METHOD FOR DETERMINING THE MOISTURE OF AGRICULTURAL SOIL

Description

Scope of the invention

[0001] The present invention relates to determining the hydric properties of a soil, in particular their real-time spatial evolution. More specifically, the invention proposes to determine the soil's hydric state by means of a measurement signal representative of the circumferential curvature of the tire. Technological background

[0002] It is indeed useful to know the soil's moisture content at all times, in order to interact with the driver or driver assistance systems, providing them with real-time information since this moisture level influences driving conditions, and potentially allowing for adjustments. In particular, knowing the soil's moisture level allows for adjusting a vehicle's operating conditions. For example, in the presence of very wet and loose soil, tire pressure can be reduced to increase the contact area between the tire and the soil, thus limiting soil compaction and the formation of ruts after the agricultural machinery has passed over it.

[0003] Combined with other parameters of the tire, such as the load carried, and of the soil, such as its mechanical resistance, knowledge of the soil's water content also makes it possible to estimate the state of soil compaction at depth.

[0004] By combining soil moisture status with synchronous geolocation data, it is possible to create a soil moisture map of a plot, potentially enriched with other soil characteristics. Such a map can be useful for determining soil management strategies for the plot, such as implementing drainage or specific irrigation tailored to soil moisture heterogeneity, or deep soil decompaction adjusted based on estimated soil compaction heterogeneity.

[0005] Finally, soil, particularly agricultural soil, is complex and changes depending on the depth of observation. It is characterized by physical quantities such as its moisture content, texture, and structure. Generally, soil moisture, or its water content, is defined by the relative mass of water remaining in the soil once its "gravitational" portion has drained away. This is called mass moisture or gross weight moisture, expressed as a percentage and denoted H. However, the amount of water actually available to the plant, or capillary water, is lower than the mass moisture content because the plant must exert a suction effort to assimilate this water. This is called water potential, expressed in bars. Typically, a plant's water potential is limited to 16 bars; beyond this point, water is not available to the plant. Conversely, a soil at its maximum water retention capacity will require very little suction effort from the plant, on the order of 0.3 bar: the soil is said to be at its "field capacity" moisture content. It is useful to quantify the capillary water available at a given time or location in the field by calculating the ratio between the measured ponderable water (H) and its maximum value at the field capacity. This capillary water, "normalized" relative to the field capacity and also expressed as a percentage, is called HCC. Normalizing to the field capacity allows us to define a moisture value independent of soil texture and representative of the water actually available to the plant. Soil texture corresponds to the proportion of clay, silt, and sand that constitute it. These three components are distinguished by their particle size. Thus, an HCC value of 99% means that the capillary water available to the plant corresponds to 99% of the field capacity, a value very close to the maximum possible.However, this value of 99% for HCC corresponds to a weight moisture H of around 40% for a soil with small grain size and high capillary water retention such as a clay soil, but only to a weight moisture H of around 20% for a soil with large grain size and low capillary water retention, such as a sandy soil.

[0006] Finally, soil structure, which corresponds to the three-dimensional arrangement of soil material that can be observed at various horizons, i.e., at different soil depths, is classically characterized by its apparent density, denoted DA. This parameter can be impacted by tillage, particularly on the surface horizons (a few tens of centimeters), but also by the repeated passes of agricultural machinery in the field.

[0007] The combination of soil moisture, texture, and structure results in a complex material with specific mechanical properties. Knowing the moisture content of a soil in isolation provides information about the soil that then makes it easier to determine other physical parameters of the soil or to identify or evaluate its overall mechanical properties.

[0008] French patent application FR3088249A3, registration number 1860481, describes a method for determining soil firmness based solely on measuring the change in curvature of a tire traveling over the soil. Soil firmness is related to the tire's flattening speed and the length of the contact patch. The soil firmness analysis, performed over a relatively significant depth, reflects the soil's bearing capacity. Other relevant examples are described in FR 3 088 426 A3 and FR 3 071 064 A1.

[0009] However, this method does not allow for the determination of all the physical or mechanical characteristics of the soil that might be of interest for specific applications. In particular, this method does not allow for the determination of the soil's water status, that is, the assessment of the amount of water contained in the soil to a depth of a few tens of centimeters, excluding, for example, water that could potentially be assimilated by the plant.

[0010] The objects of the invention which follow are intended to determine the water status of the soil in real time in order to adapt the conditions of driving the tire as needed, but also to adapt the preventive or curative actions to be carried out on the soil in order to optimize the potential of the soil. Description of the invention

[0011] The invention relates to a method for determining the moisture content of soil on which a tire mounted on a vehicle is rolling, said tire being equipped with a sensor configured to acquire a measurement signal representative of the evolution of the tire's curvature while rolling on the ground, the method comprising the following steps: acquisition by the sensor of a measurement signal representative of the evolution of the curvature of the tire during rolling, determination from the measurement signal of measurement data comprising: a) a first parameter representative of a speed, preferably angular, of flattening of the tire upon contact with the ground during one wheel revolution of the tire, and b) a second parameter representative of a speed, preferably angular, of rounding of the tire upon separation from the ground during one wheel revolution of the tire, and c) a third parameter representative of the variation in the rolling speed of the tire such as for example the difference or ratio between the speed of rounding of the tire and the speed of flattening of the tire,Determination of soil moisture as a function of the first and second parameters using a polynomial relationship linking said soil moisture and the third parameter.

[0012] The process allows for the real-time determination of soil moisture levels across the entire tire, on which a tire mounted on a vehicle travels, without consideration of pressure, load, or vehicle speed. This simple, precise, and reliable method relies solely on a measurement signal representing the evolution of the tire's curvature. By taking angular velocity into account, the process automatically eliminates the influence of the tire's rolling speed.

[0013] Here, the first and second parameters are linked to local effects of the contact patch, that is, the change in curvature at the entry or exit of the contact patch, observed by observing the rate of this change. These are not parameters at the global scale of the contact patch, such as, for example, the length of the contact patch. Because these quantities are local, their sensitivity is much greater than a global quantity at the scale of the contact patch. Consequently, their differentiated observation allows us to trace back to the moisture characteristics of the soil on which the tire travels, which is not necessarily possible, if not impossible, with a global quantity, such as the length of the contact patch, which would naturally average the effect over the entire contact patch.

[0014] This process is advantageously complemented by the following characteristics, taken alone or in any technically possible combination thereof: During rolling, over the course of one wheel revolution, the tire curvature evolves according to a cycle exhibiting: a portion out of contact with the ground, and a portion in contact with the ground. The first parameter is determined from a portion of the measurement signal corresponding to a transition in the tire curvature between the portion out of contact with the ground and the portion in contact with the ground, and the second parameter is determined from a portion of the measurement signal corresponding to a transition in the tire curvature between the portion in contact with the ground and the portion out of contact with the ground. During rolling, over the course of one wheel revolution, the tire curvature evolves according to a cycle exhibiting: - a portion out of contact with the ground characterized on the measurement signal by a stable curvature, - a portion in contact with the ground characterized on the measurement signal by a peak in the variation of contact curvature.- a transition called the input transition between the part not in contact with the ground and the part in contact with the ground, characterized on the measurement signal by a peak in the input curvature variation opposite to the peak in the contact curvature variation, - a transition called the output transition between the part in contact with the ground and the part not in contact with the ground, characterized on the measurement signal by a peak in the output curvature variation opposite to the peak in the contact curvature variation, the first parameter being determined by a slope between the peak in the input curvature variation and the peak in the contact curvature variation, the second parameter being determined by a slope between the peak in the input curvature variation and the peak in the contact curvature variation. The polynomial relation is of order n, n being an integer between 2 and 5, in the form: , F = a 0 + a 1 × ΔKS + … + a n × ΔKS n with F a moisture factor, ΔKS the third parameter, and a 0 to an predetermined fixed coefficients, soil moisture is determined by calculating a moisture factor from the first and second parameters, and comparing said moisture factor to thresholds delimiting soil moisture classes, moisture is included in the group including relative field capacity weight moisture, average weight moisture, moisture is defined over a soil depth X of less than 50 centimeters, preferably less than 40 centimeters, potentially less than 30 centimeters or 20 centimeters, a vehicle localization step during the signal acquisition step delivering at least a two-dimensional vehicle position.

[0015] According to an example not part of the invention, this description also relates to a mapping of the heterogeneity of soil moisture on a surface where a tire mounted on a vehicle is rolling, said tire being equipped with a sensor configured to acquire a measurement signal representative of the evolution of the curvature of the tire during rolling on the surface comprising at least two determinations of soil moisture corresponding to at least two two-dimensional positions of the vehicle obtained by the method according to the invention.

[0016] According to an example not forming part of the invention, this description also relates to a tire comprising a sensor sensitive to the evolution of the tire's curvature, configured to generate a measurement signal representative of the evolution of the tire's curvature when rolling on a surface, comprising an active part and an electronic board, the active part being configured to generate the measurement signal, the electronic board being configured to determine measurement data comprising: a) a first parameter representing a speed, preferably angular, of flattening of the tire during contact with the ground during one wheel revolution of the tire, and b) a second parameter representing a speed, preferably angular, of rounding of the tire during separation from the ground during one wheel revolution of the tire, the sensor being configured to transmit the measurement data outside the tire.

[0017] The invention also relates to a data processing unit configured to receive measurement data derived from a measurement signal representative of the evolution of tire curvature during rolling on a surface, said measurement data comprising: a) a first parameter representing a speed, preferably angular, of flattening of the tire upon contact with the ground during one wheel rotation of the tire, and b) a second parameter representing a speed, preferably angular, of rounding of the tire upon separation from the ground during one wheel rotation of the tire, the data processing unit being configured to define a third parameter representing the variation in the rolling speed of the tire such as for example the difference or ratio between the speed of rounding of the tire and the speed of flattening of the tire and to determine the soil moisture as a function of the first and second parameters using a polynomial relationship linking said soil moisture and the third parameter.

[0018] Preferably the data processing unit is configured to receive at least two-dimensional vehicle position associated with measurement data.

[0019] The invention also relates to a vehicle comprising: at least one tire, at least one sensor sensitive to the evolution of the tire's curvature, configured to generate a measurement signal representative of the evolution of the tire's curvature when rolling on a surface, preferably the sensor is located inside the tire, a data processing unit configured to receive measurement data derived from the measurement signal representative of the evolution of the tire's curvature when rolling on a surface and to determine the soil moisture based on the measurement data, the measurement data comprising: a) a first parameter representative of a speed, preferably angular, of the tire flattening upon contact with the ground during one wheel rotation of the tire, and b) a second parameter representative of a speed, preferably angular, of the tire rounding upon separation from the ground during one wheel rotation of the tire,the vehicle being configured to implement the method according to the invention.

[0020] Preferably, the sensor comprises an active part and an electronic board, the active part being configured to generate the measurement signal, the electronic board being configured to determine the measurement data, and in which the data processing unit is disposed outside the tire.

[0021] According to an example not included in the invention, this description also relates to a computer program product comprising program code instructions for executing the process according to the invention, when said program is executed on a computer. The computer program product may take the form of a computer-readable non-transient medium storing code instructions for executing the process according to the invention, when said computer-readable non-transient medium is read by a computer. Brief description of the drawings

[0022] The invention will be better understood upon reading the following description, given solely by way of non-limiting example and made with reference to the accompanying figures, in which the same reference numbers designate identical parts throughout and in which: there Fig. 1schematically illustrates a tire mounted on a vehicle rim; the Fig. 2 presents an example of a measurement signal recorded by a sensor sensitive to tire curvature as the tire rolls; the Fig. 3 presents a synoptic diagram of the steps in the method for evaluating soil moisture according to possible embodiments of the invention; Fig. 4a and Fig. 4b Each shows an example of the relationship and statistical classification between the two parameters derived from the measurement signal for a vehicle's front tire under different soil moisture conditions; - the Fig. 5 shows an example of soil moisture heterogeneity mapping on a plot obtained from the measurement signal for a vehicle tire; Detailed description of implementation methods

[0023] There Fig. 1Figure 1 illustrates a tire 1 mounted on a rim 2. Such a tire 1 comprises, on the one hand, a crown area 3 constituting a tread with grooves, and on the other hand, sidewalls 4 ending in lower sections. These lower sections generally include a bead and a rim strip to allow the tire 1 to be mounted on the rim 2. The rim 2 is itself connected to the vehicle 9 by an axle (not shown). The tire 1 thus provides the connection between the vehicle 9 and the ground 7.

[0024] A tire is defined as a flexible solid designed to be mounted on the rim 2 of a wheel, generally in the form of a band, to provide contact between the vehicle 9 and the ground 7. It has a tread whose circumferential radius of curvature changes when subjected to stress. The tire 1 is typically made of elastomers (e.g., rubber) and possibly other textile and / or metallic materials. The tire 1 may be airless, for example, with flexible polyurethane spokes supporting the tread. Preferably, however, a tire 1 comprises a flexible casing containing a pressurized gas, typically air. As this is the most common type of tire 1, the following description is non-limiting and refers to such a tire 1 with an internal gas pressure.

[0025] Tire 1 is subjected to a force applied by vehicle 9, via the axle and rim 2, directed towards the ground 7. This force originates from the axle load, resulting from the weight of vehicle 9. Since the rim 2 is rigid, this force acting on tire 1 deforms it when it is in contact with the ground surface 7: the crown 3 under the rim 2 flattens, increasing the contact area 6 of tire 1 with the ground, while the sidewalls 4 bulge. This deformation is more pronounced when the pressure inside the tire is low. The nature of the soil 7 also influences this deformation, and in particular the mechanical state of this soil 7. Indeed, a resistant soil does not deform or deforms very little, while a soft or loose soil deforms under the action of the tire 1, so that the deformation of the tire 1 is less, partly transferred to the soil 7.

[0026] The deformation of tire 1 results in a change in the circumferential curvature of tire 1, that is, the curvature of the apex area 3. When tire 1 rolls, this change in curvature travels around the circumference of tire 1. For a given point of tire 1, the curvature will therefore vary periodically with each wheel rotation.

[0027] The tire 1 is equipped with a sensor 10 configured to acquire a measurement signal representative of the tire's curvature evolution. This sensor 10 is located inside the tire's casing. Preferably, the sensor 10 is positioned against the crown area 3. The sensor 10 can be embedded in the structure of the tire's casing, or it can be attached to the casing and, for example, held in place by an adhesive layer. The sensor 10 has an active part 11 attached to the tire's casing, such that the deformation of the tire 1 causes a corresponding deformation of the active part 11 of the sensor 10, which generates a measurement signal that is a function of the deformation of its active part 11. The measurement signal is therefore representative of the tire's curvature evolution.

[0028] Preferably, the sensor 10 is a piezoelectric sensor, which generates a voltage proportional to the change in deflection. More precisely, the sensor 10 may, for example, include an active part 11 consisting of a piezoelectric layer between two conductive layers. It is also possible for the sensor 10 to be a resistive sensor, whose impedance is proportional to the deflection of the active part 11 of the sensor. An accelerometer can also be used, although its operation is much more complex and requires more extensive signal processing. The sensor 10 can also be adapted to measure other parameters, and in particular pressure. The sensor 10 can be integrated into other electronic equipment installed in the tire 1, such as a TMS (Tire Monitoring System) pressure and / or temperature sensor.

[0029] The sensor 10 also includes an electronic board 12 connected to the active part 11 of the sensor 10 and configured to receive the measurement signal from the active part 11. This electronic board 12 includes at least one processor and memory, and is adapted to process data such as the measurement signal, to determine measurement data from the measurement signal, and to communicate this measurement data. Preferably, the sensor 10 is associated with a wireless transmitter, in particular of the radio frequency type, and for example of the type using Bluetooth Low Energy technology or of the type of low-power device operating in the 433 MHz band (LPD 433), enabling the measurement signal to be relayed to an automated data processing unit, preferably located outside the tire 1, for further processing.The wireless transmitter can be part of the sensor 10, for example as a component of the electronic board 12, or be separate from the sensor 10. For example, an antenna can be provided inside the tire 1. In the case of wireless communication, an external receiver can receive the signals sent by the wireless communication means associated with the sensor 10, and relay them to the automated data processing unit.

[0030] Of course, the sensor 10 may include other elements enabling its proper functioning, and in particular a power supply module, for example consisting of a battery.

[0031] When the tire 1 rolls on the ground, the sensor 10 acquires (step S1) the measurement signal representing the evolution of the tire's circumferential curvature. This measurement signal can be directly related to the curvature (and therefore be a curvature measurement signal), thus tracking its evolution, or it can be indirectly related to the curvature. This is notably the case for a sensor 10 whose active part 11 is a piezoelectric sensor, since the measurement signal then corresponds to the curvature after the signal has passed through a charge amplifier. This type of sensor will be used in the examples below. The measurement signal, generated by the active part 11 of the sensor 10, is then processed by the electronic board 12 to determine measurement data from the signal. The processing of the measurement signal aims to extract the useful information from this signal, which is then used in the subsequent stages of the process.

[0032] There Fig. 2 This shows a schematic example of a measurement signal recorded by a sensor 10 sensitive to the change in curvature of the tire as the tire 1 rolls. The measurement signal is represented here by its voltage (in V), and designated by curvature, as a function of the wheel rotation expressed in degrees.

[0033] During rolling, over the course of one wheel rotation, the curvature of the tire changes according to a cycle that presents: one part out of contact with the ground, one part in contact with the ground.

[0034] The sequence illustrates two passes through the ground contact area of ​​the tire 1, where sensor 10 is located, separated by a non-ground contact area. The non-ground contact area is characterized by a stable curvature, which translates into a stable measurement signal. The ground contact area is characterized in the measurement signal by a peak in contact curvature variation 20, 30. On the Fig. 2 The peaks of contact curvature variation 20, 30 are directed downwards. Indeed, the peaks of contact curvature variation 20, 30 correspond to the flattening of the tire 1 in the contact surface 6.

[0035] The curvature also exhibits a transition, known as the entry transition, between the part not in contact with the ground and the part in contact with the ground. This transition is characterized on the measurement signal by a peak in the entry curvature variation (21, 31) opposite to the peak in the contact curvature variation (20, 30), i.e., upwards in this case. The curvature variation also exhibits a transition, known as the exit transition, between the part in contact with the ground and the part not in contact with the ground. This transition is characterized on the measurement signal by a peak in the exit curvature variation (22, 32) opposite to the peak in the contact curvature variation (i.e., upwards in this case). The entry curvature variation peak (21, 31) and the exit curvature variation peak (22, 32) correspond to the abrupt changes in the radius of curvature of the tire (1) as it enters and exits the contact patch.

[0036] Since the tire rotates, the same cycle repeats, with a stable measurement signal when not in contact with the ground, then a peak in the input curvature variation (21, 31), a peak in the contact curvature variation (20, 30), a peak in the output curvature variation (22, 32), and finally another stable measurement signal when not in contact with the ground. This cycle corresponds to one wheel rotation, therefore 360°, as shown on the diagram. Fig. 2For each cycle, the peak of the exit curvature variation 22, 32 has the significant advantage of being acute and, above all, essentially independent of the ground conditions and the tire 1. Indeed, the peak of the exit curvature variation 22, 32 corresponds to the change in curvature of tire 1 as it exits the contact patch, when the area of ​​tire 1 where sensor 10 is located abruptly changes from the flat state characteristic of the part in contact with the ground to the curved state characteristic of the part out of contact with the ground. On soft ground, as it rolls, tire 1 compacts the ground beneath it, forming a rut, and therefore a fairly firm rut bottom on which tire 1 rests as it exits the contact patch. Furthermore, the forward movement of the vehicle 9 primarily transfers the stresses towards the entrance of the contact patch.Tire 1 exiting the contact area thus exhibits an exit behavior, in terms of curvature, very close to the behavior of a tire 1 on a road.

[0037] Thus, it is easy to identify each cycle corresponding to a wheel rotation by detecting each peak in the output curvature variation 22, 32. It is also possible to identify the cycles with a dedicated device, such as a tachometer. From there, the data can be expressed as a function of the angular degree of each cycle. This makes it possible, in particular, to compare the cycles and their data independently of the vehicle's speed 9. The steps of the process require only one cycle to be implemented and can therefore be implemented for each cycle. However, in order to make the process more robust to potential one-off events (such as the presence of a stone), it is possible to use a combination of several measured cycles, for example, with a moving average.

[0038] The mechanical condition of the soil influences the characteristics of the measurement signal profile. The invention therefore aims to extract parameters from the measurement signal to deduce the mechanical condition of the soil. The method thus comprises determining (step S2), from the measurement signal, measurement data comprising at least a first parameter KSin representing the angular velocity of the tire's flattening during contact with the soil during one wheel rotation of tire 1, and a second parameter KSout representing the angular velocity of the tire's rounding during uncovering from the soil during one wheel rotation of tire 1. The measurement data may include other parameters or values ​​derived from the measurement signal.

[0039] The first KS in and second KS out parameters are determined from a portion of the measurement signal corresponding to a transition in the tire curvature between the part off the ground and the part in contact with the ground. More precisely, the first parameter KS in is determined by a slope between the peak of the input curvature variation 31 and the peak of the contact curvature variation 30. More precisely, the first parameter KS in can correspond to the maximum change (in absolute value) in curvature between the peak of the input curvature variation 31 and the peak of the contact curvature variation 30, i.e., it corresponds to the maximum slope. In the example, since the measurement signal is expressed in volts (V) as a function of angular degrees (°), the first parameter KS in can have units of V / °, i.e., the first derivative of the tire curvature 1.

[0040] The second parameter KS out is determined by a slope between the peak of the output curvature variation 32 and the peak of the contact curvature variation 30. More precisely, the second parameter KS out can correspond to the maximum curvature variation between the peak of the output curvature variation 32 and the peak of the contact curvature variation 30, i.e., it corresponds to the maximum slope. In the example, since the measurement signal is expressed in volts (V) as a function of angular degrees (°), the first parameter KS out can have the unit V / °, i.e., the first derivative of the curvature of the tire (1).

[0041] The parameters KS in and KS out can be approximated in several ways. For example, the KS parameters can correspond to the maximum (in the sense of absolute value) of the derivative of the measurement signal between the peak of the input curvature variation 31 or output 32 and the peak of the contact curvature variation 30. The derivative is estimated from the difference between two successive (or nearby) measurement points, obviously taking into account their angular separation. Since the slope of KS in in the example is decreasing, the same will be true for KS out, with an increasing slope. This maximum in the sense of absolute value corresponds to a minimum of the derivative of the measurement signal between the peak of the input curvature variation 31 and the peak of the contact curvature variation 30.It is also possible, instead of searching for an extreme value of the derivative, to choose fixed measurement points, such as those equidistant from the peaks of the input curvature variation peak 31 and the contact curvature variation peak 30, and calculate the derivative from these points. Alternatively, one can choose measurement points corresponding to a measurement signal value, such as the zero crossing in the illustrated case. More complex approaches, such as the Savitzky-Golay algorithm, are also possible. However, choosing a relatively low sampling frequency, typically less than or equal to 500 Hz, and preferably less than or equal to 400 Hz, like the 300 Hz in the example, smooths the measurement signal and allows for less computationally intensive approaches, such as those presented above.

[0042] Taken individually, the KS parameters can depend on the load, pressure, and / or speed. However, taking into account both the first parameter KS in and the second parameter KS out makes it possible to determine quantities representative of the soil water status, such as the relative gravimetric moisture at field capacity over a rather shallow depth X of the soil, HCC 0-X generated by the passage of tire 1, from these parameters alone, without knowing the load, pressure, and speed of tire 1 on the soil.

[0043] Preferably, the electronic board 12 of the sensor 10 determines, from the measurement signal, the measurement data comprising the first parameter KS in and the second parameter KS out. This measurement data is then transmitted by the sensor 10 to a data processing unit 15, which implements the remainder of the process. This data processing unit 15 is preferably located outside the tire 1, for example, in the vehicle 9, but the processing unit 15 can also be located remotely from the vehicle 9, and the data transmission may then involve intermediate transmission means. The transmission of measurement data between the sensor 10 and the data processing unit 15 is then wireless.The data processing unit 15 typically includes a processor and memory, and is adapted to receive and process measurement data during the implementation of the subsequent process for determining the soil water status.

[0044] It is possible to transmit the measurement signal to the processing unit 15 for the subsequent steps of the process. However, determining the measurement data using the sensor 10 and transmitting only this measurement data to the data processing unit 15 has the advantage of reducing the amount of data transmitted between the sensor 10 and the data processing unit 15. Since data transmission consumes a significant amount of energy, transmitting the measurement data rather than the measurement signal limits the power consumption of the sensor 10, whose power supply options in the pneumatic system 1 are limited.

[0045] Furthermore, it is advantageous not to use the electronic board 12 of the sensor 10 for implementing the rest of the process, but rather to use the data processing unit 15 to process the measurement data. This limits the calculations performed by the electronic board 12 of the sensor 10, thus saving energy and memory on the electronic board 12. In addition, it is easier to modify the implementation of the rest of the process on an easily accessible data processing unit 15, rather than on the sensor 10 inside the tire 1.

[0046] Once the processing unit 15 has received the measurement data, the processing unit 15 can determine the mechanical property of the soil as a function of the first parameter KS in and the second parameter KS in contained in the measurement data, which vary according to the surface mechanical resistance of the soil, as shown below.

[0047] THE Fig. 4a and Fig. 4b are illustrations of the determination of the soil water status which is global at the tire scale.

[0048] In these examples, the measurement data is derived from a signal acquired by a piezoelectric sensor located in the front tire of an offset trailer pulled by an agricultural tractor as the tractor passes. This allows the trailer to carry the necessary measuring equipment to determine the pressure and load applied to the tire in a stable manner. The offset of the trailer, and in particular the measuring tire relative to the tractor tires, ensures that the soil is compacted not by the tractor's passage, but only by the passage of the measuring tire.

[0049] The tractor is driving on soil with three different soil moisture conditions: A wet soil type, denoted HH, meaning moist both on the surface and at depth, corresponds to winter conditions in the Northern Hemisphere. A moist soil type, denoted HS, meaning moist on the surface but dry at depth, corresponds to autumn conditions in the same countries. A dry soil type, denoted SH or SS, meaning dry on the surface, corresponds to spring (moist at depth) or summer (dry at depth) conditions in the countries mentioned above.

[0050] The characterization of soil moisture was carried out by analyzing soil core samples taken at various depths (X). Weight or mass moisture content was determined by comparing the raw weight of the core with its dry weight, obtained after drying the soil core. Soil coring was performed simultaneously throughout the day with pneumatic tube measurements on a track adjacent to the pneumatic tube measurement track, which had the same physical characteristics of soil, i.e., the same physical properties in terms of texture, structure, and moisture content.

[0051] The results of Fig. 4a and Fig. 4b They compile data obtained on three soil types with different textures (sandy-clay, silty-clay, and silty). For each of these textures, the soils were prepared under three different structural conditions:

[0052] A compacted or pressed-down state called "WO" corresponds to the state of the soil left after harvest, which is compacted by the repeated passage of a loaded and inflated tire without any soil work after these passes.

[0053] A soil condition that has been disturbed or loosened by harrowing compacted soil to a depth of 10 or 30 centimeters, respectively named "W10" and "W30".

[0054] Finally, measurements using sensor 10 on the tire were taken under various load and pressure conditions applied to the test tire, representative of field use for an agricultural tire. Measurements were taken at varying speeds below a maximum speed of 20 km / h.

[0055] There Fig. 4aThis shows a quadratic relationship between the soil moisture content at a depth of 40 centimeters, denoted H0-40, and the variation in tire winding speed ΔKS. The dashed curves represent the 90% confidence interval. Here, the variation in tire winding speed ΔKS is obtained as the difference between the tire's winding speed (KSout) and the absolute value of the tire's flattening speed (KSin). This relationship remains valid regardless of the soil type, and in particular its structure, texture, and moisture content.

[0056] It is possible to define soil moisture or water status classes, specifically three different levels, which allows for differentiation of soil water status using laboratory-type characterization over a certain depth X. In the case of the Fig. 4aThe weight-based moisture content (H0-40) corresponds to the average moisture content of core samples taken at depths of 10, 20, 30, and 40 centimeters from various soil sections within the plots. However, it is entirely possible to incorporate these moisture measurements at shallower soil depths, but the results are generally less accurate.

[0057] As an illustrative and non-limiting example, as it is highly dependent on soil texture, the following classes can be used: [Table 1] H 0-40 (%) <= 15 15 - 20 > 20 Humidity classes dry humid wet ΔKS >= 1.2 1.2 - 0.6 < 0.6

[0058] A surprising level of correlation was identified between this laboratory measurement and the variation in tire rolling speed ΔKS obtained by measuring sensor 10 mounted on the tire. Thus, it is easy to determine the soil moisture class to a surface depth of 40 centimeters before the tire passes over the ground, simply by measuring the KS in and KS out parameters, regardless of soil texture, moisture, and structure, or the tire's operating conditions such as inflation pressure, applied load, and rolling speed.

[0059] Thus, denoting F as the soil moisture factor, here weight-based moisture H over a depth of 40 centimeters, associated with the variation in tire winding speed, and f as a function corresponding to the quadratic relationship and relating to the parameter ΔKS, we can write: F = f ΔKS

[0060] More precisely, the quadratic relation can be of the form: F = a 0 + a 1 ∗ ΔKS + a 2 * ΔKS 2 + … + a n * ΔKS n with F the soil moisture factor over a depth of 40 centimeters, ΔKS the variation in tire winding speed, as the difference between the tire winding speed KS out and the tire flattening speed KS in, and a 0 to an are fixed real coefficients previously determined.

[0061] Fixed coefficients a0 to an are preferably chosen to maximize the discrimination of soil moisture classes. A one-dimensional discriminant analysis can be used, for example. This discriminant analysis aims to maximize the differences between the centroids of each soil moisture class, while minimizing intra-class dispersion.

[0062] As an illustrative and non-limiting example, the soil moisture factor by weight over a depth of 40 centimeters (H 0-40) can be determined as follows: H 0 − 40 % = 23.63 − 4.98 x ΔKS + 3.54 x Δ KS 2 − 3.28 x Δ KS 3

[0063] There Fig. 4b This shows a quadratic relationship between the relative moisture content of the soil at a depth of 40 centimeters, denoted HCC 0-40, and the variation in tire winding speed ΔKS. The dashed curves represent the 90% confidence interval. Here, the variation in tire winding speed ΔKS is obtained as the difference between the absolute value of the tire flattening speed KS in and the tire rounding speed KS out. This relationship remains valid regardless of the soil type, and in particular its structure, texture, and moisture content, or the tire's operating conditions.

[0064] A surprising level of correlation was identified between this laboratory measurement and the variation in tire winding speed ΔKS obtained by measuring the sensor 10 mounted on the tire. Thus, it is easy to determine the relative soil moisture content at field capacity to a surface depth of 40 centimeters simply by measuring the parameters KS in and KS out, regardless of soil texture, moisture, and structure, and the tire's operating conditions such as inflation pressure, applied load, and rolling speed.

[0065] Thus, denoting F as the soil moisture factor, here relative moisture to the HCC field capacity over a depth of 40 centimeters, associated with the variation of the tire winding speed, and f as a function corresponding to the quadratic relationship and relating to the parameter ΔKS, we can write: F = f ΔKS

[0066] More precisely, the quadratic relation can be of the form: F = a 0 + a 1 ∗ ΔKS + a 2 * ΔKS 2 + … + a n * ΔKS n with F the soil moisture factor over a depth of 40 centimeters, ΔKS the variation in tire winding speed, as the difference between the tire winding speed KS out and the tire flattening speed KS in, and a 0 to an are fixed real coefficients previously determined.

[0067] Fixed coefficients a0 to an are preferably chosen to maximize the discrimination of soil moisture classes. A one-dimensional discriminant analysis can be used, for example. This discriminant analysis aims to maximize the differences between the centroids of each soil moisture class, while minimizing intra-class dispersion.

[0068] As an illustrative and non-limiting example, the relative moisture factor for soil field capacity at a depth of 40 centimeters (HCC 0-40) can be determined as follows: HCC 0 − 40 % = 90.30 + 8.50 x ΔKS − 24.07 x Δ KS 2

[0069] There Fig. 5 This is a map of soil moisture classes (H) assessed from measurements taken during the passage of a measuring tire over the soil at the scale of an agricultural plot. This plot, with a silty texture, was initially heavily irrigated across its entire width using an irrigation boom to obtain a "HH" type moisture status, i.e., winter moisture conditions.

[0070] The soil moisture content (H) is assessed by measuring the tire flattening speed (KS in) and the tire rounding speed (KS out) at each wheel rotation. Then, the change in tire rolling speed (ΔKS) is calculated by subtracting the tire flattening speed (KS in) from the rounding speed (KS out).

[0071] The columns in the map represent the longitudinal furrows of the tire equipped with sensor 10 across the agricultural plot under homogeneous tire conditions. However, from one furrow to the next, tire usage conditions can change, for example, by increasing the load applied to the tire or modifying the tire inflation pressure. Here, we assume that the entire plot has the same physical properties in terms of texture and initial structure. However, some furrows have undergone specific preparation, potentially resulting in a different furrow structure compared to other furrows. Some furrows within the same plot exhibit a WO, W10, or W30 type structure.

[0072] For the same furrow, that is, the same column for a given plot, the various lines correspond to linear units of the furrow corresponding to one wheel rotation of the measuring tire, or approximately 5 meters. For each wheel rotation, the flattening speed KS in and the rolling speed KS out of the tire are measured. From these values, a variation in the rolling speed ΔKS of the tire is deduced. The average soil moisture content H is then estimated at a depth of 40 centimeters for each wheel rotation.

[0073] The water content of the linear unit of each furrow and the water content of the furrow itself are then classified according to the three previously defined classes: wet, humid, and dry, in order to obtain the map of the Fig. 5

[0074] A good correlation was observed between the imposed water status on each furrow and the water status obtained by the method in terms of statistical classes, regardless of the physical properties of the analyzed soil. In particular, a strong heterogeneity was observed in the center of the plot, which was much drier than at its edges. This heterogeneity was attributed in part to problems with the irrigation boom nozzles and to the heterogeneous topography of the plot.

[0075] Such mapping therefore shows an obvious agricultural potential by analyzing the potential water heterogeneity of the soil of an agricultural plot, which makes it possible to adapt a priori the conditions of use of the tire to each zone of the plot considered, but also to adapt the irrigation of the plot to the spatial heterogeneity of the water state of the plot.

Claims

1. Method for determining the moisture of a ground on which a tyre (1) mounted on a vehicle (9) is running, said tyre (1) being fitted with a sensor (10) configured to acquire a measurement signal representative of the change in the curvature of the tyre as it runs over a ground (7), the method comprising the following steps: - acquiring (S1), by way of the sensor (10), a measurement signal representative of the change in the curvature of the tyre (1) while it is running, - determining (S2), from the measurement signal, measurement data comprising: a) a first parameter (KSin) representative of a rate, preferably angular rate, at which the tyre (1) flattens on contact with the ground (7) over the course of one revolution of the wheel bearing the tyre, and b) a second parameter (KSout) representative of a rate, preferably angular rate, at which the tyre (1) regains its shape on becoming separated from the ground (7) over the course of one revolution of the wheel bearing the tyre, c) a third parameter (ΔKS) representative of the variation in the recovery rate of the tyre, such as for example the difference or the ratio between the rate at which the tyre regains its shape (KSout) and the rate at which the tyre (1) flattens (KSin); - determining (S4) the moisture of the ground as a function of the first parameter (KSin) and the second parameter (KSout) using a polynomial relationship connecting said moisture of the ground and the third parameter (ΔKS).

2. Method according to Claim 1, wherein, during running, over the course of one revolution of the wheel, the curvature of the tyre (1) changes according to a cycle exhibiting: - a part where there is no contact with the ground, - a part where there is contact with the ground, wherein the first parameter (KSin) is determined from part of the measurement signal corresponding to a transition in the curvature of the tyre (1) between the part where there is no contact with the ground and the part where there is contact with the ground, and the second parameter (KSout) is determined from part of the measurement signal corresponding to a transition in the curvature of the tyre (1) between the part where there is contact with the ground and the part where there is no contact with the ground.

3. Method according to Claim 1 or 2, wherein, during running, over the course of one revolution of the wheel, the curvature of the tyre (1) changes according to a cycle exhibiting: - a part where there is no contact with the ground, characterized in the measurement signal by a stable curvature, - a part where there is contact with the ground, characterized in the measurement signal by a contact curvature variation spike (20, 30), - a transition referred to as the coming-into-contact transition between the part where there is no contact with the ground and the part where there is contact with the ground, characterized in the measurement signal by a coming-into-contact curvature variation spike (21, 22) that is the opposite of the contact curvature variation spike (20, 30), - a transition referred to as the coming-out-of-contact transition between the part where there is contact with the ground and the part where there is no contact with the ground, characterized in the measurement signal by a coming-out-of-contact curvature variation spike (22, 32) that is the opposite of the contact curvature variation spike (20, 30), the first parameter (KSin) being determined by a gradient between the coming-into-contact curvature variation spike (21) and the contact curvature variation spike (20), the second parameter (KSout) being determined by a gradient between the coming-into-contact curvature variation spike (31) and the contact curvature variation spike (30).

4. Method according to one of the preceding claims, wherein the polynomial relationship is of the order n, n being an integer of between 2 and 5, taking the following form: F = a 0 + a 1 × ΔKS + … + a n × ΔKS n where F is a moisture factor, ΔKS is the third parameter, and a0 to an are predetermined fixed coefficients.

5. Method according to one of the preceding claims, wherein the moisture of the ground is determined by calculating a moisture factor (F) from the first parameter (KSin), and the second parameter (KSout), and by comparing said moisture factor to thresholds delimiting moisture categories for the ground.

6. Method according to one of the preceding claims, wherein the moisture is included in the group comprising the moisture content by weight relative to the field capacity (HCC), the average moisture content by weight (H).

7. Method according to one of the preceding claims, wherein the moisture is defined over a ground depth X of less than 50 centimetres, preferentially less than 40 centimetres, potentially less than 30 centimetres or 20 centimetres.

8. Method according to one of the preceding claims, comprising a step of locating the vehicle (9) during the step of acquiring the signal that provides an at least two-dimensional position (Ploc) of the vehicle (9).

9. Data processing unit (15) configured to receive measurement data derived from a measurement signal representative of the change in the curvature of the tyre as it runs over a ground, said measurement data comprising: a) a first parameter (KSin) representative of a rate, preferably angular rate, at which the tyre (1) flattens on contact with the ground (7) over the course of one revolution of the wheel bearing the tyre, and b) a second parameter (KSout) representative of a rate, preferably angular rate, at which the tyre (1) regains its shape on becoming separated from the ground (7) during one revolution of the wheel bearing the tyre, the data processing unit (15) being configured to define a third parameter (ΔKS) representative of the variation in the recovery rate of the tyre, such as for example the difference or the ratio between the rate at which the tyre regains its shape (KSout) and the rate at which the tyre (1) flattens (KSin) and to determine the moisture of the ground as a function of the first parameter (KSin) and the second parameter (KSout) using a polynomial relationship connecting said moisture of the ground and the third parameter (ΔKS), and the data processing unit (15) is preferably configured to receive that at least two-dimensional position (Ploc) of the vehicle (9) that is associated with the measurement data.

10. Vehicle (9) comprising: - at least one tyre (1), - at least one sensor (10) sensitive to the change in the curvature of the tyre and configured to generate a measurement signal representative of the change in the curvature of the tyre as it runs over a ground, - a data processing unit (15) according to Claim 9, the vehicle being configured to implement the method according to any one of Claims 1 to 7.

11. Vehicle (9) according to the preceding claim, wherein the sensor (10) is disposed inside the tyre (1).

12. Vehicle (9) according to the preceding claim, wherein the sensor (10) comprises an active part (11) and an electronic circuit board (12), the active part (11) being configured to generate the measurement signal, the electronic circuit board (12) being configured to determine the measurement data, and wherein the data processing unit (15) is disposed outside the tyre.