}& b_{23}& b_{24}\\ b_{31}& b_{32}& b_{33}& b_{34}\\ b_{41}& b_{42}& b_{43}& b_{44}\end{array}\right]}_n$$

- матрица управлений, отражающая чувствительность вектора состояния почвенной среды к компонентам технологического управления; $${М}_{п}={\left[\begin{array}{ccc}{m}_{11}& m_{12}& m_{13}\\ m_{21}& m_{22}& m_{23}\\ m_{31}& m_{32}& m_{33}\\ m_{41}& m_{42}& m_{43}\end{array}\right]}_n$$

- матрица связи модели структуры биомассы, с моделью состояния почвенной среды, отражающая чувствительность состояния почвенной среды к состоянию посева.Where $$X_n^T=\left[x_{\mathrm{one}n}{x}_{2n}{x}_{3n}{x}_{\mathrm{four}n}\right]$$

- the state vector of the model with a dimension [4 × 1], the components of which are: x _1n is the moisture content of the root layer of the soil, kg / kg; x _2n is the content of available nitrogen, g / kg; x _3n is the content of available potassium, g / kg; x _4n is the content of available phosphorus, g / kg;

- time derivative of the state vector of the soil environment; U is the vector of technological control of dimension [4 × 1], the components of which are: u ₁ - water flow during irrigation, kg / m ² ; u ₂ - the consumption of nitrogen fertilizers, g / m ² ; u ₃ - the consumption of potash fertilizers, g / m ² ; u ₄ - the consumption of phosphorus fertilizers, g / m ² ; X _p (t ₀ , z, h) = X _n0 (z, h) is the state of the soil at the beginning of the growing season; ζ - [4 × 1] random interference vector in the model;

- a dynamic matrix reflecting the inertia of the process of changing the state of the soil environment; $${\mathrm{FROM}}_P={\left[\begin{array}{ccc}{c}_{\mathrm{eleven}}& c_{12}& c_{13}\\ c_{21}& c_{22}& c_{23}\\ c_{31}& c_{32}& c_{33}\\ c_{41}& c_{42}& c_{43}\end{array}\right]}_n$$

- matrix of controls, reflecting the sensitivity of the state vector of the soil environment to climatic factors; $$B_P={\left[\begin{array}{cccc}{b}_{\mathrm{eleven}}& b_{12}& b_{13}& b_{\mathrm{fourteen}}\\ b_{21}& b_{22}& b_{23}& b_{24}\\ b_{31}& b_{32}& b_{33}& b_{34}\\ b_{41}& b_{42}& b_{43}& b_{44}\end{array}\right]}_n$$

- matrix of controls, reflecting the sensitivity of the state vector of the soil environment to the components of technological management; $$M_P={\left[\begin{array}{ccc}{m}_{\mathrm{eleven}}& m_{12}& m_{13}\\ m_{21}& m_{22}& m_{23}\\ m_{31}& m_{32}& m_{33}\\ m_{41}& m_{42}& m_{43}\end{array}\right]}_n$$

- a matrix of the relationship between the model of the structure of biomass and the model of the state of the soil environment, reflecting the sensitivity of the state of the soil environment to the state of sowing.

The estimation of the parameters of mathematical models (5), (6) is carried out by solving the following optimization problem:

, $$m_{x_{п}}$$

- моделируемые значения векторов состояния посевов и почвенной среды.where g ₁ , g ₂ - weighting factors of the adaptation criterion (7), by means of which the ratio of adaptation errors of the state of crops and the state of the soil environment is established; $$m_{x_m}$$

, $$m_{x_P}$$

- simulated values of the state vectors of crops and soil environment.

, $${\stackrel{⌢}{А}}_{п}$$

, $${\stackrel{⌢}{В}}_{м}$$

, $${\stackrel{⌢}{В}}_{п}$$

, $${\stackrel{⌢}{С}}_{м}$$

, $${\stackrel{⌢}{С}}_{п}$$

, $$\stackrel{⌢}{М}$$

поступают в блок формирования оптимальных программ управления 11. В блоке 11 выбирают такие программы технологического управления, которые обеспечивают максимум прибыли с единицы площади. С учетом введенных обозначений прибыль определяют следующим образом:Estimates of the parameters of mathematical models (5), (6) $${\stackrel{⌢}{\mathrm{BUT}}}_m$$

, $${\stackrel{⌢}{\mathrm{BUT}}}_P$$

, $${\stackrel{⌢}{\mathrm{AT}}}_m$$

, $${\stackrel{⌢}{\mathrm{AT}}}_P$$

, $${\stackrel{⌢}{\mathrm{FROM}}}_m$$

, $${\stackrel{⌢}{\mathrm{FROM}}}_P$$

, $$\stackrel{⌢}{M}$$

enter the block for generating optimal control programs 11. In block 11, such technological control programs are selected that provide maximum profit per unit area. Given the designations entered, profit is determined as follows:

- матрица выхода, посредством которой из всего вектора состояния посевов выделяют товарную часть, Т - индекс транспонирования вектора или матрицы.where c is the price per unit mass of the crop, rubles., $$H^T=[\begin{array}{ccc}0& \mathrm{one}& 0\end{array}]$$

- the output matrix, through which the commodity part is extracted from the entire crop state vector, T is the transposition index of the vector or matrix.

For a given profit P ^* , in block 11, the following control optimality criterion is minimized, ensuring the achievement of a given profit from a unit area of the field, which is equivalent to minimizing deviations from a given profit

where Ω _U is the area of ecologically acceptable controls.

The minimization of the criterion (9) is carried out according to the following gradient sequence of operations

, если U(t)∉Ω_U, $$U_i^{*}(t)=U_{i-\mathrm{one}}^{*}(t)$$

if U (t) ∉Ω _U ,

where Δ _i is the optimization step.

The result of the sequence of operations (10) is a sequence of technological actions U * (t) performed at known time instants, the value of which is the same over the entire area of the main field.

поступает на блок формирования общего технологического управления 15. В блоке 15 оптимальную программу технологического управления U*(t) корректируют по площади основного поля по следующему закону:In real time, when performing technological operations, according to the information on the current optical spectral parameters Y _G , Y _X obtained during flights of a small aircraft 3, current information on the meteorological parameters F formed by the weather station 12, and spatial coordinates (z, h), formed by block 10, in the block of dynamic assessment 14 carry out the assessment of the current state of crops, information about which $${\stackrel{⌢}{X}}_m(t,h,z)$$

arrives at the unit for the formation of the general technological control 15. In block 15, the optimal technological control program U * (t) is adjusted according to the area of the main field according to the following law:

where W ^* (t, h, z) is the general technological control implemented in the spatial coordinate (z, h) and at time t, K is the matrix of parameters of the correcting controller.

The general control W ^* (t, h, z) in the form of a reference signal is transferred to the onboard controller of the technological machine 6, which moves along the main field 1 and performs technological operations, which can be watering, fertilizing with fertilizers or introducing growth and development regulators.

The presence of test sites with different fixed levels of technological impact, which creates different conditions for growing crops, periodic sampling of soil and plants from these sites in combination with optical spectral information obtained from small unmanned aerial vehicles, makes it possible to estimate the parameters of mathematical models with sufficient accuracy optical measurements of the state of crops and soil environment. In turn, the use of these models in combination with optical spectral information on the main field makes it possible to estimate the state of crops and soil medium in the main field with sufficient accuracy to control accuracy without soil and plant sampling and, based on these estimates, identify and adapt mathematical models of the state in real time crops and soil environment. This allows you to optimize the technological management program and adjust it according to the field area, taking into account the spatial heterogeneity of the state of crops and soil environment. In general, this makes it possible to increase crop yields by at least 50% and the reliability (probability) of its production to the level of 0.8-0.85 and, due to this, to obtain an economic effect of at least 10-15 thousand rubles. for crops from one hectare of agricultural land.

Claims (1)

A method for automated control of the state of crops, which includes operations to obtain information on the physical properties, chemical composition of the soil and weather conditions on the agricultural field, as well as information on the actual crop for the previous year on each fragment of the agricultural field, compared with the signals of the spatial coordinate system during the harvest, the use of mathematical models of the influence of soil and climatic factors on the final crop, the production of in the parameters of the main technologies before planting and carrying out technological actions in real time in accordance with these calculations for each fragment of the agricultural field, and before the start of the growing season, determine the optimal program for changing the average field development indicators of plants and soil parameters by searching for the maximum in the parameters of technological operations optimality criterion, taking into account the difference between the cost of the crop and the cost of obtaining it, in real The belts during the working passage of an agricultural machine with implements measure its spatial coordinates, periodically record signals from the weather station about the ambient temperature, the level of solar radiation, the intensity of precipitation, according to the measured information, specify the parameters of plant models and soil environment, for each field fragment compare the measured values of development indicators plants and soil parameters with their optimal average values, according to the comparison results form corrections to the average optical For each parameter of the field, determine the size of the total technological effect, consisting of the optimal average and local corrections, which are transmitted via modem communication in the form of a task to the onboard controller of the gun of the machine carrying out the technological effect, characterized in that the information on physical properties the chemical composition of soil and plants is obtained by periodic sampling at test sites located next to the main field, on which the same crop is cultivated, as in the main field, and which differ from each other in different fixed irrigation levels and doses of mineral fertilizers and plant growth and development regulators, simultaneously with sampling on test sites by means of aerial remote sensing, they form multispectral images of test sites and the main field, according to the received spectral information and selected samples, the mathematical model of optical measurements reflecting the relationship of the state is specified sowing and soil environment on test sites with reflection parameters in all used spectra, from the spectral information obtained over the entire area of the main field, using the mathematical model of optical measurements, the state of crops and soil medium in the main field for each moment of measurement is estimated, according to the estimates and the signals from the weather station about the temperature of the ambient air, the level of solar radiation and the intensity of precipitation are refined by the parameters of mathematical models of the state of crops and soil media, which then clarifies the optimal program for changing the average field development indicators of plants and soil parameters, in real time, with working passes of technological machines simultaneously with the measurement of spatial coordinates, they re-form a multispectral picture of the entire area of the main field, according to which the state is estimated with a given step crops and soil environment, the obtained estimates on individual fragments of the field are compared with their optimal average values obtained during the formation of of the optimal program for changing the field-average indicators of plant development and soil parameters, according to the results of the comparison, they form corrections to the average optimal values of the parameters of technological impacts and for each fragment of the field determine the size of the total technological impact, consisting of the optimal average and local correction in a given spatial coordinate.

Images