WO2022049552A1

WO2022049552A1 - Method and system for tracking, monitoring and predicting the health of a plant heritage

Info

Publication number: WO2022049552A1
Application number: PCT/IB2021/058090
Authority: WO
Inventors: Stéphane KREBS
Original assignee: Krebs Paysagistes Sa
Priority date: 2020-09-07
Filing date: 2021-09-06
Publication date: 2022-03-10
Also published as: WO2022049552A8; EP4211457A1; CH717816A1

Abstract

Described are a method and a system for tracking, monitoring and predicting the health of a plant heritage (PV) comprising at least one specimen of a plant (A.1, A.2, A.3,...). Each specimen (A.i) constituting the plant heritage (PV) is equipped with a set of sensors (10.i, 10.1, 10.2, 10.3,...) for measuring a plurality of parameters representative of the state of health of the plant (A.i) or potentially affecting the health of the plant (A.i). The measurement data ({DATAA.i}) generated by each set of sensors (10.i) is communicated to a data collection center (100). A characterization of the state of health of each specimen (A.i) is performed using the measurement data ({DATA_PV}) collected by the data collection center (100), in order to build a predictive model of the health of each specimen (A.i). An assessment of the state of health of each specimen (A.i) is made on the basis of the predictive model, and an alert is generated if the predictive model of the health of the specimen is indicative of a danger or other risk related to a worsening health of the specimen (A.i).

Description

METHOD AND SYSTEM FOR MONITORING, MONITORING AND PREDICTING THE HEALTH OF A PLANT HERITAGE

TECHNICAL AREA

The present invention generally relates to a method and a system for monitoring, monitoring and predicting the health of plant heritage comprising one or more specimens of ligneous or herbaceous plants, in particular plant heritage in an urban or rural environment, such as, in particular, tree heritage, orchards, vines, agricultural crops, market gardening, etc.

TECHNOLOGICAL BACKGROUND

Patent publication No. EP 3 421 988 A1, the content of which is incorporated by reference into the present application, describes a device and a method for continuously measuring the "staticity" of trees. By "staticity" we mean the susceptibility of each tree to undergoing movement or - in other words - the degree of stability of the tree. According to the teaching of this European patent application, the degree of stability of the tree is measured and evaluated by means of a triple-axis motion sensor, such as an accelerometer. There is also provision for the additional use of an anemometer to optionally measure the force of the wind to which the tree is subjected. This approach, although appropriate in order to anticipate the possible fall of a tree, does not however make it possible to obtain a sufficiently reliable, representative and complete indication of the state of health of the tree, the evaluation carried out being limited to a measurement of the movement of the tree, if necessary combined with a measurement of an environmental condition likely to affect this movement, namely the force of the wind.

Patent publication No. CN 103903400 A, the content of which is also incorporated by reference in the present application, describes a system for measuring the degree of inclination of a tree likely to present a risk of falling, which system is arranged in order to generate an alert when the degree of inclination of the tree is indicative of a danger. Like the patent publication No. EP 3 421 988 A1, it is planned to use a triple acceleration sensor axes, which sensor is attached to the trunk of the tree. This approach suffers from essentially the same limitations as the approach disclosed in patent publication No. EP 3 421 988 A1 discussed above.

European patent No. EP 1 793 225 B1, the content of which is incorporated by reference in the present application, describes a device and a method for the non-destructive inspection of a tree operating by acoustic tomography. This solution makes it possible to inspect the internal state of a tree in a non-destructive way, which is particularly desirable in order to determine and visualize the internal state of the trunk of the tree, in particular in order to diagnose the hardness, the existence of cavities and rot, and/or the humidity content of the tree trunk, parameters which allow an assessment of the risk of the tree falling. However, this approach is cumbersome and complex to implement in practice.

Utility Model No. CN 208969721 U, the contents of which are incorporated by reference into the present application, describes an ancient tree management system based on the Internet of Things ("Internet of Things" or "loT") namely the use of connected objects. It is essentially a question of allowing the competent service of a city to facilitate the tasks of managing the tree heritage of the city by concentrating and listing the information relating to each tree and its state of health, as noted on the field, in a centralized server, which centralized server can be interrogated or updated remotely by an operator of the city service. It is thus in particular provided to associate with each tree an identification tag carrying a two-dimensional code, which identification tag can be read by means of a mobile terminal and thus facilitate the identification of the tree and the consultation or updating the information relating thereto in the centralized server. It is also envisaged to allow, where appropriate, third-party users, such as tourists, to access descriptive information relating to the ancient trees concerned by using the identification tag in the same way in order to facilitate the identification of each tree and access to said descriptive information. However, there is no question of carrying out any real-time characterization or evaluation of the state of health of each tree. Patent publication No. EP 3 531 367 A1 describes a system and a method for predicting the appearance of a disease affecting one or more trees, in this case a pine wilt disease. Each tree is equipped with a device for predicting the appearance of the disease comprising in particular at least one measurement sensor placed on or near the tree, and a communication unit configured to transmit the measurement data to a server via a data transmission network. Several variants of prediction devices are envisaged combining various types of measurement sensors, in particular a first humidity measurement sensor placed on the tree, a second humidity measurement sensor placed in the ground where the tree is located, in particular at the level of the roots, a temperature/humidity sensor placed in the environment of the tree, a sensor for measuring the sugar content relative to the tree, and a sensor for measuring rosin (or "rosin") ).

Patent publication No. WO 2019/124657 A1 describes a method as well as a system for predicting a disease affecting trees based essentially on a humidity measurement.

Patent publication No. WO 99/44050 A1 describes an ultrasonic tomography measuring apparatus comprising a belt configured to be arranged around the trunk of a tree, which belt comprises a plurality of ultrasonic sensors capable of measuring the internal structure of the tree.

There therefore remains a need to propose a solution which makes it possible to ensure better monitoring and better monitoring of the health of plant heritage, in particular heritage consisting of ornamental trees, fruit trees, forest trees, shrubs , shrubs, vines, or any other specimen of woody or herbaceous plants.

DISCLOSURE OF THE INVENTION

A general object of the present invention is to propose a method and a system making it possible to ensure the monitoring and surveillance, with the possibility of prediction, of the health of plant heritages which remedies the problems and limitations of the known solutions. More specifically, an object of the present invention is to propose such a solution which makes it possible to ensure such monitoring and such monitoring in real or almost real time, thereby ensuring permanent or almost permanent monitoring and monitoring of the heritage in question. .

Yet another object of the present invention is to propose such a solution which allows a more reliable, representative and complete characterization and evaluation of the state of health of each specimen constituting a plant heritage.

The present invention meets these aims by proposing a method whose characteristics are listed in claim 1, namely a method for monitoring, monitoring and predicting the health of a plant heritage comprising at least one specimen of a plant. woody or herbaceous, the method comprising the following steps: a) equipping each specimen constituting the plant heritage by means of a set of sensors making it possible to measure several parameters representative of the state of health of the specimen or likely to affect the state of health of the specimen; b) communication of the measurement data generated by each set of sensors to a data collection centre; c) characterization of the state of health of each specimen using the measurement data collected by the data collection center in order to build a predictive model of the health of each specimen; d) assessment of the health status of each specimen based on the specimen health predictive model; and e) generating an alert if the specimen health predictive model is indicative of a hazard or other risk associated with a decline in specimen health status.

The invention differs notably from known solutions in that a predictive model of the health of each specimen is constructed on the basis of the measurement data collected. This predictive model makes it possible to characterize not only the current state of health of the specimen, but also its susceptibility to lead to the occurrence of a danger (for example a fall) or any other risk linked to a deterioration in the state of health of the specimen (for example a disease, a parasite, or other pathology affecting the state of health of the plant in question). It is therefore on the basis of the predictive model of the health of the specimen, which takes into account all the parameters representative of the state of health of the specimen or likely to affect the state of health of the specimen, that the assessment of the state of health of each specimen is carried out.

Preferably, the step d) of evaluating the state of health of each specimen includes the comparison of the predictive model of the health of the specimen with at least one reference model characterizing a good state of health of the specimen or a poor state of specimen health. In other words, the predictive model of the health of each specimen is compared to at least one reference model (or even several) which characterizes a good state of health or, on the contrary, a bad state of health of the specimen. It must therefore be understood that both the predictive model and the reference model are constructed on the basis of a set of data relating to the various parameters representative of the state of health of the specimen or likely to affect the state of health. of the specimen. According to this preferred approach, it is thus possible to identify and characterize the signature of a breast specimen, respectively that of a specimen at risk or suffering from a disease, a parasite, or other pathology affecting the state of health of the specimen. .

Even more preferably, step c) of characterizing the state of health of each specimen is carried out taking into account the typology specific to each specimen and the characteristics specific to the typology of each specimen. Indeed, the state of health of any plant specimen cannot necessarily be characterized by the same set of parameters or measurement data. Each type of plant can thus lead to the need to build a predictive model of the state of health that is different from one type to another.

In this last context, we will therefore advantageously proceed to the definition of a basic predictive model specific to each typology of specimen (among a plurality of typologies) and we will ensure that the predictive model of the health of each specimen concerned is constructed on the basis of the model basic predictor corresponding to the typology of the specimen concerned and the measurement data of the specimen concerned.

Preferably, each set of sensors is configured to carry out several measurements from among the set of following measurements, namely a measurement of water stress undergone by the specimen, a measurement of sap flow circulating in the trunk of the specimen, a measurement of structural integrity of the trunk of the specimen, a measurement of the temperature of the specimen and/or its environment, a measurement by imaging the specimen in a determined wavelength or range of wavelengths, a measurement of the movement of the specimen, a measurement of mechanical stresses to which the specimen is subjected, a measurement of environmental conditions undergone by the specimen, a measurement of electrical conductivity of the specimen, and a measurement of the surrounding gaseous components. However, this is not necessarily an exhaustive list, other additional measures may be considered if necessary. It will be understood that the greater the diversity of the measurements taken, the more representative the predictive model of the state of health of the specimen concerned will be. Although it is possible to install only one sensor, it is preferable to proceed with the installation of several sensors.

In general, it will be understood that the set of sensors considered must make it possible to optimize an overall management of the specimens considered in correlation with their history, for the benefit of their health and/or the safety of people and the environment. where each specimen is implanted, particularly in terms of robustness, dangerousness, statics and vitality, by measuring (directly or indirectly) and analyzing multiple physical and physiological factors, exogenous and/or endogenous, such as:

- the robustness of the specimen;

- the stability of the trunk and/or the root base;

- the thickness of the stems and trunks;

- elongation of stems and trunks;

- the solidity of the trunk;

- the resistance of the structural branches;

- root absorption; - the circulation of raw and elaborated sap;

- photosynthesis ;

- evapotranspiration;

- electrical conductivity;

- the temperature of the plant;

- Room temperature ;

- imaging at different wavelengths (visible, infrared, etc.);

- the presence of surrounding gaseous components, in particular oxygen, carbon dioxide, nitrogen, etc. ;

- breathing;

- sap analysis;

- stored reserves;

- the water available in the soil; and

- weather conditions and forecasts.

Step a) of equipping each specimen may in particular include: equipping a plurality of dedicated sensors arranged on or near the specimen concerned; the equipment of a short-distance receiver device configured to receive the measurement data from one or more sensors among said plurality of dedicated sensors by means of a wireless link; and the equipment of a transmitter device coupled to said short-distance receiver device, and if necessary directly to one or more sensors among said plurality of dedicated sensors, which transmitter device is configured to communicate the measurement data collected locally to said central data collection.

This solution offers increased flexibility in that the various sensors required can be distributed on and/or in the vicinity of the specimen concerned, without being constrained by the need to arrange all of the sensors in a single and unique device attached to the specimen. All or part of the sensors can thus communicate the measurement data to the associated short-distance receiver device by means of a wireless link in order to allow their retransmission by the transmitter device to the control center. data gathering. It is alternatively or additionally conceivable that all or part of the sensors directly transmit the measurement data to the data collection center, in which case each sensor concerned will be equipped with transmission means suitable for this purpose.

According to a preferred variant, step b) of communicating measurement data is carried out in real time or periodically, and step c) of characterizing the state of health of each specimen includes updating in real time or the specimen health predictive model periodically based on the collected measurement data. In this context, steps d) and e) are preferably repeated as soon as the predictive model of the health of the specimen concerned is updated.

The present invention also meets the aforementioned aims by proposing a system whose characteristics are listed in claim 7, namely a system for monitoring, monitoring and predicting the health of a plant heritage for the implementation of the method according to the invention, comprising: at least one set of sensors intended to equip each specimen constituting the plant heritage; a data collection center configured to collect the measurement data generated by each set of sensors and compile a set of collected measurement data relating to the plant heritage; and a transmitter device associated with each set of sensors and configured to ensure the communication of the measurement data generated by the set of sensors to said data collection center.

Steps c) to e) of the method according to the invention can be implemented by the data collection center or by means of one or more ad hoc data processing systems able to access the measurement data collected by the central data collection.

As a preference, each set of sensors comprises a plurality of dedicated sensors arranged on or near the specimen concerned and the system further comprises a short-distance receiver device associated with each set of sensors and configured to receive the measurement data from one or several sensors among said plurality of dedicated sensors by means of a wireless link. In this context, each transmitter device is then configured to ensure the communication of measurement data collected locally to said data collection center.

The present invention also relates to a measuring device capable of being used in particular within the framework of the system according to the invention. This measuring device comprises an extensible belt configured so as to be able to be arranged around the trunk of a specimen of a ligneous plant, said extensible belt being coupled by a wired or non-wired connection to at least one sensor and/or bearing at least a sensor, which sensor is capable of measuring a parameter representative of the state of health of the specimen or capable of affecting the state of health of the specimen. The measuring apparatus comprises alternatively or additionally a measuring device which can be inserted at least partially inside the trunk of a specimen of a woody plant, which measuring device comprises at least one sensor capable of measuring a parameter representative of the health of the specimen or likely to affect the health of the specimen. The measuring equipment comprises alternatively or additionally one or more sensors placed in the vicinity of the specimen, on its foliage, its branches, or its plant ramifications.

Other aspects of the invention are set out in the remainder of this description.

BRIEF DESCRIPTION OF THE DRAWINGS

The characteristics and advantages of the present invention will appear more clearly on reading the following detailed description of embodiments of the invention, which are presented solely by way of non-limiting examples and are illustrated by the appended drawings where:

- Figure 1 shows an overall schematic view of a system applied to the monitoring and surveillance of the health of a plant heritage, in this case, and more specifically, a tree heritage comprising a plurality of trees, according to an embodiment of the invention; - Figure 2 shows a schematic view of an isolated tree among the set of trees of the tree heritage of Figure 1 as well as a set of associated sensors making it possible to measure several parameters representative of the state of health of the tree or likely to affect the state of health of the tree;

- Figure 3 is a schematic diagram summarizing the various steps of the method implemented by means of the system of Figure 1 and each set of sensors, such as the set of sensors of Figure 2;

- Figure 4 is a schematic representation of an embodiment of step d) of the method of Figure 3 according to a preferred embodiment;

- Figure 5 is a schematic diagram of an embodiment of step c) of the method of Figure 3 according to a preferred embodiment;

- Figure 6 is a schematic representation of a measuring device according to one embodiment of the invention, which measuring device comprises in particular an extensible belt arranged around the trunk of a tree as well as a measuring device at least partially inserted into the trunk of the tree; and

- Figure 7 is a schematic representation of the measuring device of Figure 6 partially inserted into the trunk of the tree.

MODES FOR CARRYING OUT THE INVENTION

The present invention will be described with reference to various preferred embodiments as illustrated in particular by Figures 1 to 7.

Figure 1 is a schematic overview of a system for tracking, monitoring and predicting the health of a plant heritage according to an embodiment of the present invention, which plant heritage is designated globally by the reference PV . This plant heritage PV is constituted here by way of illustrative example of a tree heritage comprising a plurality of trees A.1, A.2, A.3, etc. (each tree of the tree heritage PV also being designated indifferently by the reference Ai) distributed over the whole of a determined territory, for example the territory of a city. In the remainder of this description, any reference to the tree heritage PV must be understood as an illustrative reference to an example of plant heritage (in this case a ligneous plant heritage) within the meaning of the invention. Similarly, any reference to each tree Ai should be understood as an illustrative reference to an example of a plant specimen (in this case a ligneous plant) within the meaning of the invention.

The present invention is particularly applicable to the management of plant heritage (in particular but not exclusively ligneous) in an urban environment, but is not necessarily limited to such an application. The invention is thus perfectly applicable to the management of a plant heritage in a rural environment, for example a forest or an arboreal exploitation such as an orchard comprising one or more species of fruit trees, or else vines, shrubs, crops agriculture or market gardening, this list not being exhaustive. The particular nature of each plant is not limiting and it can indifferently be any ornamental or fruit tree, established naturally or not in the environment, vines, shrubs, shrubs, or any other specimen of plants.

The present invention is also applicable where appropriate to the management of herbaceous plant heritage, namely for any application in the field of agriculture and market gardening.

It will be understood that, for the implementation of the invention, it will not be necessary for all the plant specimens appearing in a given territory to be subject to monitoring and surveillance. It will suffice to select a representative sample of specimens for which it is desired to ensure the monitoring and surveillance according to the invention. By "plant heritage", it must therefore be understood that it is a question of a heritage made up of one or more specimens of plants (woody or herbaceous) which one specifically wishes to ensure the monitoring and surveillance, which heritage can be representative of a wider range of specimens located in the targeted territory. The heritage considered may, where appropriate, consist of one and the same species of plants or, conversely, of a mix of different species. According to the embodiment of the invention, each tree Ai of the tree heritage PV considered is equipped with a set of sensors 10.i (sets which are designated by the references 10.1 to 10.9 in Figure 1), each set of sensors 10.i being arranged to allow the measurement of various parameters representative of the state of health of the associated tree Ai or likely to affect the state of health of the associated tree Ai As will be seen later, each set of sensors 10.i is configured to carry out several measurements indicative of or liable to impact the state of health of the associated tree Ai

Each set of sensors 10.i thus allows the collection of a corresponding set of measurement data, designated {DATAA}, specific to each tree A.i. These measurement data {DATAA} are collected, for each tree Ai of the tree heritage PV, and communicated to a data collection center 100 which compiles these data in the form of a set of data {DATAPV} relating to the tree heritage considered PV. This data {DATAPV} may be processed, in accordance with the provisions mentioned below, directly at the level of the data collection center 100 or be made available to one or more ad hoc data processing systems 100*. It will also be understood that the measurement data collected by each sensor can, if necessary, be processed or preprocessed at the level of the sensor concerned or at the level of the set of sensors 10.i before being transmitted to the data collection center. 100 data for an overall review and analysis of the health of the affected Ai tree. In this respect, optimization methods such as “edge computing” (or “periphery computing”) or “fog computing” (or “geodistributed computing”) can be perfectly implemented.

The collection of measurement data not only makes it possible to characterize and evaluate the state of health of each specimen Ai, but also to build a history of the evolution of the state of health of each specimen Ai, according to seasons, environmental conditions, the age of the specimen, its own history, etc. Figure 2 is a schematic view of an isolated tree Ai among the set of trees of the tree heritage PV and of the set of associated sensors 10.i making it possible to measure various parameters representative of the state of health of the tree Ai or likely to affect the state of health of the tree Ai More specifically, the set of sensors 10.i comprises a plurality of dedicated sensors 10A-10F arranged on or near the tree Ai

In the illustration of Figure 2, the sensor 10A is placed close to the rooting of the tree A.i in order to perform a direct measurement at the level of the roots of the tree A.i. It may in particular be a sensor partially or completely buried at the level of the roots of the tree A.i in order to measure the degree of humidity of the soil at the level of the rooting of the tree A.i. Such a measurement makes it possible in particular to derive a measurement of the water stress undergone by the tree A.i.

The sensors 10B, 10D and 10E are arranged on or near the plant, respectively the trunk T of the tree A.i in order to carry out various direct or indirect measurements at the level of the trunk T of the tree A.i. The sensor 10B can thus be a sensor making it possible to carry out a measurement (invasive or non-invasive) of the flow of sap circulating in the trunk T of the tree A.i. The sensor 10D can itself be a sensor making it possible to measure the structural integrity of the trunk of the tree A.i, for example by means of a probe inserted into the trunk T of the tree. The sensor 10E can be a sensor making it possible to measure the movement of the shaft A.i, such as an accelerometer, or a measurement of the mechanical stresses to which the shaft A.i is subjected, such as a strain gauge. One can also imagine equipping the shaft A.i with a sensor making it possible to perform an electrical conductivity measurement.

The sensor 10C is placed close to the tree Ai in order to carry out an additional measurement, such as a thermal measurement of the tree Ai (for example a measurement of the average temperature of the foliage F) or a measurement by imaging of all or part of the Ai tree by means of a camera. This may be a measurement by imaging in the visible (by means of a color camera or not) or in the invisible (for example by means of an infrared camera or any other camera sensitive to a wavelength or range of wavelengths outside the visible range) making it possible to derive a qualitative measurement of all or part of the tree and its evolution over time. Imaging techniques can thus be implemented in order to determine the occurrence of a noticeable change in the appearance and/or posture of the tree Ai

The sensor 10F is for its part arranged close to the tree Ai in order to proceed to a measurement of environmental conditions undergone by the tree, such as the force of the wind, the sunshine, the ambient temperature, the level of precipitation, the possible lightning strikes, degree of environmental pollution, etc.

A thermal or environmental measurement also makes it possible to detect the possible occurrence of a fire affecting the specimen concerned and/or its surrounding environment, which is of interest for the purposes of detecting, for example, the occurrence of forest or brush fires.

It is obviously understood that other sensors could still be provided in addition to or replacement of the sensors 10A to 10F mentioned above.

At least a portion of the sensors may suitably be mounted on an expandable belt placed around the trunk of the specimen, resistant to weather, theft and vandalism. Some sensors can also take the form of sensors inserted directly into the trunk of the tree, if necessary connected to the aforementioned stretch belt.

Figure 6 is a schematic illustration of a measuring device suitable for use according to one embodiment of the invention. This measuring apparatus comprises in particular an extensible belt C arranged around the trunk T of a tree as well as a measuring device 15 at least partially inserted into the trunk T of the tree. The expandable belt C can carry a sensor 10b (or even several) and/or be coupled by a wired or wireless connection to one or more remote sensors, such as a sensor 10c mounted on a branch B or a sensor 10d coupled to the foliage F of the tree. The expandable belt C can also carry other components necessary for the implementation of the required functionalities, such as the short-distance receiver device 20 and the transmitter device 50 discussed below, and/or any other component, for example an energy source 30 (battery or other), an audible and/or visual alert transmitter 40, etc. The measuring device 15 can for example be inserted into the trunk T close to the expandable belt C and be connected to the latter by means of a wired (as illustrated) or non-wired connection.

The measuring device 15 is illustrated schematically in Figure 7. It can advantageously take the form of a probe or essentially hollow screw 11 insertable at least partially into the trunk T and accommodating within it one or more sensors 10a. This probe 11 can be closed at its outer end by a head 11a closing off the probe 11 in order to protect the sensors 10a from any environmental interference and to allow their maintenance. The sensor or sensors 10a can be functionally connected to the stretch belt C by means of a wired connection 15a in order to ensure the transmission of measurement data. The transmission of the measurement data coming from the sensor(s) 10a could alternatively be carried out by means of a wireless link.

Multiple static measurements by sensors (direct or indirect) can be carried out, including in particular the robustness of the tree, the stability of the root base, the stability of the trunk (for example by monitoring the inclination of the trunk), the solidity of the trunk at critical points (for example by on-board tomography), the resistance of the structural branches (stays), the tension of the stays, etc.

It will be understood that the measurements made in relation to each specimen of the plant heritage can be endogenous, namely relating to the specimen A.i itself, or exogenous, namely relating to the environment in which the specimen A.i is found. Thus, one can consider in addition the taking into account of exogenous data related for example to meteorological conditions or predictions affecting or likely to affect the environment in which the specimen A.i.

Multiple physiological measurements by sensors (direct or indirect) can also be carried out, including in particular root absorption, circulation of raw and elaborated sap, photosynthesis, evapotranspiration, respiration, sap analysis, stored reserves, available water in the soil, etc.

Although not illustrated, it will be understood that each specimen Ai can still be equipped with a location system (for example GPS) making it possible to determine the location of the specimen concerned, information likely to also be communicated to the central collection of data 100 in addition to the measurement data.

In addition, each specimen may also be equipped with a danger warning system capable of emitting an audible and/or visual alert in the event of the occurrence of a danger or other risk identified in relation to the state of health of the specimen. specimen Ai and/or in relation to the arrival of a threat, a storm or any other danger that could affect the population (for example a terrorist risk).

Finally, it is still possible to envisage equipping each specimen with a machine-readable identifier (such as a QR code or any other suitable coding), or even an information system dedicated to amateurs and/or professionals in order to provide descriptive information specific to each specimen (species, age, date of establishment, history, specificities, etc.). A smartphone application can thus be provided in order to read a QR code associated with each specimen and provide access to said descriptive information. Any other device or information system is possible.

Figure 2 also illustrates the presence of a transmitter device 50 associated here with the aforementioned set of sensors 10.i and configured to transmit the measurement data {DATAA} collected locally by the set of sensors 10.i to the data collection center 100 previously mentioned.

Data transmission can be done in various ways. The transmitter device 50 can thus be a connected device (or “loT” device) capable of directly transmitting the measurement data {DATAA}, preferably in a secure manner, over a local or extended data network, such as the Internet. Many cities already offer wireless Internet coverage, freely accessible or not, which would ensure the transmission of measurement data {DATAAJ} directly on the Internet, it being understood that the data collection center 100 would then preferably take the form of a hardware or dematerialized server hosting the tools and services necessary for collecting data from all of the tree heritage PV concerned. The measurement data { DATAPV } of all the PV tree heritage concerned could thus be automatically downloaded into a dedicated cloud hosted on a server accessible via the Internet.

In the absence of adequate wireless Internet coverage, the transmission of the data {DATAA} can be operated through a cellular telephone network or any other radio transmission network, in which case the transmitter device 50 will then be configured to format and transmit the measurement data { DATAA } on the radio frequency band allocated or made available for this purpose.

It will also be understood that the transmitter device 50 can be directly connected to a wired communication network, whether it is an optical fiber network or any other suitable physical channel making it possible to ensure the wired transmission of data.

Independently of the particular nature of the transmitter device 50, it is preferable to adopt an architecture further comprising a short-distance receiver device 20 configured to receive the measurement data from one or more sensors among the plurality of dedicated sensors 10A-10F through a wireless link. In this regard, Figure 2 illustrates by way of example that the sensors 10B, 10D and 10E are directly connected to the transmitter device 50, thus allowing the direct transmission of measurement data to the transmitter device 50, while the sensors 10A, 10C and 10F are sensors that do not have a direct link with the transmitter device 50. Each of these sensors 10A, 10C, 10F is in wireless communication with the short-distance receiver device 20 in order to transmit the measurement data to it. It will be understood that this wireless link can be ensured in any suitable manner, for example, but not exclusively by means of a local wireless network link (or "WLAN") established according to a Wi-Fi protocol based on the standard IEEE 802.11. Although FIG. 2 shows the transmitter device 50 and the short-distance receiver device 20 as two distinct devices, it will be understood that the functions of each device can be perfectly integrated into one and the same transmitter-receiver device.

It is also possible to envisage that one or more of the dedicated sensors 10A-10F directly communicate the measurement data which are specific to him/her/them to the data collection unit 100 (for example by means of a direct connection to an extended data network such as the Internet, in which case we can speak of connected sensors or loT sensors), the functionalities of the transmitter device 50 then being integrated into each sensor concerned. The use of a transmitter device 50 common to each set of sensors 10.i nevertheless appears preferable in the sense that such use makes it possible to transmit the measurement data {DATAA} specific to each tree Ai in the form of a same data packet, if necessary accompanied by identification data specific to the tree Ai concerned, including for example an indication of the type of tree Ai (ornamental or fruit, essence, species, etc.) and/or other information such as the location of the Ai tree, its age, its location, etc.

For the implementation of the invention, preference will be given to the use of low-cost, proven sensors, allowing a multiplication of the number and a diversification of the measurements carried out on each tree, as well as a large-scale deployment, without however compromise the investment necessary for such an implementation. However, this does not exclude the use of any other sensor likely to allow direct or indirect measurement of a parameter representative of the state of health of the specimen in question or likely to affect the state of health of the specimen in question.

Figure 3 is a schematic diagram summarizing the various steps of the method implemented by means of the system of Figure 1 and of each set of sensors, such as the set of sensors 10. i of Figure 2. The method according to invention is essentially divided into five main steps, namely: a) the equipment (or apparatus) of each specimen Ai constituting the plant heritage concerned PV by means of a set of sensors 10.i making it possible to measure several parameters representative of the state of health of the specimen Ai or likely to affect the state of health of the specimen Ai (as discussed above); b) the communication of the measurement data {DATAAJ} generated by each set of sensors 10.i to the data collection center 100 (according to one or the other of the methods already discussed); c) the characterization of the state of health of each specimen Ai by means of the measurement data {DATAPV} collected by the data collection center 100 in order to construct a predictive model, designated [HEALTHAJ], of the health of each specimen Have ; d) evaluating the health status of each specimen Ai based on the predictive model [HEALTHAJ] of the health of the specimen Ai; and e) generating an alert if the predictive model [HEALTHA ] of the health of specimen Ai is indicative of a hazard or other risk related to a deterioration in the state of health of specimen Ai

The innovative approach is thus based on the characterization of the state of health of each specimen in a global and “multisensory” way, rather than on the basis of a single measurement parameter or a very small set of measurement parameters. It is in fact a question of determining and measuring the parameters which characterize a specimen in good health versus a subject presenting a reduced vitality, even a risk for the population or its immediate environment. The characterization of the state of health of each specimen which is carried out according to the invention must therefore be understood as being based on a multi-parameter analysis making it possible to characterize the health of the specimen concerned from multiple angles and according to multiple variables. We thus commonly speak of an approach based on the "fusion of sensor data" (or "sensor fusion" in Anglo-Saxon terminology) an approach which is usually applied in the machinery industry or in aeronautics in order to predict the occurrence of faults or problems and/or anticipate maintenance operations.

Knowing the great diversity of species of trees or other plants, situations and exposures of subjects in town or rural areas, etc., the effectiveness of a static method cannot be guaranteed. It is therefore rather a question of describing and characterizing a normal envelope of variation of the various relevant indicators for each specimen, before determining the measurement points located outside of a "normal" variability envelope and which may constitute indicators of health and alert triggers.

The various measurement variables specific to each specimen or type of specimen may in particular be described by Gaussians or other statistical parameters characterizing a "normal" variation envelope through an initial calibration or learning phase, which may require implementation of machine learning techniques and algorithms.

According to a preferred variant, step b) of communicating measurement data is carried out in real time or periodically, and step c) of characterizing the state of health of each specimen includes updating in real time or periodicity of the predictive model [HEALTHA ] of the health of the specimen Ai according to the collected measurement data. In this context, steps d) and e) are preferably repeated as soon as the predictive model of the health of the specimen concerned is updated.

Figure 4 is a schematic representation of an embodiment of step d) of the method of Figure 3 according to a preferred embodiment of the invention. As schematically illustrated in Figure 4, rather than evaluating the state of health of each specimen Ai on the basis of one or more parameters measured by comparing it or them to determined reference thresholds, it is preferable to proceed to a comparison of the predictive model [HEALTHAJ] of the health of specimen Ai with at least one reference model, designated [HEALTHREF]. This reference model [HEALTHREF] can be a reference model characterizing a good state of health of the specimen Ai or a reference model characterizing, conversely, a poor state of health of the specimen Ai (for example due to a disease, a parasite , or other pathology). This reference model [HEALTHREF] can be constructed on the basis of an earlier predictive model [HEALTHA ] of the health of the specimen Ai, judged to be representative of a good state of health, in which case the comparison leads to determining any variations in the state of health of the specimen Ai over time. Alternatively, the reference model [HEALTHREF] can be built on the basis of a standard predictive model of the health of a specimen of the same nature elaborated on the basis of a statistical model taking into account a large population of specimens of same nature. In fact, a comparison can be made between the updated [HEALTHAJ] predictive model of the specimen concerned and a multitude of reference models in order to determine whether any discrepancies are revealed in a statistically decisive and overriding manner or it is simply an artifact that can be ignored.

Figure 5 is a schematic diagram of an embodiment of step c) of the method of Figure 3 according to a preferred embodiment of the invention. According to this preferred variant embodiment, the step of characterizing the state of health of each specimen Ai, namely the development of the predictive model [HEALTHAJ], is carried out taking into account the typology, designated TYPEAJ, specific to each specimen Ai and characteristics specific to the TYPEA typology of each specimen Ai Indeed, the state of health of any plant cannot necessarily be characterized by the same set of parameters or measurement data. Each type of plant can thus lead to the need to build a predictive model of the state of health that is different from one type to another.

More specifically, as schematically illustrated in FIG. 5, the typology corresponding to the specimen concerned Ai is first selected from a predetermined set of typologies designated TYPEi, TYPE2, TYPE3, etc. On the basis of this selection, one then proceeds to the selection of a basic predictive model, designated [HEALTHBASELINE], which is predetermined for the selected typology TYPEAJ. Once this operation has been carried out, the proper characterization of the state of health of the specimen concerned Ai can be carried out by means of the measurement data {DATAAJ} of the specimen concerned Ai In other words, a “typical” predictive model is previously determined for each typology of specimen concerned, and the predictive model of each specimen is built from this "typical" predictive model and feeds using the current measurement data { DATAA } of the specimen concerned Ai

It will generally be understood that various modifications and/or improvements obvious to those skilled in the art can be made to the embodiments described in the present description without departing from the scope of the invention defined by the appended claims.

In particular, although Figures 1 and 2 show a tree heritage PV made up of several trees A.i, it will again be understood that the invention is applicable to any ligneous, even herbaceous plant heritage, comprising one or more plant specimens.

In addition, the plant heritage can be made up of a set of specimens belonging to the same plant species or, on the contrary, be made up of a mix of plant species.

Claims

1. A process for monitoring, monitoring and predicting the health of a plant heritage (PV) comprising at least one specimen of a plant (A.1, A.2, A.3, etc.), the method comprising the following steps: a) equipping each specimen (Ai) constituting the plant heritage (PV) by means of a set of sensors (10.i, 10.1, 10.2, 10.3, etc.) making it possible to measure several parameters representative of the state of health of the specimen (Ai) or likely to affect the state of health of the specimen (Ai); b) communication of the measurement data ({ DATAAJ }) generated by each set of sensors (10.i) to a data collection center (100); c) characterization of the state of health of each specimen (Ai) by means of the measurement data ({ DATAPV }) collected by the data collection center (100) in order to construct a predictive model ([HEALTHAJ]) of the health of each specimen (Ai); d) evaluation of the state of health of each specimen (A.i) based on the predictive model ([HEALTHA ]) of the health of the specimen (A.i); and e) generating an alert if the predictive model ([HEALTHA ]) of the health of the specimen (A.i) is indicative of a danger or other risk related to a deterioration in the state of health of the specimen (A.i).

2. The method according to claim 1, characterized in that step d) of evaluating the state of health of each specimen (Ai) includes the comparison of the predictive model ([HEALTHAJ]) of the health of the specimen ( Ai) with at least one reference model ([HEALTHREF]) characterizing a good state of health of the specimen (Ai) or a poor state of health of the specimen (Ai).

3. The method according to claim 1 or 2, characterized in that step c) of characterizing the state of health of each specimen (Ai) is carried out taking into account the typology (TYPEAJ) specific to each specimen ( Ai) and characteristics specific to the typology (TYPEAJ) of each specimen (Ai).

23

4. The method according to claim 3, characterized in that a basic predictive model ([HEALTHBASELINE]) is predetermined for each typology of specimen (TYPEAJ) among a plurality of typologies of specimen (TYPEi, TYPE ₂ , TYPE _3I . .. ), the predictive model ([HEALTHAJ]) of the health of the specimen (Ai) being built on the basis of the basic predictive model ([H EALTHBASELINE]) corresponding to the typology (TYPEA ) of the specimen concerned (Ai) and measurement data ({ DATAAJ }) of the specimen concerned (Ai).

5. The method according to any one of the preceding claims, characterized in that each set of sensors (10.i) is configured to carry out several measurements selected from the set of the following measurements, namely a measurement of water stress undergone by the specimen (Ai), a measurement of sap flow circulating in the trunk (T) of the specimen (Ai), a measurement of structural integrity of the trunk (T) of the specimen (Ai), a measurement of the temperature of the specimen (Ai) and/or of its environment, a measurement by imaging of the specimen (Ai) in a determined wavelength or range of wavelengths, a measurement of movement of the specimen (Ai), a measurement of mechanical stresses to which the specimen (Ai), a measurement of environmental conditions undergone by the specimen (Ai), a measurement of electrical conductivity of the specimen (Ai), and a measurement of the surrounding gaseous components are submitted.

6. The method according to any one of the preceding claims, characterized in that step a) of equipping each specimen (Ai) includes: equipping a plurality of dedicated sensors (10A-10F) arranged on or near the specimen concerned (Ai); the equipment of a short-distance receiver device (20) configured to receive measurement data from one or more sensors (10A, 10C, 10F) among said plurality of dedicated sensors (10A-10F) by means of a wireless link; and the equipment of a transmitter device (50) coupled to said short-distance receiver device (20), and if necessary directly to one or more sensors (10B, 10D, 10E) among said plurality of dedicated sensors (10A-10F) , which transmitter device (50) is configured to communicate the locally collected measurement data ({ DATAAJ }) to said data collection center (100).

7. The method according to any one of the preceding claims, characterized in that step b) of communicating the measurement data ({DATAA}) is carried out in real time or periodically, in that step c) of characterization of the health status of each specimen (Ai) includes the real-time or periodic updating of the predictive model ([HEALTHAJ]) of the health of the specimen (Ai) according to the collected measurement data ({ DATAPV } ).

8. The method according to claim 7, characterized in that steps d) and e) are repeated as soon as the predictive model ([HEALTHAJ]) of the health of the specimen concerned (A.i) is updated.

9. A system for tracking, monitoring and predicting the health of a plant heritage (PV) for implementing the method according to any one of the preceding claims, comprising: at least one set of sensors (10. i, 10.1, 10.2, 10.3, etc.) intended to equip each specimen (Ai) constituting the plant heritage (PV); a data collection center (100) configured to collect the measurement data ({ DATAAJ }) generated by each set of sensors (10. i) and to compile a set of collected measurement data ({ DATAPV }) relating to the plant heritage (PV); and a transmitter device (50) associated with each set of sensors (10.i) and configured to ensure the communication of the measurement data ({DATAAJ}) generated by the set of sensors (10.i) to said collection center of data (100).

10. The system according to claim 9, characterized in that steps c) to e) of the method are implemented by the data collection center (100) or by means of one or more data processing systems ad hoc (100*) that can access the measurement data ({ DATAPV }) collected by the data collection center (100).

11. The system according to claim 9 or 10, characterized in that each set of sensors (10.i) comprises a plurality of dedicated sensors (10A-10F) arranged on or near the specimen concerned (Ai), and in that that the system further comprises a short-distance receiver device (20) associated with each set of sensors (10.i) and configured to receive measurement data from one or more sensors (10A, 10C, 10F) among said plurality dedicated sensors (10A-10F) by means of a wireless link, each transmitter device (50) being configured to ensure the communication of locally collected measurement data ({ DATAA }) to said data collection center (100) .

12. A measuring device adapted to be used in particular within the framework of the system according to any one of claims 9 to 11, characterized in that said measuring device comprises an extensible belt (C) configured so as to be able to be arranged around of the trunk (T) of a specimen (Ai) of a ligneous plant, said extensible belt (C) being coupled by a wired or wireless connection to a sensor (10a-10d) and/or carrying at least one sensor ( 10a-10d), which sensor (10a-10d) is capable of measuring a parameter representative of the state of health of the specimen (Ai) or capable of affecting the state of health of the specimen (Ai).

13. A measuring device adapted to be used in particular within the framework of the system according to any one of claims 9 to 11, characterized in that said measuring device comprises a measuring device (15)

26 insertable at least partially inside the trunk (T) of a specimen (Ai) of a ligneous plant, which measuring device (15) comprises at least one sensor (10a) capable of measuring a parameter representative of the state of health of the specimen (Ai) or likely to affect the state of health of the specimen (Ai).

14. A measuring device suitable for use in particular in the context of the system according to any one of claims 9 to 11, characterized in that said measuring device comprises one or more sensors (10c, 10d) arranged (s) near the specimen (Ai), on its foliage (F), its branches (B), or its plant ramifications.

27