Crop protection system for an agricultural vehicle and associated protection method

Abstract

The invention relates to a system for protecting a crop plot for an agricultural vehicle comprising a protection device (140, 240) configured to: - measure at least one characteristic quantity of the crops being treated (52); the system further comprising a control unit (110, 210) configured to: - receive a measurement of the at least one characteristic quantity from the protection device (140, 240); - identify the state of the crop being treated (52), the state being a deteriorated state or a preserved state; - transmit a protection signal from the crop plot (50) to the agricultural vehicle (100, 200), or to a remote terminal, when the crop being treated (52) is in a deteriorated state.

Description

[0001] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ Crop protection system for agricultural vehicle and associated protection method TECHNICAL FIELD [1] The present invention relates to the field of agriculture and in particular agricultural vehicles, such as agricultural machines or autonomous agricultural robots. [2] More specifically, the invention relates to a system for protecting a crop plot for an agricultural vehicle, designed to protect the crop plots from possible damage caused by the agricultural vehicle itself, or by its tools. [3] The invention finds applications in particular in agricultural vehicles, for example having automated or autonomous functions, which are dedicated to cultivation in low, medium or high rows, such as salads, wheat or even vines.STATE OF THE ART [4] Agricultural vehicles configured to treat crops using treatment tools are known from the prior art. These tools can be positioned manually by an operator, directly or indirectly via a control system. [5] Agricultural robots moving and treating crops autonomously are also known from the prior art. These agricultural robots include treatment tools. Said treatment tools may include an autonomous positioning system, allowing them, in an optimal operating configuration, to be well positioned relative to the crops to carry out the treatment. [6] In both cases, the treatment operations may require positioning accuracy of the treatment tools of the order of a centimeter.For example, weeding knives can be used in crop plots, such as salads, and incorrect positioning of said weeding knives, incorrect use, the intrusion of an external element into the ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ system (stones, branches, plastic residue, etc.) or untimely adjustment of the tools (due to vibrations, for example) can damage the crop being treated. [7] Incorrect positioning or incorrect use of the treatment tools may be due to a malfunction of the autonomous positioning system, or a malfunction of the treatment tools, the intrusion of an external element into the system (stones, branch, etc.) or an untimely adjustment of the tools (due to vibrations, for example).Also, a malfunction of a control unit or a location device can lead to inaccuracies in the treatment of crops and damage them. [8] In order to verify that the treatment tools do not damage the crop plot, human supervision is currently necessary. [9] This solution is not satisfactory, since the agricultural vehicle cannot be steered freely without supervision of the treatment carried out on the crop plots. In the case of an autonomous agricultural robot, the latter does not operate in complete autonomy in the crop plots.

[0010] Nevertheless, without human supervision, at present, the entire crop can be damaged following a previously mentioned malfunction.

[0011] None of the current systems can simultaneously meet all the required needs, namely to propose a technique for autonomously protecting a plot of crops in real time, reliable and simple to implement.

[0012] Furthermore, no known technique can effectively prevent underground deterioration and destruction of crops, such as deterioration of the roots, which cannot be identified from the surface.

[0013] However, it is desirable to identify such underground deterioration and destruction, because they can have a harmful influence on the growth of the crops, and are only identifiable at a late stage. DISCLOSURE OF THE INVENTION

[0014] The present invention aims to remedy all or part of the drawbacks of the state of the art cited above.

[0015] To this end, the invention relates to a system for protecting a plot of crops, comprising a protection device adapted to be mounted on an agricultural vehicle comprising at least one crop treatment tool, said vehicle comprising advancement means enabling it to advance in a direction referred to as advancement within a plot of crops, said crops being said to be upstream when they are located upstream of the at least one treatment tool before the at least one treatment tool passes over the crops relative to the direction of advancement, and said crops being said to be being treated when they are located at the level of the at least one treatment tool during the passage of at least one treatment tool on crops.

[0016] The crop plot protection device is configured to measure at least one characteristic quantity of crops being treated.

[0017] Said system further comprises a control unit configured to: - receive a measurement of the at least one characteristic quantity from the protection device, in particular the size or position in space of a crop; - identify a state of said crop being treated, said state possibly being a so-called deteriorated state, or a so-called preserved state; - transmit a crop plot protection signal to the agricultural vehicle or to a remote terminal, when the crop being treated is in a deteriorated state.

[0018] The identification of the state of said crop being treated may in particular be a classification of the state of the crop. Indeed, the control unit can discriminate the crops according to a deteriorated state or a preserved state.However, the classification is not necessarily a step of classification into two groups in the algorithmic sense. Indeed, it is possible that only the deteriorated states are identified directly, and possibly recorded in a storage memory, the preserved states being deduced indirectly from the fact that they are not deteriorated states. Thus it is possible that, when a culture is in a preserved state, no particular event (neither alert, nor data archiving, nor action on the system) is observed at the level of the protection system. Indeed, when the cultures are in preserved states during the treatment, this means that no anomaly is detected, and that no particular action on the system is necessary.

[0019] The transmission of a protection signal is carried out when a deterioration of said culture during treatment, or a high probability of such deterioration, has been detected.It is understood that the crop being treated has generally already been damaged at this stage, but the following crops, downstream, are protected by the emission of the protection signal.

[0020] The invention advantageously makes it possible to establish in real time whether a crop being treated is likely to be damaged or not by the treatment tool. The analysis of a characteristic quantity of a crop during its treatment by the treatment tools allows the identification of the state of the crop being treated, which may be a damaged or preserved state. Thus, the invention advantageously makes it possible to protect other crops in the plot when a crop has been damaged, or when the protection device is defective.

[0021] In particular, the invention makes it possible to detect deterioration of crops which is not visible after treatment of a crop.In particular, all-terrain treatments may involve crop movements, which are detectable only during treatment. Thus, the invention makes it possible to detect types of crop damage that were previously not detectable using prior art techniques.

[0022] For example, when a treatment tool, such as a hoeing knife, passes over a crop, and abnormal crop behavior is detected, this may mean that there is a high probability that the treatment tool is hitting the crop, or damaging its roots, for example. In particular, the physical quantity measured, or captured, may be an image. The abnormal behavior of a crop may be manifested by significant movement (such as lifting of the crop out of the ground, lateral shifting, etc.).The device according to the invention makes it possible to identify such significant movements, which may be synonymous with deterioration of the crop, even if the crop returns to its position after the treatment tool has passed over it, and the deterioration is no longer visible. In other words, the invention makes it possible to monitor the occurrence of deterioration in real time at the time the treatment tools pass over it. This makes it possible to ensure that no potential deterioration of crops is ignored, and that appropriate measures can be taken as quickly as possible after the deterioration has occurred.

[0023] Thus, unlike known techniques, the invention proposes to monitor the “instantaneous” state of a crop, during its treatment, to determine whether it may have been damaged, without being interested in its state after treatment, since this may not be representative of its deterioration.^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0024] The invention also relates to a vehicle as described above, comprising a system as described above.

[0025] In a particular embodiment, to identify the state of said crop being treated, the control unit is further configured to: - make a prediction of the state of said crop being treated by means of a machine learning algorithm previously trained from a set of data representative of at least one characteristic quantity of the crop being treated, the state of said crop being treated being said to be deteriorated if said prediction exceeds a predefined threshold, and the state of said crop being treated being said to be preserved if said prediction does not exceed said predefined threshold.

[0026] By using a trained machine learning algorithm, an identification of the state of the crop being treated can be obtained without requiring measurements other than those of quantities characteristic of the crop being treated.

[0027] In a preferred embodiment, said machine learning algorithm is a neural network trained by supervised learning from said data set representative of the at least one characteristic quantity of the crop being treated, in which the data are identified as corresponding to a deteriorated state, or as corresponding to a preserved state.

[0028] Such a type of learning, called "supervised", has the advantage of being able to precisely indicate the state of a crop to the algorithm during the learning phase, resulting in better identification, or classification, during the phase of use of the algorithm.

[0029] In another particular embodiment of the invention, to identify the state of said crop being treated, the control unit is further configured to: - compare the measurement of the at least one characteristic quantity of the crop being treated with data representative of at least one characteristic quantity of an upstream crop, corresponding to the comparison between the at least one characteristic quantity of said crop before the passage of the at least one treatment tool, and the at least one characteristic quantity of said crop during the passage of the at least one treatment tool, the state of said crop being said to be deteriorated if said comparison exceeds a predefined threshold ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^, and the state of said culture being processed being said to be preserved if said comparison does not exceed said predefined threshold.

[0030] Thus, it is possible to determine the state of the crop by a simple comparison with a reference datum. Such an operation has the advantage of great simplicity, in particular at the level of the algorithm implemented in the control unit which executes the aforementioned steps. In addition, such an operation is possible even with a limited number of comparison data, or even a single comparison datum per type of crop.

[0031] In a preferred embodiment, the protection device is further configured to measure at least one characteristic quantity of upstream crops, and in which the data representative of at least one characteristic quantity of the upstream crop is a measurement of at least one characteristic quantity of the upstream crop.

[0032] Thus, the reference data is particularly relevant for making the comparison with the measurement of the characteristic quantity of the crop being treated, in particular because it is the same crop compared at two different times.

[0033] In another preferred embodiment, the data representative of at least one characteristic quantity of the upstream crop is pre-existing data stored in a non-volatile storage memory of the control unit, or in a remote non-volatile storage memory.

[0034] Thus, it is not necessary to carry out an additional measurement, making it possible to save on equipment. In addition, it is possible to compare the measurement of the characteristic quantity of the crop being treated with a single reference data item, or with a limited number of data items.In other words, each crop is not necessarily associated with its own representative data (such as a characteristic quantity of the same crop when it is located upstream), but with a generic representative data, which may be common to several crops being processed.

[0035] These characteristic quantities advantageously make it possible to characterize a crop. If the crop is damaged (for example, cut, moved or twisted) by the processing tools, the characteristic quantity of said crop before it passes through the processing tools is different from the characteristic quantity of said crop after it has been processed by the processing tools.^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0036] In a particular embodiment, the at least one characteristic quantity measured by the protection device is chosen from an image of the crop, the width of the crop, the height of the crop, the angle formed by the crop with the ground, the position of the crop, a spacing between the crop and another crop in the plot.

[0037] In a particular embodiment, the treatment tool(s) are capable of adopting a working position and a rest position. The control unit is configured to transmit the crop plot protection signal to the at least one treatment tool of the agricultural vehicle, so that the at least one treatment tool is positioned in the rest position, among the working position and the rest position that it can adopt.

[0038] This advantageously makes it possible to protect the other crops in the plot when a crop has been damaged, or when the protection device is defective, and to guide the agricultural vehicle to a location where maintenance operations can be carried out easily.

[0039] In a particular embodiment, the control unit is configured to transmit the crop plot protection signal to the forward movement means to cause the agricultural vehicle to stop.

[0040] In a particular embodiment, a crop protection device comprises an image sensor and / or a laser sensor.

[0041] These embodiments offer the advantage to said protection device of limiting its size and mass.

[0042] In a particular embodiment, the system comprises a location device, in which the control unit is configured to associate the measurement of the at least one characteristic quantity of the crop being treated with a position of the crop being treated estimated by the location device.

[0043] This embodiment has the advantage of making the association of the measurements of the characteristic quantities of the same crop, in particular upstream and during treatment, at two given times, more reliable.

[0044] In a particular embodiment, the control unit is remote from the agricultural vehicle, and in which the protection device comprises wireless communication means adapted to transmit data from the control device to the control unit, and vice versa.^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0045] In this way, the operations performed by the control unit can be outsourced and performed remotely. This has the advantage of having, for example, a greater computing capacity, or a greater storage capacity.

[0046] The invention also relates to an assembly comprising a crop treatment tool and a crop protection system, in which the protection device and the control unit of said system are arranged on said treatment tool.

[0047] Thus, the assembly can be adapted to any ordinary agricultural vehicle, which thus becomes an agricultural vehicle equipped with a crop protection system.

[0048] The invention also relates to an agricultural vehicle configured to receive such an assembly, said treatment tool being mechanically connected to said agricultural vehicle.

[0049] The invention also relates to an autonomous agricultural robot, comprising a protection system according to the invention.

[0050] According to a particular embodiment, the robot comprises a protection system, in which the control unit of the protection system corresponds to a central control unit of said robot.

[0051] The invention also relates to a method for protecting a plot of crops implemented by a protection system, comprising the steps of: - Measurement by the protection device of at least one characteristic quantity of a crop being treated; - Transmission of said measurement to the control unit; - Identification of the state of said crop being treated, said state possibly being a so-called deteriorated state, or a so-called preserved state; - Transmission of a signal for protecting the plot of crops to the agricultural vehicle or to a remote terminal, when the crop being treated is in a deteriorated state.

[0052] In a particular mode of implementation, implemented by a protection system according to a first embodiment, the classification step comprises the steps of: - Performing a prediction of the state of said crop being treated by means of a machine learning algorithm previously trained from a set of data representative of at least one characteristic quantity of the crop being treated, the state of said crop being treated being said to be deteriorated if said prediction exceeds a predefined threshold, and the state of said crop being treated being said to be preserved if said prediction does not exceed said predefined threshold.

[0053] In another particular mode of implementation, implemented by a protection system according to a second embodiment, the classification step comprises the steps of: - Association by the control unit of the measurement of a crop being treated from the protection device with data representative of at least one characteristic quantity of an upstream crop; - Comparison by the control unit of said measurement with said data representative of at least one characteristic quantity of an upstream crop, the state of said crop being treated being said to be deteriorated if said comparison exceeds a predefined threshold, and the state of said crop being treated being said to be preserved if said comparison does not exceed said predefined threshold.

[0054] In a preferred embodiment, implemented by a protection system comprising a location device, the method comprises an additional preliminary step of recording the location of a crop by the location device, and the step of association by the control unit comprising a first association of a measurement of at least one characteristic quantity of a crop being treated from the protection device with a position estimated by the location device of the characterized crop and a second association of a measurement from the protection device and a data item representative of at least one characteristic quantity of an upstream crop associated with the position of the crop being treated.

[0055] This makes it possible to make reliable the association of measurements of characteristic quantities, from the protection device, of the same crop at two given times, before and during treatment by the treatment tools.BRIEF DESCRIPTION OF THE FIGURES

[0056] Other advantages, aims and particular characteristics of the present invention will emerge from the following non-limiting description of at least one particular embodiment of the devices and methods which are the subject of the present invention, with reference to the appended drawings, in which: ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ − Figure 1 is a schematic side view of an autonomous agricultural robot equipped with a protection system at a time t1; − Figure 2 is a schematic side view of an autonomous agricultural robot equipped with a protection system in the same embodiment illustrated in Figure 1 at a time t2; − Figure 3 is a schematic side view of a tractor equipped with a treatment tool with a protection system, in another embodiment; − Figure 4 is a flowchart illustrating the method of protecting a crop plot.DETAILED DESCRIPTION OF THE INVENTION

[0057] The present description is given without limitation, each characteristic of an embodiment being able to be combined with any other characteristic of any other embodiment in an advantageous manner.

[0058] It should be noted, from now on, that the figures are not to scale.

[0059] Figure 1 and Figure 2 illustrate a side view of an autonomous agricultural robot 100. Said autonomous agricultural robot comprises a chassis 160 on which a control unit 110 is arranged, and processing tools 120 making it possible to process crops 50.

[0060] The control unit 110 comprises, in a conventional manner, a computer, or processor.

[0061] The autonomous agricultural robot 100 can move in a crop plot 50. A crop plot 50 can comprise several crops of the same type or of different types.

[0062] The processing tools 120 may comprise, in a non-limiting manner, kress fingers, lump-breaking discs, mechanical interceptors, notched discs, or even knives.

[0063] The processing tools 120 may also be movable between two positions, a so-called working position and a rest position, in particular a raised position. In the working position, the treatment tools 120 treat the crops 50. In the rest position, the treatment tools 120 have no contact with the crops 50. All or part of the treatment tools 120, in the rest position, can be inclined substantially parallel to the ground on which the autonomous agricultural robot 100 is moving. All or part of said treatment tools 120 can also be offset transversely relative to the direction of advance 101.

[0064] The autonomous agricultural robot 100 may also comprise an automatic device for positioning the treatment tools 120. Said device is conventionally connected to the control unit 110. Said control unit is then configured to transmit to it orders for positioning the treatment tools 120, said treatment tools then positioning themselves in the transmitted position.

[0065] In a non-limiting manner, the crops 50 are crops arranged in rows in the embodiment described. The crops 50 may be large, such as vines, medium-sized such as wheat, or small, such as lettuce.

[0066] The autonomous agricultural robot 100 comprises advancement means 130. Said advancement means allow the autonomous agricultural robot 100 to advance in a direction 101 called advancement.

[0067] The control unit 110 is conventionally connected to the treatment tools 120 and to the advancement means 130. It can be configured to send an order to place the treatment tools 120 in the rest position. In this case, the treatment tools 120 are positioned in the rest position. Similarly, the control unit 110 can be configured to send a stop order to the advancement means 130. In this case, the advancement means brake until the autonomous agricultural robot 100 stops.

[0068] The autonomous agricultural robot 100 can advance at a speed called the treatment speed when treating the crop plots.

[0069] The autonomous agricultural robot 100 also comprises a protection device 140 for the crop plot.Crop plot protection means that if a crop 50 (or a limited number of crops 50) has been damaged by the treatment tools 120, the other crops in the crop plot 50 are protected from the treatment tools 120.

[0070] In the present exemplary embodiment, the control unit of the protection system corresponds to the control unit 110 of the robot. The protection device 140 and the control unit 110 together form a crop plot protection system. ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0071] Upstream crop 51 denotes a crop located upstream of the treatment tools 120 before the treatment tools 120 pass over the crops 50 relative to the direction of advance 101.

[0072] Crop being treated 52 denotes a crop located at the level of the treatment tools 120 during the passage of the treatment tools 120 over the crops 50.

[0073] Therefore, an upstream crop 51 at a time t1 can become a crop being treated 52 at a time t2, if t1 is less than t2.

[0074] The protection device 140 measures characteristic quantities of the crops 50.

[0075] The characteristic quantities measured are intended to be representative of the state of the crops 50.

[0076] The protection device 140 of the plot of crops 50 can comprise a first characterization device 141 for the upstream crops 51 and the crops being treated 52. The characterization device 141 is then arranged so that it can characterize the upstream crops 51 and the crops being treated 52 without changing orientation. For example, an orientation at an angle relative to the ground on which the agricultural robot is moving can be chosen, so as to cover an area at the level of the treatment tools 120 and an area upstream of the agricultural robot.

[0077] The sensor can be connected to the control unit 110.

[0078] In a first embodiment shown in Figures 1 and 2, the characterization device 141 can comprise image sensors. These image sensors can be CMOS sensors.

[0079] The characterization device 141 can also comprise, additionally or complementary, laser sensors.

[0080] These sensors can be oriented towards the ground. They can also be oriented in a plane normal to the direction of advance 101, or even take an intermediate orientation, at an angle for example.

[0081] The measured characteristic quantities can be an image of the crop.

[0082] The characteristic quantities can also be the width of the crop 50, the width being defined as a dimension of the crop in a plane parallel to the ground on which the autonomous agricultural robot 100 is moving and perpendicular to its direction of advance 101.^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0083] The control unit 110 can be configured to extract the width of a crop from an image extracted from a CMOS sensor following image processing.

[0084] The characteristic quantities can also be the height of the crop 50. The height is defined as the dimension of the crop 50 along an axis substantially perpendicular to the ground on which the autonomous agricultural robot 100 moves.

[0085] The control unit 110 can be configured to extract the height of a crop 50 from an image extracted from a CMOS sensor following image processing.

[0086] The characteristic quantities can also be the angle that the crop 50 forms in relation to the ground on which the autonomous agricultural robot 100 is moving.

[0087] The characteristic quantities can also be the position of the crop 50.

[0088] Similarly, the characteristic quantities can also be a spacing between the crop 50 and another crop in the plot.

[0089] Preferably, the spacing is measured between two neighboring crops 50, that is to say that one is processed one just after the other by the processing tools 120.

[0090] The control unit 110 can be configured to extract the angle that a crop 50 forms with the ground on which the autonomous agricultural robot 100 is moving from an image extracted from a CMOS sensor following image processing.

[0091] The characterization device 141 can be positioned on the autonomous agricultural robot 100 respectively at the level of the treatment tools 120 according to the direction of advancement 101. Preferably, it is fixed on the autonomous agricultural robot 100.

[0092] The control unit 110 is configured to receive the measurements of the characteristic quantities from the protection device 140 of the crops 50.It can also be configured to process the measurements of the characteristic quantities recorded by the protection device 140 of the same crop 50 during the passage of the treatment tools 120, and optionally before the passage of the treatment tools 120.

[0093] The control unit 110 is configured to receive said measurements of the characteristic quantities from the protection device 140.

[0094] The control unit 110 is also configured to identify a state of the crop being treated 52, said state possibly being a so-called deteriorated state, or a so-called preserved state. ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0095] The identification of the state of the culture being treated 52 can be carried out in several ways, which will be explained in the remainder of the description.

[0096] When the state of the crop being treated is identified as being in a deteriorated state, the control unit 110 can be configured to send a protection signal to the autonomous agricultural robot 100, or to an external terminal.

[0097] The protection signal can take the form of a protection order sent to the autonomous agricultural robot.

[0098] In this case, the protection signal can take the form of a stop order sent directly to the agricultural vehicle, which can also take several forms.

[0099] Thus, the control unit 110 can be configured to send a stop order to the advancement means 130.

[0100] Also, the control unit 110 can be configured to transmit an order to place the treatment tools 120 in the rest position.

[0101] Furthermore, the protection signal can take the form of a protection signal sent to a remote terminal.Such a protection signal then takes the form of an alert or an error message to an external medium intended for an operator, for example inviting an operator to act on the agricultural vehicle.

[0102] These protection signals can be cumulative.

[0103] Optionally, before transmitting a protection signal, it is possible to carry out an automatic adjustment of the treatment tools 120. Such an adjustment can for example consist of a repositioning of all or part of the treatment tools 120.

[0104] As presented above, the state of the crop being treated 52 can be identified as being a so-called deteriorated state, or a so-called preserved state.

[0105] According to a first variant, the identification of the state of the crop being treated 52 can be carried out by means of a machine learning algorithm, also known as “machine learning”.

[0106] More specifically, the control unit 110 is then configured to make a prediction of the state of said culture being treated 52 by means of a machine learning algorithm.

[0107] Such a machine learning algorithm may for example be a neural network. ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0108] The neural network may also be of the convolutional type (known by the acronym "CNN").

[0109] The machine learning algorithm is previously trained from a set of data representative of at least one characteristic quantity of the crop being processed 52.

[0110] The training data may have been generated during prior measurements. It is also possible to increase the measured data in order to generate a larger training data set.For example, in the case of a characteristic quantity corresponding to an image, a captured image can be decolorized, detextured, rotated, cropped, etc.

[0111] According to a first possibility, the learning from the dataset is unsupervised, that is to say that the data in the dataset are not labeled or identified. In other words, it is not specified whether a piece of data is associated with a deteriorated state or a preserved state.

[0112] According to a second possibility, the learning from the dataset is preferentially supervised. In this case, the data are identified as corresponding to a deteriorated state, or as corresponding to a preserved state.

[0113] In particular, the characteristic quantity corresponds to an image. Such a type of characteristic quantity is particularly suitable, in particular, for use with neural networks, for the purpose of image classification.

[0114] The image of a preserved crop can for example be identified by a regular orientation of the crops.

[0115] The image of a damaged crop can for example be identified by an irregular orientation of all or part of the crops 50. Indeed, for example when the treatment tools 120 work underground, an involuntary deterioration of the roots can be manifested by a movement of the crops 50 during the passage of the treatment tools 120.

[0116] For example, when the characterization device 141 is an image sensor or camera, the latter can capture a stream of images, for example a stream of 5 to 60 images per second.

[0117] Each image of the stream can be processed by the previously trained neural network. The output of the neural network corresponds to a prediction of the state of said crop being processed 52.This prediction may take the form of a probability, for example expressed by a number between 0 and 1, that the culture is in a deteriorated state, or conversely, that it is in a preserved state.

[0118] The culture being treated 52 is then classified as being in a deteriorated state if the prediction exceeds a predefined threshold, otherwise it is classified as being in a preserved state, if the prediction does not exceed said predefined threshold.

[0119] The predefined threshold may be set by an operator of the autonomous robot 100. For example, the predefined threshold may be set between 0.7 and 0.95, corresponding to a probability ranging from 70% to 95% that the crop being treated 52 is in a deteriorated state.

[0120] According to a second variant, the identification of the state of the crop being treated 52 can be carried out by means of a comparison with data representative of a characteristic quantity of an upstream crop 51.

[0121] The control unit 110 can then be configured to carry out a comparison between the measurement of a characteristic quantity of the crop 50 being treated 52 and data representative of a characteristic quantity of an upstream crop 51, corresponding to the comparison between a characteristic quantity of the crop before the passage of the treatment tools 120, and a characteristic quantity of said crop during the passage of the treatment tools 120.

[0122] According to a first possibility, measurements of the characteristic quantities of the same crop 50 before and during the passage of the treatment tools 120 are recorded by the protection device 140.

[0123] In such a case, the protection device 140 is configured to measure at least one characteristic quantity of upstream crops 51, for example also by means of the characterization device 141.

[0124] The data representative of a characteristic quantity of the upstream crop 51 is then a measurement of a characteristic quantity of the upstream crop 51.

[0125] According to a second possibility, the control unit 110 is configured to carry out a comparison of the measurements of the characteristic quantities recorded by the protection device 140 of the same crop 50 during the passage of the treatment tools 120 with a data representative of at least one characteristic quantity of an upstream crop 51, which is different from a measurement.^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0126] The control unit 110 is then configured to associate the measurement of a characteristic quantity of a crop being treated 52 with data representative of a characteristic quantity of an upstream crop 51.

[0127] For example, the data representative of a characteristic quantity is pre-existing data stored in a non-volatile storage memory of the control unit 110, or in a storage memory, in particular non-volatile, remote.

[0128] The pre-existing data may for example have been generated during a previous use of the agricultural robot, or during an independent data generation operation. For example, it is possible to generate a set of pre-existing data by flying over the crop plot using a drone equipped with a crop characterization device, such as a camera.

[0129] It is also possible for the pre-existing data to be a single image of a crop 50 in a preserved state. Such an image then serves as a reference for all the crops being processed 52.

[0130] Whatever the type of data representative of a characteristic quantity of an upstream crop 51, the control unit 110 is configured to identify the state of the crop being processed 52, in particular to classify the crop being processed 52 as being in a deteriorated state if the comparison exceeds a predefined threshold, or as being in a preserved state if the comparison does not exceed the predefined threshold.

[0131] For example, when the characteristic quantity is a width of the crop 50, the pre-existing data is the width of an upstream crop 51. The measurement of the width of the crop being processed 52 is compared to said width of the upstream crop 51.For example, a width ratio may be defined, such as a ratio of the width of the upstream crop 51 to the width of the crop being treated 52. Alternatively, it may be defined that when this ratio is greater than a predefined threshold equal to 1 (indicating lateral movement of all or part of the crops), then the crop being treated 52 is classified as being in a deteriorated state.

[0132] According to other examples of characteristic quantities presented further upstream, when the characteristic quantity is an image, if a difference in pixels is observed between the two images of the same culture 50 before and during the ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ processing by the processing tools 120, this may mean that the culture 50 has been degraded.

[0133] In particular, if the compared images do not have the same orientation, for example, it may be necessary to straighten the images in order to make the comparison. According to other possibilities, it may also be necessary to crop or rotate the image. It may also be useful to flatten the image, in particular to take into account differences in height when the ground is not flat. The techniques for carrying out this type of processing are known in the art.

[0134] When the characteristic quantity is a width or a length of the crop, a difference in the characteristic quantities of the same crop 50 before and during processing by the processing tools 120 may in particular highlight a cutting or tearing, in particular underground, of the crop 50.

[0135] When the characteristic quantity is a position or a spacing between the crop 50 and another crop in the plot, a difference in the characteristic quantities of the same crop 50 before and during treatment by the treatment tools 120 may in particular highlight a displacement or a tearing, in particular underground, of the crop 50.

[0136] When the characteristic quantity is the angle that the crop 50 forms with respect to the ground on which the autonomous agricultural robot 100 is moving, a difference in the characteristic quantities of the same crop 50 before and during treatment by the treatment tools 120 may in particular highlight a crushing of the crop 50, particularly if it is a tall crop.

[0137] Furthermore, the protection system, or the agricultural robot, may also comprise a location device 170.The location device 170 comprises for example a satellite positioning system, such as the GPS system (from the English acronym Global Position System).

[0138] The control unit is then configured to associate with each measurement of a characteristic quantity of a crop 50 a positioning of said crop, determined by the location device 170. Characteristic quantities associated with the same positioning are considered as being associated with the same crop 50. Example of a second particular embodiment ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0139] Figure 3 illustrates an agricultural vehicle 200 equipped with a system for protecting a plot of crops according to the invention. The agricultural vehicle 200 is here a tractor-type vehicle, equipped with a crop treatment tool 220 and a crop protection system. The agricultural vehicle 200 is equipped with forward movement means 230.Optionally, the agricultural vehicle 200 is equipped with a location device 270.

[0140] The crop protection system comprises a control unit 210 and a protection device 240, equipped with a characterization device 241 which may be an image or laser sensor.

[0141] In particular, the protection device 240 and the control unit 210 are arranged on the crop treatment tool 220.

[0142] However, it is conceivable that these elements are arranged at any other suitable location on the agricultural vehicle 200. In particular, the control unit 210 may be remote from the agricultural vehicle 200.

[0143] The protection system of the agricultural vehicle 200 is similar to the protection system of the autonomous agricultural robot 100, in its structure and operation.

[0144] In general, the control unit 210 is distinct from a control unit of the agricultural vehicle 200, when it has one.However, it is not excluded that the control unit 210 corresponds to a control unit of the agricultural vehicle 200.

[0145] In general, the protection signal sent by the control unit is intended for an operator. The operator then chooses to stop the agricultural vehicle 200, and / or to position the crop treatment tool 220 in the rest position. Such a signal can be displayed on a screen of the agricultural vehicle 200, or on a mobile terminal. For example, the mobile terminal can be a mobile phone or a connected watch. However, it is conceivable that the protection signal is sent directly to the agricultural vehicle 200, in the form of a protection order as mentioned earlier in the description.

[0146] Furthermore, the invention also relates to a method 300 for protecting a plot of crops 50, illustrated in FIG. 4.This method can be implemented by a protection system equipping an autonomous agricultural robot 100 or an agricultural vehicle 200 previously described, in any of its embodiments, when the autonomous agricultural robot 100 or the agricultural vehicle 200 is running. ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^

[0147] The method 300 comprises a first step 310 of measuring a characteristic quantity of a crop being treated 52 by the protection device.

[0148] The protection device then transmits the measurement to the control unit during a step 320. The transmission of the measurements is preferably carried out by wire. It can also be carried out by other transmission means.

[0149] During an identification step 330, the control unit identifies the state of the culture being processed 52.In particular, the control unit classifies (or classifies) the state of the crop being treated 52 as being in a so-called deteriorated state, or as being in a so-called preserved state.

[0150] The method 300 also comprises a step 340 of transmitting a signal for protecting the crop plot by the control unit to the agricultural vehicle, or to a remote terminal, if the crop being treated is identified or classified as being in a deteriorated state.

[0151] The protection signal may be transmitted by the control unit to the advancement means of the agricultural vehicle during a step 350. The advancement means then cause the autonomous agricultural vehicle to stop. The crop plot 50 is thus protected.

[0152] The protection signal may be transmitted by the control unit to the treatment tools during a step 360. It may be an order to put the vehicle in the rest position.The processing tools then no longer have contact with the rest of the crop plot 50. The crop plot 50 is thus protected.

[0153] The protection signal can be transmitted by the control unit to an external operator via an external terminal or medium (not shown). The operator can then control the agricultural vehicle to secure the crop plot 50.

[0154] The protection order can comprise, for an agricultural vehicle comprising an automatic device for positioning the processing tools during a step 370, an analysis of the comparison and a transmission to the processing tools of an order for positioning the processing tools. Thus, if the characteristic quantity, for example, in a non-limiting manner, is an image, an analysis can be carried out on the distribution of the pixels between the image of the upstream crop 51 and the image of the crop being processed 52.For example, this comparison ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ may highlight that the crop being treated 52 has been damaged (cut for example) at one of its ends. This may mean that the treatment tools have become misaligned and therefore incorrectly positioned.

[0155] According to a first variant, the identification is carried out using a machine learning algorithm.

[0156] The identification step 330 then comprises the following steps.

[0157] A step 331 consists of making a prediction of the state of the culture being treated 52 by means of a machine learning algorithm previously trained from a set of data representative of at least one characteristic quantity of the culture being treated 52.

[0158] The structure of the machine learning algorithm, which may for example be a neural network, is known from the state of the art.The input data of the network consists of a characteristic quantity of the crop being treated 52, and the output consists of a prediction taking for example the form of a probability (between 0 and 1) that a crop being treated is deteriorated.

[0159] A step 332 consists of determining the state of the crop being treated, and in particular of classifying the crop being treated 52 as being in a deteriorated state if said prediction exceeds a predefined threshold, or a classification of said crop being treated as being in a preserved state if said prediction does not exceed said predefined threshold.

[0160] The predefined threshold corresponds to a probability value beyond which a crop is considered to be deteriorated.

[0161] According to a second variant, the identification is carried out by means of a comparison of the measurement of a culture being processed 52 with data representative of a characteristic quantity of an upstream culture 51.

[0162] The identification step 330 then comprises the following steps.

[0163] A step 333 consists of associating the measurement of a crop being treated with data representative of a characteristic quantity of an upstream crop 51.

[0164] When the method is implemented by an agricultural vehicle comprising the satellite positioning system, such as the GPS system (from the English acronym Global Position System), the method 300 may comprise a complementary preliminary step 315 ^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^. Step 315 comprises a survey of the location of the crops 50 using the satellite positioning system.

[0165] Step 333 can then comprise a first association of each measurement of characteristic quantities of a crop 50 by the protection device with a positioning estimated by the device for locating the characterized crop 50.

[0166] Step 333 can also comprise a second association of a measurement of characteristic quantities of an upstream crop 51 from the protection device and a measurement of characteristic quantities of a crop being treated 52 from the protection device 40, the two measurements being associated with the same positioning of the characterized crop. A tolerance threshold can be applied to the measurement of the positioning.

[0167] The method comprises a step 334. Step 334 comprises a comparison of the measurement of a characteristic quantity of a crop being treated 52 with the data representative of a characteristic quantity of an upstream crop 51.

[0168] The method also comprises a step 335 for determining the state of the crop being treated 52. The crop being treated is considered to be in a deteriorated state if the comparison exceeds a predefined threshold, to within a tolerance, or to be in a preserved state if said comparison does not exceed said predefined threshold, to within a tolerance.

[0169] Generally speaking, the predefined threshold can be set by an operator, via an external support (not shown) connected to the control unit for example. Said threshold can depend on the tolerance of the comparison of the characteristic quantities compared accepted by the operator and on the precision of the protection device.

[0170] The analysis of the comparison can make it possible to determine the distance by which the treatment tools must be repositioned so that said tools do not damage the plot of crops 50.The control unit extracts said distance and transmits it to the automatic device for positioning the treatment tools.

[0171] When the comparison is greater than the predefined threshold, the control unit can also generate a slowdown order to the advancement means.

[0172] If after step 332 or 335, the predefined threshold is no longer exceeded for a predefined duration, the control unit can generate an order to return to the treatment speed to the advancement means. This means that the treatment tools have been set and no longer pose a danger to the crop plot 50. The predefined duration can be integrated into the control unit by the operator.

[0173] If after step 332 or 335, the predefined threshold is exceeded for a predefined duration, the control unit can generate an order to stop the agricultural vehicle.This means that the treatment tools have not been adjusted appropriately and that the crop plot 50 is not safe. The predefined duration can be integrated into the control unit by the operator.

[0174] Similarly, if after step 332 or 335, the predefined threshold is exceeded for a predefined duration, the control unit can generate an order to place the treatment tools in the rest position. This means that the treatment tools have not been automatically adjusted appropriately and that the crop plot is not safe. The predefined duration can be integrated into the control unit by the operator.

[0175] The method can comprise an additional step of maintaining the treatment tools. The treatment tools can be adjusted by an operator.

[0176] The method can also comprise an additional step of maintaining the device for protecting a crop plot 50. ^.

Claims

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ Claims 1. System for protecting a plot of crops, comprising a protection device (140, 240) adapted to be mounted on an agricultural vehicle (100, 200) comprising at least one tool (120, 220) for treating crops (50), said vehicle comprising advancement means (130, 230) allowing it to advance in a so-called advancement direction (101) within a plot of crops (50), said crops being said to be upstream (51) when they are located upstream of the at least one treatment tool (120, 220) before the passage of the at least one treatment tool (120, 220) over the crops (50) relative to the direction of advancement (101), and said crops being said to be in the process of being treated (52) when they are located at the level of the at least one treatment tool (120, 220) during the passage of the at least one treatment tool (120, 220) over the crops (50);said protection device (140, 240) being characterized in that it is configured to: - measure at least one characteristic quantity of crops being treated (52), in particular the size or position in space of a crop (50); said system further comprising a control unit (110, 210) configured to: - receive a measurement of the at least one characteristic quantity from the protection device (140, 240); - identify a state of said crop being treated (52), said state possibly being a so-called deteriorated state, or a so-called preserved state;- transmit a signal for protecting the crop plot (50) to the agricultural vehicle (100, 200) or to a remote terminal, when the crop being treated (52) is in a deteriorated state.

2. System according to claim 1, in which, to identify the state of said crop being treated (52), the control unit (110, 210) is further configured to: - perform a prediction of the state of said crop being treated (52) by means of a machine learning algorithm previously trained from a set of data representative of the at least one quantity ^; ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ characteristic of the crop being treated (52), the state of said crop being treated (52) being said to be deteriorated if said prediction exceeds a predefined threshold, and the state of said crop being treated (52) being said to be preserved if said prediction does not exceed said predefined threshold.

3. System according to claim 2, wherein said machine learning algorithm is a neural network trained by supervised learning from said set of data representative of the at least one characteristic quantity of the crop being treated (52), in which the data are identified as corresponding to a deteriorated state, or as corresponding to a preserved state. 4.

5. System according to claim 1 wherein, to identify the state of said crop being treated (52), the control unit (110, 210) is further configured to: - compare the measurement of the at least one characteristic quantity of the crop being treated (52) with data representative of at least one characteristic quantity of an upstream crop (51), corresponding to the comparison between the at least one characteristic quantity of said crop before the passage of the at least one treatment tool (120, 220), and the at least one characteristic quantity of said crop during the passage of the at least one treatment tool (120, 220), the state of said crop being treated (52) being said to be deteriorated if said comparison exceeds a predefined threshold, and the state of said crop being treated (52) being said to be preserved if said comparison does not exceed said predefined threshold.System according to claim 4, in which the protection device (140, 240) is further configured to measure at least one characteristic quantity of upstream crops (51), and in which the data representative of at least one characteristic quantity of the upstream crop (51) is a measurement of at least one characteristic quantity of the upstream crop (51).

6. System according to claim 4, in which the data representative of at least one characteristic quantity of the upstream crop (51) is pre-existing data ^. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ stored in a non-volatile storage memory of the control unit (110, 210), or in a remote non-volatile storage memory.

7. System according to any one of the preceding claims, in which the at least one characteristic quantity measured by the protection device (140, 240) is chosen from an image of the crop (50), the width of the crop (50), the height of the crop (50), the angle that the crop (50) forms with the ground, the position of the crop (50), a spacing between the crop (50) and another crop in the plot. 8.

9. System according to any one of the preceding claims, in which the control unit (110, 210) is configured to transmit the crop plot protection signal to the at least one treatment tool (120, 220) of the agricultural vehicle (100, 200), so that the at least one treatment tool (120, 220) of the agricultural vehicle (100, 200) is positioned in a rest position, among a working position and a rest position that it can adopt.

9. System according to any one of the preceding claims, in which the control unit (110, 210) is configured to transmit the crop plot protection signal to the advancement means (130, 230) to cause the agricultural vehicle (100, 200) to stop.

10. System according to any one of the preceding claims, wherein the protection device (140, 240) comprises an image sensor and / or a laser sensor. 11.System according to any one of the preceding claims, comprising a location device (170, 270), in which the control unit (110, 210) is configured to associate the measurement of the at least one characteristic quantity of the crop being treated (52) with a position of the crop being treated estimated by the location device (170, 270). ^. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 12. System according to any one of the preceding claims, wherein the control unit (110, 210) is remote from the agricultural vehicle (100, 200), and wherein the protection device comprises wireless communication means adapted to transmit data from the control device to the control unit (110, 210), and vice versa.

13. An assembly comprising a crop treatment tool (120, 220) and a crop protection system according to any one of claims 1 to 12, wherein the protection device (140, 240) and the control unit (110, 210) of said system are arranged on said treatment tool (120, 220).

14. An agricultural vehicle (100, 200) configured to receive an assembly according to claim 13, said treatment tool (120, 200) being mechanically connected to said agricultural vehicle (100, 200). 15.Autonomous agricultural robot (100), comprising a protection system according to any one of claims 1 to 12.

16. Robot (100) according to claim 15, in which the control unit (110) of the protection system corresponds to a central control unit (110) of said robot (100). 17.Method (300) for protecting a crop plot (50) implemented by a protection system according to any one of claims 1 to 12, comprising the steps of: - (310) Measurement by the protection device (140, 240) of at least one characteristic quantity of a crop being treated (52); - (320) Transmission of said measurement to the control unit (110, 210); - (330) Identification of the state of said crop being treated (52), said state possibly being a so-called deteriorated state, or a so-called preserved state; - (340) Transmission of a signal for protecting the crop plot (50) to the agricultural vehicle or to a remote terminal, when the crop being treated (52) is in a deteriorated state. ^. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 18. Method (300) for protecting a plot of crops (50) according to claim 17, implemented by a protection system according to claim 2 taken in combination with claim 3 or any one of claims 7 to 12, in which the identification step (330) comprises the steps of: - (331) Performing a prediction of the state of said crop being treated (52) by means of a machine learning algorithm previously trained from a set of data representative of the at least one characteristic quantity of the crop being treated (52), the state of said crop being treated (52) being said to be deteriorated if said prediction exceeds a predefined threshold, and the state of said culture being treated (52) being said to be preserved if said prediction does not exceed said predefined threshold. 19.Method (300) for protecting a plot of crops (50) according to claim 17, implemented by a protection system according to claim 4, taken in combination with any one of claims 5 to 12, in which the identification step comprises the steps of: - (333) Association by the control unit (110, 210) of the measurement of a crop being treated from the protection device (140, 240) with data representative of at least one characteristic quantity of an upstream crop (51); - (334) Comparison by the control unit (110, 210) of said measurement with said data representative of at least one characteristic quantity of an upstream culture (51), the state of said culture being treated (52) being said to be deteriorated if said comparison exceeds a predefined threshold, and the state of said culture being treated (52) being said to be preserved if said comparison does not exceed said predefined threshold. 20.Method (300) for protecting a plot of crops (50) according to claim 18 or 19, implemented by a protection system based on claim 11, comprising a preliminary additional step (315) of recording the location of a crop (50) by the location device (170, 270), and the association step (333) by the control unit (10) comprising a first association of a measurement of the at least one characteristic quantity of a crop being treated (52) from the protection device to an estimated position ^. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^ by the location device (170, 270) of the characterized culture (50) and a second association of a measurement from the protection device (140, 240) and data representative of at least one characteristic quantity of an upstream culture (51) associated with the position of the culture being treated (52). ^

Images