Field of the Invention

The present invention relates to the field of indoor plant growing systems. More particularly, the invention relates to an autonomous plant growing system for reproducing a given plant having a given plant quality without human intervention.

Background of the Invention

In order to meet the needs of consumers, plant growers are interested in marketing plants with certain favorable characteristics. These favorable characteristics are not only dependent upon the genetic properties of the plant, but also on the way the plant is grown and on the time period during which the plant is harvested. Indoor plant growers are able to maximize their profits when the plants are grown in a large farm to produce a correspondingly large yield and when the man hours needed to grow the plants are minimized.

Typical favorable characteristics (hereinafter referred to as "quality") include level of sweetness, size, taste, color, and growth time for fruits and vegetables. For medicinal plants, the quality may be in terms of providing efficacy at a given dosage. For example, a medicinal cannabis plant may be grown for desired levels of TFIC or CBD.

It is an object of the present invention to provide an autonomous plant growing system for reproducing plants with a predetermined quality.

It is an additional object of the present invention to provide a plant growing system for automatically identifying when some plants being grown are foreseen to be lacking the predetermined quality and for automatically correcting the identified deficiencies.

Other objects and advantages of the invention will become apparent as the description proceeds.

Summary of the Invention

An autonomous plant growing system for automatically reproducing a given plant type with desired quality comprises a plant growth unit for growing a given plant type in an indoor controlled environment, said plant growth unit comprising one or more sensors for acquiring data representative of plant growth conditions within an interior thereof, a plurality of electromechanical devices responsive to said acquired data including one or more imaging devices, and a control unit in data communication with each of the sensors and electromechanical devices, wherein said control unit is configured to automatically activate the plurality of electromechanical devices in accordance with an optimal plant growth protocol, acquire real-time visual data related to the given plant type that has been captured by the one or more imaging devices, and command operation of one or more of the electromechanical devices in accordance with a corresponding calculated corrective action if the real-time acquired visual data is found to significantly deviate from expected values, to ensure compliance with the optimal plant growth protocol following performance of each of the corresponding corrective actions.

The plant growing system preferably further comprises a computerized device in data communication with said control unit, said computerized device comprising a database in which is stored data representative of the optimal plant growth protocol that defines a process for reproducing the given plant type with desired quality, and an analysis module for analyzing plant derived data, wherein the control unit is configured to receive data representative of the optimal plant growth protocol from the computerized device, activate, in response, the plurality of electromechanical devices in accordance with the optimal plant growth protocol, and transmit a signal representative of the acquired visual data to the computerized device, wherein the computerized device is operable to determine when the acquired visual data significantly deviates from the expected values, wherein the computerized device is additionally operable to transmit to the control unit a signal indicative of the corresponding calculated corrective action that is needed to be performed following determination that the acquired visual data significantly deviates from the expected values, and the control unit, in response, is operable to activate the one or more electromechanical devices by one or more corrected device-specific activation signals to ensure compliance with the optimal plant growth protocol following performance of the corresponding calculated corrective action.

In one aspect, the analysis module is configured with a plant-specific model for helping to predict how the given plant type will be influenced by a change in value of a plant growth parameter.

Brief Description of the Drawings

In the drawings:

Fig. 1 is a front view of an embodiment of a plant growth unit, shown when a closure thereof is opened; Fig. 2 is a top view of the plant growth unit of Fig. 1, shown when an upper cover thereof is removed;

Fig. 3 is a rear view of the plant growth unit of Fig. 1;

Fig. 4 is a rear view of the plant growth unit of Fig. 1, shown when a rear panel thereof is removed;

Fig. 5 is a schematic illustration of an embodiment of a reproducible plant growing system;

Fig. 6 is a perspective view of apparatus for supporting distributed plant growth units;

Fig. 7 is a flow chart for optimizing the plant growth protocol;

Fig. 8 is a flow chart for generating a plant-specific model; and

Fig. 9 is a flow chart for automatically initiating a recovery process within a plant growth unit.

Detailed Description of the Invention

In the reproducible plant growing system, a plurality of substantially closed plant growth units, which may be distributed, are provided, in each of which plants are grown in an indoor controlled environment in terms of air temperature and humidity, light intensity, wavelength and direction, and irrigation solution including nutrient additives.

The system is also able to automatically sample, monitor and control each parameter that is conducive to optimizing the growth of plants having a predetermined quality, using closed loop feedback in real-time at all growing stages. The controlled environment is adapted for the growth of a specific plant or strain with a predetermined quality and in accordance with a desired yield, e.g. growth rate vs. quality. The system uses data collection capabilities with high-quality imaging devices and various sensors, computer vision and machine learning capabilities to generate a model serving as a basis of comparison for the real-time growth conditions of each plant or strain.

Figs. 1-4 schematically illustrate an exemplary plant growth unit 10, which comprises body 5, e.g. rectilinear, defining an interior volume 6 in which one or more plants are aeroponically or hydroponically grown in plant support structure 8 and which houses electronic and electromechanical control equipment. A front closure 4, which may be pivotally connected to body 5, is openable to access interior volume 6. The closing of front closure 4 primarily the entire day provides a substantially disease-free environment within interior volume 6. The control equipment may include at least one angularly adjustable lighting assembly 11 that is pivotable about a horizontal axis to ensure illumination of the entire plant while growing without causing shadowing effects, a conduit circuit 13 through which irrigation solution circulates, one or more irrigation control members 14 such as a pump and a sprayer for directing irrigation solution to each plant being grown, and a fertilizer injector 17 for injecting fertilizer from a nutrition supply source into conduit circuit 13 to supply the plant being grown with needed minerals. Air conditioning equipment 19 which may also include one or more compressors, fans and ducts for suitably circulating temperature-controlled air after being introduced through inlet 23 throughout interior volume 6 until being discharged by outlet 21 is also provided. One or more sensors 19 are deployed within interior volume 6 in order to acquire data representative of plant growth conditions, such as air temperature, light parameters, humidity level, EC and PH levels within the irrigation solution. The plant growth conditions may be divided into a first group that is indicative of environmental parameters and into a second group that is indicative of plant uptake parameters.

A control unit 26, preferably implemented as a printed circuit board, is in data communication with each sensor 19, and is configured to automatically control each lighting assembly 11, irrigation control member 14, fertilizer injector 17 and air conditioning equipment 19. Control unit 26 is also in data communication with one or more imaging devices 28 adapted to acquire visual data regarding each plant being grown, and particularly regarding the leaves of the plant being grown. If the real-time acquired visual data significantly deviates from expected values, control unit 26 automatically controls operation of one or more types of electromechanical control equipment, such as controlling a fertilization process and balancing the pH of the irrigation solution, by means of wired or wireless signals.

The wireless signals may be provided by the Near Field Communication (NFC) protocol or by another protocol that enable electronic components to establish radio communication with each other. Control unit 26 may communicate with a server or any other external computer device via a suitable data network such as the cellular network, generally by means of remote communication equipment and a dedicated application.

An important aspect of the system is the ability to visually monitor the plant being grown in such a way so as to accurately determine its instantaneous growth traits. Through suitable modeling, an association may be made between the instantaneous growth traits and the ability or inability of the monitored plant to produce the predetermined quality.

Some growth traits relate to aspect traits, including the height and width of the plant, as well as the branching pattern and branching density, the quantity of leaves, flowers, and fruits, and the shape uniformity of the leaves. Other growth traits relate to the size and the color of the leaves, the height of the flowers within the plant, and the clustering of the flowers.

These growth traits are able to be accurately determined when a high-resolution digital imaging device is employed, for example one having a resolution of at least 1920 ^c 1080. The imaging device may be stationary or may be displaced in order to capture different regions of the plant or of the plant growth unit.

Fig. 5 schematically illustrates one embodiment of an autonomous plant growing system, generally indicated by numeral 30. System 30 comprises a plurality of distributed plant growth units (PGUs) 40, a plurality of control units (CUs) 26, generally each of which deployed at a corresponding plant growth unit 40, a user computerized device 32, e.g. a landline or mobile device such as a smartphone, for setting an initial user-selected plant growth protocol, a server 34 in data communication with each of the control units 26 via a suitable data network, and a database (DB) 39 in which data derived from server 34 is storable. Each of the plant growth units 40 has a unique identifier.

Each of the plant growth units 40 may be identical to a plant growth unit 10 shown in Fig. 1, and may be physically and geographically distributed from other plant growth units, such as by a distance of at least 0.5 km, or even by a distance of at least 100 km, while providing a large variation in the plant derived data, to assist in generating a model adapted to predict plant behavior. Alternatively, each of the plant growth units 40 may be deployed within a common frame 45 shown in Fig. 6, which defines vertically or horizontally spaced plant growth units.

Server 34 is configured with an analysis module 37, which is adapted to analyze acquired plant derived data, for example by means of machine learning, and to determine whether it is reflective of the predetermined quality. Analysis module 37 may be configured with a dedicated machine learning algorithm that facilitates classifying acquired data and updating boundary conditions of each classification upon determining that they significantly deviates from sampled values. Server 34 may be interconnected with one or more other servers to store the plant derived data according to a cloud-based content management system, to function as a repository for the big data characteristic of all plant derived data associated with system 30.

After data D related to the user-selected plant growth protocol is transmitted by a suitable signal via data network 33 from user device 32 to each of control units 26, or, alternatively, to server 34 and the server then transmits the data to each of the control units 26, or to the control units 26 and server 34 in parallel, each of the control units 26 operates in accordance with the selected plant growth protocol without human intervention. That is, a control unit 26 transmits a device specific activation signal A, which is indicative of the value of various operation-sensitive parameters, to the corresponding electromechanical devices controlled thereby, generally at predetermined times, in order to ensure or monitor a desired growth pattern defined by the selected plant growth protocol. A non-exhaustive list of operation-sensitive parameters to be controlled includes light intensity, wavelength and propagation direction to be generated by a lighting assembly, air temperature, humidity and carbon dioxide concentration to be generated by air conditioning equipment, volume and concentration of fertilizer to be discharged by the fertilizer injector, and dosage of each irrigation constituent to be discharged by irrigation control members.

An activation signal A is also transmitted to an imaging device in order to capture an image of the plant being grown so that its instantaneous growth traits may be determined. In response, the control unit 26 transmits a signal V indicative of the real-time acquired visual data to server 34. Server 34 is accordingly able to analyze the received data with analysis module 37 in order to suitably classify the acquired visual data, for example according to a classification that is reflective of plant diseases, and then to command the corresponding control unit 26 to adjust the value of various operation-sensitive parameters so that the plant disease will be effectively treated. Server 34 is also configured to optimize the plant growth protocol and generate a suitable plant-specific model.

Fig. 7 illustrates a flow chart for optimizing the plant growth protocol. The control unit of a plant growth unit first receives the user-selected plant growth protocol in step 42 and transmits one or more device-specific activation signals in step 43 in order to achieve a desired growth pattern defined by the user-selected plant growth protocol. After the elapse of a predetermined duration that is reflective of the time period during which the given plant becomes fully developed, the grown plant is graded by the control unit in step 44. The score given to the fully grown plant is weighted based on several factors such as yield, plant growth duration and quality.

These three factors of yield, plant growth duration and quality are mutually exclusive in terms of optimizing the growth of a plant. If a plant grower is interested, for example, in achieving a short growth duration, the yield and quality generally would have to be compromised. Likewise, a high plant quality will be at the expense of reduced yield or of increased plant growth duration, or a high yield will be at the expense of reduced quality or of increased plant growth duration. While most of the factors may be automatically determined, the predetermined quality, such as taste or dosage efficacy, needs to be tested and recorded by a human. A test signature, which is indicative of the score and association with the user-selected plant growth protocol, as well as with an identifier of the plant growth unit, is then stored in the database in step 45.

In step 46, values of one or more environmental parameters or plant uptake parameters are changed, or the plant growth duration is changed, and then steps 42-45 are repeated for the updated plant growth protocol. After a large number of test signatures are stored in the database, for example in the order of thousands, those test signatures that have a score greater than a predetermined threshold are prioritized in step 47. One of the prioritized test signatures is user- selected in step 48, such as by a manager, as an optimal plant growth protocol that reflects the favorable characteristics of the plant desired to be reproduced.

For the selected optimal plant growth protocol, the analysis module associates in step 50 different types of visual data with corresponding growth traits of a given plant at the related plant growth conditions, for predetermined levels of development.

The function of the plant-specific model will now be described with reference to Fig. 8, which illustrates a flow chart for generating a plant-specific model.

The plant-specific model is adapted to help predict how the given plant will be influenced by a change in the value of each individual plant growth parameter or in the value of a combination of plant growth parameters. As an example for the functionality of the plant-specific model, the model is able to predict the change that the given plant will undergo after the control unit of a plant growth unit changes the temperature within the interior volume from 25°C, as defined by the plant growth protocol, to 30°C after 3 weeks of growth. In order to predict a change in the plant growth pattern, two plant growth units will have to be monitored simultaneously, wherein the temperature within the interior volume of the first is set at 25°C and the temperature within the interior volume of the second is set at 30°C. If it is revealed that the leaf color changes from green to green-brown in the second plant growth unit, the plant-specific model is expected to predict that the impact on the plant caused by an increase in temperature of 5°C within the interior volume is a change in the leaf color.

Machine learning and statistical techniques may be employed in order to understand the sensitivity and influence of each plant growth parameter.

It will be appreciated that the permutations of changes in plant growth parameters needed to suitably generate a plant-specific model that can reliably predict plant behavior is huge. Since a correspondingly large number of plant growth units is required, one for each combination of plant growth parameters, the database and server are consequently configured with sufficient resources and processing power to suitably command the operation of each control unit and to accommodate the big data generated by the experimentation within each plant growth unit, although the reliance on some scientific fields may assist in reducing the number of plant growth units that are needed.

The control unit of a plant growth unit first receives the previously selected optimal plant growth protocol in step 52 and transmits one or more device-specific activation signals in step 53 in order to achieve a desired growth pattern that supports the optimal plant growth protocol. After the elapse of a predetermined duration that is reflective of the time period during which instantaneous growth traits of the given plant become discernable, sensed data representative of instantaneous plant growth conditions, which have been changed, is acquired in step 54. The imaging device is commanded in step 55 to acquire visual data related to the instantaneous plant growth conditions simultaneously with the acquisition of the sensed data, whereupon the control unit transmits to the server in step 57 the acquired visual data and the associated plant growth conditions, time stamp, and plant growth unit identifier.

The analysis module of the server then classifies the acquired visual data in step 59 according one of a plurality of plant-related classifications that are predefined in the dedicated algorithm, including a normative classification that represents the desired growth pattern for the optimal plant growth protocol at a similar level of development, as well as deviative classifications relative to the normative classification. The dedicated algorithm defines boundary conditions for each of the classifications. The plant being grown in a plant growth unit is then sampled by a human in step 61 at the same level of development that corresponds to that of the acquired visual data. When the acquired visual data is found in step 63 to significantly deviate from sampled data, a model administrator enters refined boundary conditions in step 65 for the normative classification and each of the deviative classifications, whereupon the model is generated in step 67. Typical deviative visual data relate to a deviative leaf color, such as a brownish leaf color instead of a green leaf color. These steps are repeated for other plant growth units. The model may undergo additional refining in conjunction with machine learning while the accuracy in prediction may improve.

The model may likewise be used to determine which input of changed plant growth conditions is needed to produce a desired output regarding a deviation value of visual data. A selected output is able to counterbalance previously acquired deviative visual data using closed loop feedback so that desired growth traits will be able to be manifested.

Fig. 9 illustrates a flow chart for automatically initiating a recovery process within a plant growth unit if the real-time acquired visual data significantly deviates from expected values. Following planting, the control unit transmits one or more device-specific activation signals in step 72 to achieve a desired growth pattern defined by the optimal plant growth protocol and in accordance with corresponding optimal plant growth conditions. Plant-specific visual data representative of instantaneous growth traits is then acquired in step 74 at predetermined times corresponding to different levels of development and transmitted to the server in step 76 for analysis. If the analysis module determines in step 78 that the acquired visual data significantly deviates from expected data, e.g. expected visual data, associated with the optimal plant growth protocol at a similar level of development in accordance with instructions defined in the dedicated algorithm, a correcting value for correcting the previously determined deviation is calculated in step 80. The calculated correcting value is entered into the model and a corresponding combination of changed plant growth conditions is derived in step 82, whereupon the server transmits data to the control unit in step 84 which is indicative of the derived combination of plant growth conditions to be changed. In response, the control unit transmits one or more corrected device specific activation signals in step 86 to the corresponding electromechanical devices in order to ensure that growth pattern defined by the optimal plant growth protocol will be complied with following the corrective actions. Thus a plant having a predetermined quality may be automatically reproduced in accordance with the recovery process.

While some embodiments of the invention have been described by way of illustration, it will be apparent that the invention can be carried out with many modifications, variations and adaptations, and with the use of numerous equivalents or alternative solutions that are within the scope of persons skilled in the art, without exceeding the scope of the claims.