WO2018178468A1 - Method for the treatment of crops - Google Patents

Abstract

The invention relates to a method for the treatment of crops comprising the following phases: passing an unmanned aerial vehicle over a cropland capturing multispectral and/or tomographic data for different points with the camera; determining a vegetative factor for each of said points based on the information obtained by the camera; generating a recommendation for each of the different points based on the vegetative factor; applying the recommendation to the cropland using a system provided with a fertilizing device and a control system which controls the quantity of fertiliser based on said fertilisation command; and acquiring data relating to the incline of the land for each of said points. The generated fertilisation command depends on the incline of the land at the calculation point and on the typology of the land.

Description

 CULTURE TREATMENT PROCEDURE DESCRIPTION The objective of the present invention is a crop treatment procedure. More specifically, the present invention provides the recommendation and guidance of applications and media management.

The growth of the world population and environmental awareness make better organization and use of farmland necessary, and this need will surely increase in the coming years.

There are different techniques that increase the yields of our crops, but agronomic practices such as irrigation and fertilizer dosing are of vital importance, because they affect two unsustainable resources, such as available water and organic and mineral elements assimilable by plants. On the other hand, in treatments such as pruning or planting, the optimization of your operation can result in very high cost savings.

The differentiation of the casuistry of the determination of the dosage of fertilizer and irrigation is due to the fact that the assimilation and use of these resources by the plants are carried out by two well differentiated paths, from the physiology of the plant. Therefore, for its determination we must introduce ourselves in these physiological processes, for a correct management of the cultural treatments of the crops. Similarly, in pruning and planting, differential treatments can be determined by quantifying vegetation volumes and determining their physiology. The determination of areas with different dosages, without their determination, is not sufficient for proper crop management. It is imperative, the determination of doses that allow to reduce costs of consumption of water and fertilizer, as well as quantify volumes to prune or density of seeds to be used. Japanese Patent Document JPH1 1235124 discloses a process for treatment of crops comprising the phases of:

- Pass an unmanned aerial vehicle equipped with GPS and camera through a farmland, taking visual data with a camera for various points.

- Determine an amount of chlorophyll from the color information obtained by the camera.

 - Determine a growth factor for each of these points.

 - Generate a fertilization order for each of the various points from the growth factor.

 - Fertilize the cultivation land using a vehicle equipped with a fertilization device and a control system that controls the amount of fertilizer from the aforementioned fertilization order.

In JPH1 1235124 it uses a measurement in the visible spectrum to determine the chlorophyll in the crop and based on it give a subscriber recommendation. The procedure disclosed in JPH1 1235124 is suitable for typical oriental rice crops, in which the land is flooded with water. However, according to the applicant's studies, it is excessively inaccurate for non-flooded crops. In particular, the document disclosed is not valid for non-homogeneous land. Due to the types of soil, textures, depths, orientations, water needs, sanitary problems, etc., the same plot or land to be cultivated has a variability that makes fertilizer recommendation systems based on chlorophyll development inappropriate. It is an objective of the present invention to disclose a method that does not have the aforementioned drawbacks.

It is another objective of the present invention to provide a method for controlling the application of fertilizer that is both accurate and easily commercially implemented, without requiring specialized knowledge from the farmer. It is also an object of the present invention to provide a method of controlling crop management that can be implemented in an economical manner.

More particularly, the present invention discloses a method for the treatment of crops comprising the phases of:

- to pass an unmanned aerial vehicle equipped with a GPS and preferably a multispectral camera, through a crop field, taking, for various points, data with said camera, preferably multispectral data.

- determine at least one agronomic index for each of these points based on the information obtained by the camera,

 - generate a crop management recommendation for each of the various points from the aforementioned at least one agronomic index,

 - apply the recommendation in the field of cultivation using a system equipped with a treatment device and a control system that controls an input dosage from the aforementioned recommendation, which also includes the phase of: - acquisition of slope data ground for each of the mentioned points,

and because the recommendation generated depends, in addition to the aforementioned agronomic index, on the slope of the land at the calculation point and on the type of terrain.

Preferably, the process also comprises a phase of determining each crop point, each plant or each tree, as a single and independent unit. The recommendation generated can also be made dependent, for example on meteorology, on the volumes and surfaces of vegetation and on non-aerial point data such as leaf or soil analytics, or probe data. Preferably, the agronomic index comprises a vegetative factor. Preferably, the cultivation recommendation comprises an order or recommendation for application of input, more preferably fertilizer.

The present invention solves the aforementioned problem by considering, for each point for which a recommendation for the application and / or management of crops (for example, a fertilization order), the slope of the land and the type of soil is generated for the generation of a fertilization order. Studies conducted by applicants confirm that fertilization orders generated in this way are more efficient. The applicant has verified that the improvement is due to the fact that the slope and type of terrain influence the availability, assimilation and concentration of the fertilizer or fertilizer dosage. In fact, the existence of a slope can cause a loss of fertilizers due to the influence of gravity, or drag of irrigation or rainwater. The type of soil also influences the retention of the fertilizer or fertilizer supplied.

By means of the present invention it is allowed to take into account that the lands are not homogeneous due to the types of soil, textures, depths, orientations, water needs, etc. The present invention allows identifying the variability within the same crop plot and treating the crop based on said variability.

The aforementioned factors (agronomic index, slope and terrain typology) may preferably be combined by weighted overlap.

In a particularly preferred embodiment, the method comprises the stage of taking altimetric terrain data from the unmanned aerial vehicle to determine the slope data. Alternatively, altimetry data can be taken from a cartographic service. However, in situ data collection is more accurate and beneficial for the final results.

Especially preferably, the present invention also provides for the generation of irrigation orders. The present invention also provides for the generation of orders for pruning, planting and even applications of plant protection products.

For this, preferably, the present invention also comprises the phases of:

- temperature data collection at the mentioned points by means of a thermal chamber,

 - generate an irrigation dosing order for each point from the data obtained through the thermal chamber.

Advantageously, the irrigation dosing order also depends on the slope of the land at the point of calculation and on the type of terrain. The dosage order may also depend on the meteorology, volumes and vegetation surfaces and non-aerial point data such as leaf or soil analytics, or probe data.

Preferably, the vegetative factor, the slope of the terrain, the typology of the terrain and / or the data obtained by the thermal chamber are combined by weighted overlap.

The present invention has the advantage of being able to complement the on-site data collection by means of an unmanned vehicle through a remotely accessible platform. This provides numerous advantages, such as user interactivity.

To do this, preferably, after obtaining data by means of an unmanned aerial vehicle, the data is sent remotely to a remote platform to carry out determinations and generation of said orders. More preferably, the remote platform comprises a user interface for receiving feedback.

Even more preferably, the platform automatically modifies values associated with the weighted overlay based on the aforementioned feedback.

The weighting of the different parameters must be done on a case-by-case basis. The use of a remote access platform allows automatic implementation of learning algorithms that optimize the parameters based on experiences produced in different places. In a particularly advantageous manner, the present invention provides that the dosage order of input (for example, fertilizer) depends on the data taken by the thermal chamber. In fact, a lack of chlorophyll development can be caused by water stress in well-paid areas, especially in areas with a noticeable slope, which results in more difficult occurrence in low-lying areas that have to be runoff receivers.

Advantageously, the aforementioned recommendations and / or orders are sent to an automatic control device of a dispenser installed in an agricultural vehicle. More particularly, the present invention allows generating orders that can be directly integrated into tractor screens and new generation irrigation programs. The present invention can also be used with traditional irrigation systems. For this, the present invention provides for the generation of visual maps with recommendations and guidance of applications and crop management. These maps can be remotely accessible from tablets or mobile phones with GPS (from the "cloud" or "cloud") that can be easily installed on tractors in car navigator mode. The procedure object of the present invention allows the generation of maps and / or agronomic recommendations for decision making in the field. This allows the achievement of objectives such as increasing crop production, optimizing costs and increasing the quality of the resulting crops. The adjustment of the recommendations and orientations of applications and crop management (for example, irrigation and / or fertilizer dosages and / or phytosanitary products) allows to reduce the contamination of soils and underground and surface waters, by reducing the use of nitrates.

For better understanding, some drawings of an embodiment of the advertising arrangement, object of the present invention, are attached by way of explanatory but not limitative example.

Figure 1 shows an unmanned aerial vehicle ("drone") usable in the process object of the present invention.

Figure 2 shows schematically a method object of the present invention.

Figure 1 shows an unmanned vehicle ("drone") -1 - with which the data collection phases of an example procedure according to the present invention can be carried out. Drone -1 - in the example is an n-copter rotor model, valid for relatively small land. For larger land, a fixed-wing drone may be more suitable.

For data collection, drone -1 - in the example comprises a multispectral camera -1 1 - and a thermal camera -12-, which are located in the lower part of drone -1 -. The drone -1 1 - also includes a geolocation device (such as a GPS -13-) that allows geolocation of images taken by cameras -1 1 -, -12-. The drone also comprises a control board -14- for storing the data taken. It would also be possible for the drone to communicate the geolocated data directly without intermediate storage.

Figure 2 schematically shows an exemplary embodiment of a method according to the present invention.

The procedure begins with a data collection phase -100- in which the drone -1 - is blown above the cultivation land -50-. The drone takes data, for various points, using its -1 1 optical camera - and its thermal camera -12-. Through your GPS -13-, these data are geolocated.

The data acquired by the drone are used in a second phase -200- to determine the orders (or recommendations and orientation of applications and crop management) of fertilization and / or irrigation. From the data of the multispectral chamber a vegetative factor can be generated using any type of known calculation, such as the reflectance ratio of plants at 550 nm and 560 nm (NDVI ₅₆ o / NDVI ₆ 6o), indicative of the amount of chlorophyll through the evaluation of nitrogen concentration. Other known indices may also be used, such as the so-called TCARI / OSAVI, SAVI ("Soil-adjusted Vegetation Index"), or the NDRE ("Normalized Difference Red Edge Index").

From the thermal chamber data -12-, water stress indices can be calculated, such as the so-called CWSI ("Crop Water Stress Index"), which can be calculated thanks to the use of a Unmanned vehicle with a suitable sensor, but other indices can also be used, such as the calculation of reference evapotranspiration (Eto) from the Pensman-Montheith model (Alien and others 1998), which requires a crop coefficient which takes into account the crop and its phenological state.

Both indices (vegetative and water stress) present a large uncertainty. To reduce uncertainty, the present invention provides for the use of slope and soil type data, meteorology, volumes and vegetation surfaces and non-aerial point data such as leaf or soil analytics, or probe data. The slope is a property of the land that greatly affects the uncertainty of fertilizer / irrigation recommendations based on the rates of known type calculated, because it affects the retention / accumulation capacity of fertilizers / water in the land. In accordance with the present invention, the slope has to be considered independently of other land properties, such as the type of soil. The slope information can be obtained from altimetric information obtained from the drone. Geographical information can also be acquired or generated to generate an MDT (Digital Terrain Model) of the land to be cultivated.

For a better adjustment of the irrigation and / or fertilizer dosages, consideration may also be taken in any embodiment of the present invention of the phenological state of the crops when generating an irrigation and / or fertilizer order.

A great advantage of this solution is that when working with high resolution images it is possible to work point to point which means plant to plant or tree to tree. Unlike other data analysis mechanisms we are able to identify these points and give them descriptors or factors that allow us to make a recommendation.

The different factors to consider can be combined in a weighted way. A preferred method of weighting is that of weighted overlay. The weighted overlay presents as advantages the possibility of carrying out distributed learning that results in a great ease of adaptation to different cases.

Once the different indexes chosen and weighted in the chosen way have been calculated, you can proceed to make fertilizer recommendations and / or irrigation point by point. This implies dividing the cultivated land -50- into points or zones and assigning a specific fertilizer and / or irrigation recommendation to each zone or point -51 -, -52-, -51 '-, -52'-. When the recommendations are geolocated, a map with the recommendations can be generated. This map can be consulted remotely by accessing the platform or by downloading it on a mobile or tablet -500-, which can be arranged in a -400- agricultural vehicle. In the fertilizer or irrigation phase -300-, the tablet -500- can be used as a reference for irrigation or semi-manual fertilizer, or it can be connected to an automatic irrigation or fertilizer device, so that the tablet controls the irrigation or subscription based on the map generated (or the file with the orders downloaded from the platform). The present invention, thanks to the use of a remote access platform in combination with data collection by unmanned aerial vehicle, allows the accumulation and consultation of information, as well as the possibility of interacting with information for recreation, adjustment and Improvement of the calculation mechanisms of the irrigation and / or fertilizer orders.

The remote access platform will preferably allow access to information in the form of different layers (eg, raster and shp), query of evolution of values as a function of time, elaboration of query scripts between sectors of the same producer agricultural. The platform also allows, for any of the embodiments of the present invention, to complement the data obtained by unmanned aerial vehicle, for example by uploading to the same geolocated photos, which can provide additional information for detection of nutrient deficiencies or pest attacks. and diseases The platform may also feed on information from other sensors and / or databases for the preparation of the agronomic recommendation.

Although it has not been reflected in the figures, the drone can also comprise a topographic team to determine the altimetry of the terrain. The drone -1 - can also typically comprise other equipment such as communication devices, a autopilot, and a battery.

Although the invention has been presented and described with reference to embodiments thereof, it will be understood that these are not limiting of the invention, so that multiple constructive or other details that could be evident to the technicians of the sector could be variable after Interpret the subject matter disclosed in the present description, claims and drawings. Thus, all variants and equivalents will be included within the scope of the present invention if they can be considered to be within the broadest scope of the following claims.

Claims

one . Procedure for the treatment of crops comprising the phases of: - passing an unmanned aerial vehicle equipped with GPS and camera through a crop field taking data with said camera for various points,  - determine at least one agronomic index for each of these points based on the information obtained by the camera,  - generate a crop management recommendation for each of the various points from the aforementioned at least one agronomic index,  - apply the recommendation in the field of cultivation using a system equipped with a treatment device and a control system that controls an input dosage from the aforementioned recommendation, characterized in that it comprises the phase of: - Acquisition of land slope data for each of the aforementioned points, and why the recommendation generated depends, in addition to the aforementioned agronomic index, on the slope of the land at the calculation point and on the type of terrain. 2. Method according to claim 1, characterized in that it comprises the stage of taking altimetric data of the terrain from the unmanned aerial vehicle to determine the slope data. 3. Method according to claim 1 or 2, characterized in that it also comprises the phases of: - temperature data collection at the mentioned points by means of a thermal chamber,  - generate an irrigation dosing order for each point from the data obtained through the thermal chamber. 4. Method according to claim 3, characterized in that the irrigation dosing order also depends on the slope of the land at the calculation point and on the type of terrain. 5. Method according to any one of claims 1 to 4, characterized in that the agronomic index, the slope of the land, the type of terrain and / or the data obtained by the thermal chamber are combined by weighted overlap. 6. Procedure, according to any of the previous embodiments, characterized in that after obtaining data by means of an unmanned aerial vehicle, the data is sent remotely to a remote platform for carrying out determinations and generation of said orders. 7. Method according to claim 6, characterized in that the remote platform comprises a user interface for receiving feedback. 8. Method according to claim 7 and claim 5, characterized in that the platform automatically modifies values associated with the weighted overlay based on said feedback. 9. Method according to any of the preceding claims, characterized in that said recommendations and / or orders are sent to an automatic control device of a dispenser installed in an agricultural vehicle. Method according to any one of claims 3 to 9, characterized in that said recommendation depends on the data taken by the thermal chamber. eleven . Method according to any of the preceding claims, characterized in that the camera is a multispectral camera that takes multispectral data. 12. Method according to any of the preceding claims, characterized in that the input comprises fertilizer. 13. Method according to any of the preceding claims, characterized in that the agronomic index comprises a vegetative factor. 14. Method according to any of the preceding claims, characterized in that it comprises a phase of determining each cultivation point, each plant or each tree, as a single and independent unit. 15. Method according to any of the preceding claims, characterized in that said at least one agronomic index is at least two agronomic indices 16. Method according to any of the preceding claims, characterized in that said recommendation includes orders for sowing and / or pruning and / or application of phytosanitary products.