WO2021099497A1 - Dispositif de commande, système et procédé d'irrigation - Google Patents

Dispositif de commande, système et procédé d'irrigation Download PDF

Info

Publication number
WO2021099497A1
WO2021099497A1 PCT/EP2020/082742 EP2020082742W WO2021099497A1 WO 2021099497 A1 WO2021099497 A1 WO 2021099497A1 EP 2020082742 W EP2020082742 W EP 2020082742W WO 2021099497 A1 WO2021099497 A1 WO 2021099497A1
Authority
WO
WIPO (PCT)
Prior art keywords
irrigation
time period
model
processing resource
schedule
Prior art date
Application number
PCT/EP2020/082742
Other languages
English (en)
Inventor
Andrew Peacock
David CORNE
David Kane
Abhimanyu BHARGAVA
Original Assignee
Farm-Hand Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Farm-Hand Ltd filed Critical Farm-Hand Ltd
Priority to GB2208397.6A priority Critical patent/GB2605321A/en
Publication of WO2021099497A1 publication Critical patent/WO2021099497A1/fr

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B17/00—Systems involving the use of models or simulators of said systems
    • G05B17/02—Systems involving the use of models or simulators of said systems electric
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G25/00—Watering gardens, fields, sports grounds or the like
    • A01G25/16—Control of watering
    • A01G25/165—Cyclic operations, timing systems, timing valves, impulse operations
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G25/00—Watering gardens, fields, sports grounds or the like
    • A01G25/16—Control of watering
    • A01G25/167—Control by humidity of the soil itself or of devices simulating soil or of the atmosphere; Soil humidity sensors
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
    • G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/20—Pc systems
    • G05B2219/26—Pc applications
    • G05B2219/2625—Sprinkler, irrigation, watering

Definitions

  • a controller for controlling an irrigation apparatus an irrigation system comprising the controller and the irrigation apparatus, and irrigation methods are described herein for use in particular, though not exclusively, for the irrigation of one or more plants or crops in one or more irrigation areas.
  • Automated irrigation systems are well known, and sophisticated irrigation models have been developed. In large scale agriculture in developed markets such as in the US, Europe and Australia, such irrigation models may use a range of inputs, including weather forecasts and real-time or almost real-time measurements of various physical parameters and crop growth to provide precisely tailored irrigation.
  • irrigation systems may have to operate under conditions which are less predictable than the conditions existing in developed markets. For example, in some developing economies switching on or off of irrigation systems may be dependent on electricity supply to the irrigation system being available. In extreme cases, in view of free or subsidised electricity being available for irrigation purposes in India, irrigation systems can be in an always on state, with irrigation being provided when the electricity company is able to provide electricity to the farmer. The irrigation systems in such cases can sometimes be used effectively for load-balancing purposes by electricity companies, and over- or under-watering can be a regular occurrence.
  • An improved or at least alternative irrigation system is desired, for example an irrigation system that can provide improved irrigation capabilities in an efficient and simple manner in varied or difficult farming conditions.
  • a controller for controlling an irrigation apparatus comprising: a memory storing an irrigation model; and a processing resource, wherein the processing resource is configured to: determine a first irrigation schedule for the irrigation apparatus for a first time period based on the irrigation model; and control the irrigation apparatus so that the irrigation apparatus provides water to one or more locations within an irrigation area during the first time period according to the first irrigation schedule.
  • the processing resource may be configured to receive input data associated with the first time period, optionally including user input data which represents a user assessment of, for example, the irrigation provided in the irrigation area during the first time period.
  • the processing resource may be configured adjust the irrigation model, optionally according to an irrigation model machine learning process, based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation model.
  • the processing resource may be configured to determine a second irrigation schedule for the irrigation apparatus for a second time period subsequent to the first time period based on the adjusted irrigation model.
  • the processing resource may be configured to control the irrigation apparatus so that the irrigation apparatus provides water to the one or more locations within the irrigation area during the second time period according to the second irrigation schedule.
  • Such an irrigation controller may be used to control an irrigation apparatus even when the conditions under which the irrigation controller operates are unpredictable.
  • Such an irrigation controller may also take into account user input data or user preferences.
  • the processing resource may be configured to receive input for a plurality of further time periods and to adjust the irrigation model further based on the input for the plurality of further time periods.
  • the irrigation model may gradually be refined and become more specific to a particular irrigation area, for example a farm or part of a farm.
  • the irrigation model may be refined over a longer period, for example weeks, months or years. Any suitable irrigation mode may be used as a starting point and then adjusted by the controller.
  • the irrigation model may include a crop coefficient or look-up table that can be applied to determine irrigation schedule for a time period (for example, amount of water for a time period) and the adjusting of the irrigation model may comprise an adjusting of the crop coefficient or look-up table.
  • the processing resource may be configured to store the adjusted irrigation model in the memory.
  • the user input data may comprise an indication of whether more water or less water was needed for irrigation of the irrigation area during the first time period.
  • the input data may include data measured by one or more sensors in the irrigation area during the first time period.
  • the input data may include data relating to the forecast weather and/or environmental conditions in a region which includes the irrigation area for the second time period.
  • the input data may comprise remote sensing data, optionally satellite data.
  • the remote sensing data may comprise data representative of at least one of ground reflectance, surface temperature, rainfall, ground moisture level, fractional vegetative cover, ground aquifer level.
  • the processing resource may be configured to receive the data relating to the forecast weather and/or environmental conditions in the region for the second time period from one or more remote resources located remotely from the irrigation system such as one or more remote servers and/or one or more remote databases, for example via the cloud.
  • the processing resource may be configured to adjust the data relating to the forecast weather and/or environmental conditions in the region for the second time period received from the one or more remote resources based on measured data relating to the weather and/or environmental conditions in the irrigation area during the first time period to thereby generate a local forecast of the weather and/or environmental conditions in the irrigation area for the second time period.
  • the processing resource may be configured to adjust the irrigation model, optionally according to the irrigation model machine learning process, based at least in part on the local forecast of the weather and/or environmental conditions in the irrigation area for the second time period.
  • the processing resource may be configured to receive data relating to the forecast weather and/or environmental conditions in the region for the first time period from the one or more remote resources.
  • the processing resource may be configured to use the data relating to the forecast weather and/or environmental conditions in the region for the first time period received from one or more remote resources as a local weather model for the irrigation area for the first time period.
  • the processing resource may be configured to store the local weather model for the irrigation area for the first time period in the memory.
  • the processing resource may be configured to adjust the local weather model, optionally according to a weather model machine learning process, based at least in part on data relating to the forecast weather and/or environmental conditions in the region for the second time period received from the one or more remote resources and measured data relating to the weather and/or environmental conditions in the irrigation area during the first time period to thereby determine an adjusted local weather model.
  • the processing resource may be configured to adjust the irrigation model, optionally according to the irrigation model machine learning process, based at least in part on the adjusted local weather model to thereby obtain the adjusted irrigation model.
  • the irrigation model may comprise a water balance model and an irrigation policy.
  • the processing resource may be configured to construct the water balance model for the first time period based at least in part on static information relating to at least one of the configuration of the irrigation apparatus, the configuration of the irrigation area and/or the configuration of one or more plants in the irrigation area.
  • the processing resource may be configured to construct the water balance model for the first time period based at least in part on the local weather model for the first time period.
  • the processing resource may be configured to determine a provisional first irrigation schedule for the irrigation apparatus for the first time period based on the water balance model.
  • the processing resource may be configured to use the provisional first irrigation schedule as the first irrigation schedule for the irrigation apparatus for the first time period.
  • the processing resource may be configured to adjust the water balance model based at least in part on the received input data associated with the first time period and the adjusted local weather model to thereby obtain an adjusted water balance model.
  • the processing resource may be configured to determine a provisional second irrigation schedule for the irrigation apparatus for the second time period based on the adjusted water balance model.
  • the processing resource may be configured to adjust the provisional second irrigation schedule according to the irrigation policy to thereby determine the second irrigation schedule for the irrigation apparatus for the second time period.
  • Adjusting the irrigation model, optionally according to the irrigation model machine learning process, based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation model may comprise adjusting the irrigation policy, optionally according to the irrigation model machine learning process, based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation policy.
  • the processing resource may be configured to store the water balance model and the irrigation policy in the memory.
  • the processing resource may be configured to store the irrigation policy at one or more remote locations such as one or more remote servers and/or one or more remote databases.
  • the irrigation policy may be common to a plurality of different irrigation systems including the irrigation system, wherein each different irrigation system comprises a corresponding irrigation apparatus configured to provide water to one or more locations within a corresponding irrigation area.
  • the irrigation areas corresponding to the different irrigation systems may contain the same plant type, plant kind and/or plant variety.
  • an irrigation system comprising: the controller as described above; and the irrigation apparatus.
  • the irrigation apparatus may comprise one or more controllable devices.
  • the processing resource may be configured to control the one or more controllable devices so as to control the provision of water to the one or more locations within the irrigation area according to the first and second irrigation schedules.
  • the one or more controllable devices may comprise at least one of: one or more pumps; one or more valves; one or more actuators, for example one or more relays or switching arrangements; and one or more inverters.
  • the system may comprise one or more sensors for measuring data.
  • the processing resource may be configured to receive data measured by the one or more sensors during the first time period as part of the input data.
  • One or more of the sensors may be configured to measure data relating to environmental and/or weather conditions in the irrigation area.
  • One or more of the sensors may be configured to measure at least one of: precipitation such as rainfall, for example in the irrigation area; air temperature, for example in the irrigation area; air pressure, for example in the irrigation area; humidity, for example in the irrigation area; irradiance and/or lux, for example in the irrigation area; and wind speed, for example in the irrigation area.
  • One or more of the sensors may be configured to measure soil conditions in the irrigation area.
  • One or more of the sensors may be configured to measure at least one of: soil moisture; soil temperature; and soil salinity.
  • One or more of the sensors may be configured to measure a condition of one or more plants in the irrigation area.
  • One or more of the sensors may be configured to provide canopy image data and/or thermal image data of one or more plants in the irrigation area.
  • One or more of the sensors may comprise: a camera or image sensor such as a visible or infrared camera or image sensor; a thermal camera or a thermal image sensor; or a hyperspectral camera or a hyperspectral image sensor a thermocouple or array of thermocouple for measuring canopy temperature
  • a camera or image sensor such as a visible or infrared camera or image sensor
  • a thermal camera or a thermal image sensor or a hyperspectral camera or a hyperspectral image sensor a thermocouple or array of thermocouple for measuring canopy temperature
  • One or more of the sensors may be configured to measure a status, and/or monitor the operation, of the irrigation apparatus.
  • One or more of the sensors may comprise: a phase detector; a voltmeter; a sensor for measuring grid frequency; an ammeter; a water pressure sensor; a water flow meter; or a water tank level sensor.
  • the system may comprise a user interface.
  • the user interface may be implemented by, or provided on, a mobile electronic device such as a smart phone.
  • the system may comprise an app for a mobile electronic device, which app, when executed by the mobile electronic device, provides the user interface.
  • the controller may be located at least partly at a location remote from the irrigation apparatus, for example at a remote server.
  • an irrigation method comprising: determining a first irrigation schedule for an irrigation apparatus for a first time period based on an irrigation model, the irrigation apparatus being configured to provide water to one or more locations within an irrigation area; controlling the irrigation apparatus so that the irrigation apparatus provides water to the one or more locations within the irrigation area during the first time period according to the first irrigation schedule; receiving input data associated with the first time period, optionally including user input data which represents a user assessment of the irrigation provided in the irrigation area during the first time period; adjusting the irrigation model, optionally according to an irrigation model machine learning process, based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation model; and determining a second irrigation schedule for the irrigation apparatus for a second time period subsequent to the first time period based on the adjusted irrigation model.
  • the method may comprise controlling the irrigation apparatus so that the irrigation apparatus provides water to the one or more locations within the irrigation area during the second time period according to the second irrigation schedule.
  • the method may comprise storing the adjusted irrigation model.
  • the method may comprise using data relating to the forecast or predicted weather and/or environmental conditions in a region including the irrigation area for the first time period received from one or more remote resources as a local weather model for the irrigation area for the first time period.
  • the method may comprise storing the local weather model for the irrigation area for the first time period.
  • the method may comprise adjusting the local weather model, optionally according to a weather model machine learning process, based at least in part on data relating to the forecast or predicted weather and/or environmental conditions in the region for the second time period received from the one or more remote resources and measured data relating to the weather and/or environmental conditions in the irrigation area during the first time period to thereby determine an adjusted local weather model.
  • the method may comprise adjusting the irrigation model, optionally according to the irrigation model machine learning process, based at least in part on the adjusted local weather model to thereby obtain the adjusted irrigation model.
  • the irrigation model may comprise a water balance model and an irrigation policy.
  • the method may comprise constructing the water balance model for the first time period based at least in part on static information relating to at least one of the configuration of the irrigation apparatus, the configuration of the irrigation area and/or the configuration of one or more plants in the irrigation area.
  • the method may comprise constructing the water balance model for the first time period based at least in part on the local weather model for the first time period.
  • the method may comprise determining a provisional first irrigation schedule for the irrigation apparatus for the first time period based on the water balance model.
  • the method may comprise adjusting the provisional first irrigation schedule according to the irrigation policy to thereby determine the first irrigation schedule for the irrigation apparatus for the first time period.
  • the method may comprise adjusting the water balance model based at least in part on the received input data associated with the first time period and the adjusted local weather model to thereby obtain an adjusted water balance model.
  • the method may comprise determining a provisional second irrigation schedule for the irrigation apparatus for the second time period based on the adjusted water balance model.
  • the method may comprise adjusting the provisional second irrigation schedule according to the irrigation policy to thereby determine the second irrigation schedule for the irrigation apparatus for the second time period.
  • Adjusting the irrigation model, optionally according to the irrigation model machine learning process, based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation model may comprise adjusting the irrigation policy, optionally according to the irrigation model machine learning process, based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation policy.
  • the irrigation policy may be common to a plurality of different irrigation systems including the irrigation system, wherein each different irrigation system comprises a corresponding irrigation apparatus configured to provide water to one or more locations within a corresponding irrigation area.
  • the irrigation areas corresponding to the different irrigation systems may contain the same plant type, plant kind and/or plant variety.
  • an automatic irrigation system for a farm or other irrigation location comprising irrigation apparatus for providing water to location(s) on the farm or other irrigation location; a processing resource for determining an irrigation schedule for the irrigation apparatus that comprises calculated volumes of water to be provided by the irrigation apparatus; a plurality of inputs that provide data to the processing resource; a memory storing an irrigation model, wherein the processing resource is configured to determine the irrigation schedule according to the model and said inputs; the system further comprise a user interface that is configured to receive input from the farmer or other user representing an assessment of the irrigation provided during a preceding time period (for example, the preceding day or week); and the processing resource is configured to alter the irrigation model and/or the irrigation schedule for current or future periods dependent on the user input representing an assessment of the irrigation provided during said preceding time period.
  • the controller may be configured to alter the irrigation model and/or the irrigation schedule, optionally according to a machine learning process, and optionally the user input comprises an indication of whether more water or less water was needed for said preceding time period.
  • the user interface may comprise a mobile phone user interface.
  • the processing resource may be located at least partly at a location remote from the irrigation system, for example at a remote server.
  • the processing resource and/or model may be used by a plurality of farms or other irrigation locations, and the model as used by different farms or other irrigation locations may diverge over time, optionally based on the machine-learning process, and the different user inputs for different farms or other irrigation locations.
  • the input data may include a localised weather forecast for the farm or other irrigation location.
  • the localised weather forecast may be determined based on the location of the farm or other irrigation location and based on local weather-related or other environmental measurements using weather sensors or other environmental sensors at the farm or other irrigation location.
  • the processing resource may be configured to receive weather forecast data from a remote source, and to apply a machine learning process to alter said weather forecast data based on learned weather behaviour determined from the sensors at the farm or other irrigation location.
  • the inputs may comprise inputs from a plurality of sensors, optionally wherein the sensors comprises at least one of a flow meter, a soil moisture sensor, a temperature sensor, a humidity sensor, a water tank level sensor.
  • the controller may comprise a dedicated microcontroller that comprises a memory that is configured to communicate with a plurality of sensors and said processing resource.
  • the controller may be remotely controllable by a user via a mobile phone.
  • FIG. 1 is a schematic illustration of an irrigation system irrigating an irrigation area
  • FIG. 2 is a schematic block diagram of a part of the irrigation system of FIG. 1 ;
  • FIG. 3 is a flow diagram illustrating a first irrigation method performed by the irrigation system of FIG. 1 ;
  • FIG. 4 is a flow diagram illustrating a second irrigation method performed by the irrigation system of FIG. 1 ;
  • FIG. 5 is a flow diagram illustrating a third irrigation method performed by the irrigation system of FIG. 1.
  • an irrigation system generally designated 2 including an irrigation apparatus generally designated 4 for providing water to a plurality of locations 6 within an irrigation area generally designated 8, a controller generally designated 10 for controlling the irrigation apparatus 4, and a plurality of sensors 12.
  • the irrigation system 2 irrigates one or more plants (not shown) located in the vicinity of the plurality of locations 6 within the irrigation area 8.
  • the irrigation apparatus 4 includes a reservoir of water 13 or a connection to a reservoir of water 13, a pump 14, one or more valves 16, and one or more fluid conduits such as one or more hoses 18 connecting the pump 14 to the reservoir 13, connecting the pump 14 to the valves 16, and connecting each valve 16 to a corresponding plurality of the locations 6.
  • Each hose 18 connecting a valve to 16 to a corresponding plurality of the locations 6, has one or more apertures, outlets and/or nozzles at each location 6.
  • the reservoir may, for example, be a water tank, pond, river, lake, well or ground water source.
  • the irrigation apparatus 4 includes one or more actuators such as one or more relays, inverters, switching arrangements, or the like (not shown in FIG.1) for actuating the pump 14 and the valves 16.
  • the controller 10 selectively provides power to the actuators via one or more cables 20 as indicated by the dashed lines in FIG. 1 so as to selectively actuate the pump 14 and the valves 16 for the irrigation of one or more plants (not shown) located in the vicinity of the plurality of locations 6 within the irrigation area 8.
  • the controller 10 is configured for wireless communication via the cloud 28 with one or more mobile electronic devices 22 and with one or more remote resources such as one or more remote servers 24 or databases 26.
  • FIG. 2 schematically illustrates the controller 10 and the sensors 12 in more detail.
  • Power for the controller 10 can be provided by a grid connection or by on-site generation (e.g. a solar PV cell with battery - not shown).
  • the sensors 12 include one or more sensors 12a for measuring the environmental and/or weather conditions in the irrigation area 8 such as one or more sensors for measuring at least one of: precipitation such as rainfall, air temperature, air pressure, humidity, irradiance, lux, output of the solar PV cell (e.g. Wh) and wind speed.
  • the sensors 12 further include one or more sensors 12b for measuring soil conditions in the irrigation area 8 such as soil moisture, soil temperature, and soil salinity.
  • the sensors 12 further include one or more sensors 12c for measuring plant conditions in the irrigation area 8 such as thermocouples, visible or IR cameras or image sensors, thermal cameras or image sensors, and hyperspectral imaging cameras or image sensors.
  • the sensors 12 further include one or more irrigation management sensors 12d for measuring the status and/or operation of the irrigation apparatus 4 such as one or more of: phase detectors, voltmeters, grid frequency sensors, ammeters, water flow detectors or meters, water pressure sensors and water tank level sensors.
  • sensors Although a variety of sensors have been mentioned above, it is a feature of certain embodiments that only a small number of sensors may be used, for example only one or more irrigation management sensors. In conjunction with simple user feedback provided via the user interface, discussed further below, and the use of the machine learning algorithm, even such simple systems can provide for improved irrigation and the gradually tailoring of an irrigation schedule over time to a particular irrigation area, for example a particular farm. The use of more sensors can nevertheless provide increased accuracy, sensitivity or robustness and increased information regarding useful for assessing farm management.
  • the controller 10 includes a micro-controller 30 having a processing resource 32 and a memory 34.
  • the controller 10 further includes one or more expansion slots 36, a user interface 38, storage 40 and one or more communications modules 42.
  • the controller 10 receives input data via the mobile electronic device 22, the remote resources 24, 26 and the sensors 12a, 12b, 12c and 12d. At least some of the input data received by the controller 10 is static input data which does not change with time and which includes static information relating to the configuration of the irrigation apparatus 4, static information relating to the irrigation area 8, and static information relating to the one or more plants in the irrigation area 8. At least some of the input data received by the controller 10 is dynamic input data which does change with time and which includes dynamic input data which is measured during a historical first time period which has expired, and dynamic input data which is predicted or forecast data for a future second time period subsequent to the first time period.
  • the input data received by the controller 10 comprises remote sensing data, optionally satellite data, as well as or instead of local sensor data.
  • the remote sensing data may comprise data representative of at least one of ground reflectance, surface temperature, rainfall, ground moisture level, fractional vegetative cover, ground aquifer level
  • the processing resource 32 of the controller 10 constructs an initial irrigation model 50 for the first time period from static site, crop and local information 52 and from a local weather model 54 for the first time period and stores the initial irrigation model 50 in the memory 34.
  • Any suitable type of know irrigation model may be used as, or to contruct, the initial irrigation model, for example an irrigation model as described in Walter, I. A., Allen, R.G., Elliott, R., Jensen, M.E., Itenfisu, D., Mecham, B., Howell, T.A., Snyder, R., Brown, P., Echings, S. and Spofford, T., 2000.
  • ASCE's standardized reference evapotranspiration equation. In Watershed management and operations management 2000 (pp. 1-11).
  • the static site, crop and local information 52 can be provided by a user such as a farmer via a farmer app on the mobile electronic device 22.
  • static site, crop and local information 52 can be pre-stored and can be the same for all farms or other sites in a region.
  • the static information can be the same for all sites.
  • the static site, crop and local information 52 includes static information relating to the configuration of the irrigation apparatus 4, static information relating to the one or more plants in the irrigation area 8, and static information relating to the irrigation area 8.
  • the static information relating to the one or more plants in the irrigation area 8 can include the number and distribution of the plants in the irrigation area 8, the nature of the plants including at least one of the plant type, plant kind and plant variety, the size, age, and agronomic coefficients of the plants in the irrigation area 8.
  • the static information relating to the irrigation area 8 can include the soil type in the irrigation area 8 and the topography of the irrigation area 8.
  • the processing resource 32 receives data relating to the forecast or predicted weather and/or environmental conditions in the wider region which includes the irrigation area 8 for the first time period from the one or more remote resources 24, 26.
  • the processing resource 32 initially uses the data relating to the forecast or predicted weather and/or environmental conditions in the wider region for the first time period received from the one or more remote resources 24, 26 as the local weather model 54 for the irrigation area 8 for the first time period.
  • the processing resource 32 stores the local weather model 54 in the memory 34.
  • the processing resource 32 determines a first irrigation schedule 56 for the irrigation apparatus 4 for the first time period based on the initial irrigation model 50.
  • the processing resource 32 controls the irrigation apparatus 4 so that the irrigation apparatus 4 provides water to the one or more locations 6 within the irrigation area 8 during the first time period according to the first irrigation schedule as shown at step 58 of FIG. 3.
  • the processing resource 32 receives dynamic input data at step 60 of FIG. 3 relating to the first time period.
  • the dynamic input data relating to the first time period includes data measured by the sensors 12a, 12b, 12c and 12d during the first time period.
  • the dynamic input data relating to the first time period also includes user input data from a user interface of the mobile electronic device 22, wherein the user input data represents a user assessment of the irrigation provided in the irrigation area 8 during the first time period.
  • the user input data may comprise an indication of whether more water or less water was needed for irrigation of the irrigation area 8 during the first time period.
  • the processing resource 32 adjusts the local weather model 54 according to a weather model machine learning process based at least in part on data relating to the forecast or predicted weather and/or environmental conditions in the wider region for the second time period received from the one or more remote resources 24, 26 and based at least in part on the data relating to the weather and/or environmental conditions in the irrigation area 8 as measured by the one or more of the environmental/weather sensors 12a during the first time period to thereby obtain an adjusted or improved local weather model 54 for the second time period.
  • a weather model machine learning process is described in more detail in US patent publication no. US 2017/0131435.
  • the use of a local weather model can be an important feature of certain embodiments.
  • the processing resource 32 then adjusts the irrigation model 50 according to an irrigation model machine learning process based on the received input data and the adjusted local weather model 54 so as to obtain an adjusted or improved irrigation model 50.
  • the processing resource 32 determines a second irrigation schedule 56 for the irrigation apparatus 4 for the second time period based on the adjusted irrigation model 50.
  • the processing resource 32 then controls the irrigation apparatus 4 so that the irrigation apparatus 4 provides water to the one or more locations 6 within the irrigation area 8 during the second time period according to the second irrigation schedule at step 58.
  • One of ordinary skill in the art will understand that the cycle described above with reference to FIG. 3 may continue for one or more further time periods subsequent to the second time period.
  • Such continuance of the cycle of adjustment for multiple time periods is an important feature of certain embodiments, and enable gradual refinement of the irrigation model over a lengthy period of time, for example weeks, months or years based on the machine learning model and simple feedback from a user day-by-day (for instance, in some embodiments feedback indicating simply too much or too little water) such that the irrigation model gradually becomes tailored to a specific farm or even field or other area of a farm.
  • a system user e.g. a farmer
  • Data on the irrigation schedule including irrigation volumes and timing, and feedback provided by the system user is captured and stored on a remote server 24 or a database 26 via the cloud 28.
  • the user interface provided to the farmer or other user can be simple in nature.
  • the user interface provides an input arrangement (for example one or more areas on a touch screen, or one or more buttons) that can be used by the farmer or other user to indicate that the more or less water would have been desirable for a preceding or current time period, or that more or less water is desired for a current or future period.
  • Such input arrangements for example touchscreen areas or buttons can be indicated using arrows or other pictograms rather than text in some embodiments.
  • Such simplified user interface arrangements can be particularly useful in rural communities in third world countries, where for example there may be a variety of local languages or where levels of literacy may not be high.
  • the app or other user interface or user input device can, in some embodiments, be used to provide control commands as well as to provide feedback on the irrigation that has taken place.
  • the mobile device may be configured to communicate with the micro-controller and to over-ride the irrigation schedule and/or to switch on or off irrigation on command of a user received via the user interface.
  • Such user over-ride and/or remote control of the irrigation system can be particularly useful in some rural communities where there can be safety issues for farmers tending or inspecting their fields or crops or irrigation systems late at night or early in the morning.
  • FIG. 4 shows a variant of the irrigation method of FIG. 3.
  • the irrigation method of FIG. 4 comprises the processing resource 32 constructing a water balance model (WBM) 150a and an irrigation policy 150b.
  • the processing resource 32 constructs the WBM 150a for the first time period based at least in part static information relating to at least one of the configuration of the irrigation apparatus 4, the configuration of the irrigation area 8 and/or the configuration of one or more plants in the irrigation area 8.
  • the processing resource 32 constructs the WBM 150a for the first time period based at least in part on the local weather model 154 for the first time period.
  • the processing resource 32 stores the WBM 150a in the memory 34.
  • the processing resource 32 uses data relating to the forecast or predicted weather and/or environmental conditions in a region including the irrigation area 8 for the first time period received from the one or more remote resources 24, 26 as the local weather model 154 for the irrigation area 8 for the first time period.
  • the processing resource 32 stores the local weather model 154 for the irrigation area 8 for the first time period in the memory 34.
  • the processing resource 32 determines a provisional first irrigation schedule 156a for the irrigation apparatus 4 for the first time period based on the WBM 150a and static information relating to at least one of the configuration of the irrigation apparatus 4 and uses the provisional first irrigation schedule 156a as the first irrigation schedule 156b for the irrigation apparatus 4 for the first time period.
  • the initial irrigation policy 150b for the first time period makes no changes to the provisional first irrigation schedule 156a.
  • the processing resource 32 stores the initial irrigation policy 150b for the first time period in the memory 34.
  • the processing resource 32 then controls the irrigation apparatus 4 for the first time period according to the first irrigation schedule 156b at 158.
  • the processing resource 32 receives dynamic input data at step 160 relating to the first time period.
  • the dynamic input data relating to the first time period includes data measured by the sensors 12a, 12b, 12c and 12d during the first time period.
  • the dynamic input data relating to the first time period also includes user input data from a user interface of the mobile electronic device 22, wherein the user input data represents a user assessment of the irrigation provided in the irrigation area 8 during the first time period.
  • the user input data may comprise an indication of whether more water or less water was needed for irrigation of the irrigation area 8 during the first time period.
  • the processing resource 32 adjusts the local weather model 154 according to a weather model machine learning process based at least in part on data relating to the forecast or predicted weather and/or environmental conditions in the wider region for the second time period received from the one or more remote resources 24, 26 and based at least in part on the data relating to the weather and/or environmental conditions in the irrigation area 8 as measured by the one or more of the environmental/weather sensors 12a during the first time period to thereby obtain an adjusted or improved local weather model 154 for the second time period.
  • the processing resource 32 also adjusts the WBM 150a based at least in part on the received input data associated with the first time period and the adjusted local weather model 154 to thereby obtain an adjusted WBM 150a.
  • the processing resource 32 adjusts the irrigation policy 150b according to an irrigation model machine learning process based at least in part on the received input data associated with the first time period to thereby obtain an adjusted irrigation policy 150b.
  • the processing resource 32 determines a provisional second irrigation schedule 156a for the irrigation apparatus 4 for the second time period based on the adjusted water balance model 150a.
  • the processing resource 32 adjusts the provisional second irrigation schedule 156a according to the irrigation policy 150b to thereby determine the second irrigation schedule 156 for the irrigation apparatus 4 for the second time period.
  • the WBM is executed in a regular cycle (once per day, or multiple times per day).
  • the WBM calculates the amount of water required by the one or more plants or the crop in the irrigation area 8 during the next time period or cycle, based on current conditions and forecast conditions.
  • the key inputs to the WBM are static information about the crop (plant variety, agronomic coefficients, etc%), together with (but not required) suggestions and constraints on the irrigation schedule from a system user such as a farmer or crop manager and dynamic, accurate site-specific weather forecasts (temperature, light levels, wind-speed, rainfall, humidity etc.) from the local weather model 154.
  • the WBM produces an estimate 156a of the amount of water required by the one or more plants or the crop in the irrigation area 8 during the next time period or cycle, and it optimises the planned execution of the irrigation to deliver this estimate 156a in a way that matches preferences supplied by the system user (if any). These estimates 156a are then sent to the Irrigation Policy 150b.
  • the Irrigation Policy 150b adjusts the output 156a of the WBM.
  • the Irrigation Policy 150b may adjust the amount of water to be supplied, or the timing of delivery, or both. These adjustments are made on the basis of a hierarchical reinforcement learning (RL) algorithm that adapts and maintains the Irrigation Policy 150b, based on the feedback 160.
  • RL hierarchical reinforcement learning
  • the irrigation control decisions are executed physically by the irrigation system 2. This will normally be via automated means.
  • Timestamped feedback 160 of multiple kinds is fed back into the system. Examples include: (i) a system user indicates an adjustment to the amount of water supplied (or to be supplied); (ii) onsite weather sensing provides actuals for contrast with forecasts; (iii) soil moisture sensors indicate actual moisture levels; (iv) estimates of water stress provided by smartphone apps via image analysis;
  • All feedback is supplied to the irrigation policy 150b.
  • Some of the feedback 160 is used directly by the local weather model, which will be incorporated into the regular process of building site-specific machine learning weather models.
  • Some of the feedback 160 goes directly to the WBM e.g. to provide immediate adjustment based on revised system user suggestions.
  • the irrigation policy 150b comprises an irrigation policy cloud (I PC) 250b as illustrated in FIG. 5.
  • the irrigation policy cloud 250b is common to a plurality of different irrigation systems 2, 2’, 2” etc. including the irrigation system 2, wherein each different irrigation system 2, 2’, 2” comprises a corresponding irrigation apparatus configured to provide water to one or more locations within a corresponding irrigation area, and wherein the irrigation areas corresponding to the different irrigation systems 2, 2’, 2” contain the same plant type, plant kind and/or plant variety.
  • the irrigation policy cloud (I PC) 250b is stored or maintained on a remote server or a remote database and accessed by the micro-controller of any of the irrigation systems 2, 2’, 2” via the cloud.
  • Each micro-controller may read the I PC 250b via appropriate messaging protocols.
  • Each micro-controller may write to the I PC 250b so as to update the I PC 250b with data measured by the sensors of the corresponding irrigation system 2, 2’, 2” via appropriate messaging protocols. Any suitable messaging protocol may be used.
  • a single IPC manages all irrigation systems 2, 2’, 2” (e.g. all farms and crops), enabling cross-transfer of irrigation policy learning across the same crops (e.g. the same plant types, plant kinds or plant varieties) in different irrigation areas (e.g. different farms) over time.
  • the IPC connects all irrigation systems 2, 2’, 2”.
  • irrigation at Farm/Crop 1 is under the control and management of a first irrigation system 2 that is set up specifically for that farm/crop; irrigation at Farm/Crop 2 is under the control and management of a second irrigation system 2’ that is set up specifically for that farm/crop; irrigation at Farm/Crop 3 is under the control and management of a third irrigation system 2” that is set up specifically for that farm/crop; and likewise for all other farms/crops.
  • each of these irrigation systems 2, 2’, 2” uses the same IPC, and are hence connected by the same IPC, as illustrated by FIG. 5.
  • the reinforcement learning process which control adjusts and corrects the irrigation policy on the basis of various forms of feedback, can learn far more speedily and effectively, by experiencing and learning from multiple environments in a short time from multiple farms, rather than waiting for several years to build up such a variety of experience from a single farm. Learning is applied where most relevant - e.g. farm 1 may have a corn crop experiencing water stress only one week after planting for the first time, in particularly hot conditions.
  • the irrigation policy will then incorporate and reflect learned adjustments from other farms that have experienced similar conditions with their corn crop at the same growth stage.
  • the micro-controller 30 is configured to provide information from a plethora of weather and crop/farm management sensors 12a, 12b, 12c, 12d to the irrigation policy cloud 250b. This information is used in machine learning modules in the local weather model 154 and the WBM 150a to modify and improve the accuracy of the irrigation schedules.
  • the micro-controller 30 is configured to provide information from the plethora of irrigation management sensors 12a, 12b, 12c, 12d to the irrigation policy cloud 250b.
  • the information from the plethora of irrigation management sensors 12a, 12b, 12c, 12d is used to verify that the issued irrigation schedules have been enacted.
  • the micro-controller 30 is configured to receive via an appropriate messaging protocol, irrigation schedules from the irrigation policy cloud 250b.
  • the micro-controller 30 is configured to actuate, turn on and off, or open and close controllable or actuatable devices on the farm (e.g. the water pump 14 and valves 16) to allow irrigation to pass to the one or more locations 6 or other defined areas of the irrigation area 8 (e.g. farm) commensurate with the received irrigation schedules.
  • the micro-controller 30 may control or actuate the actuatable devices 14, 16 by selectively providing power to the actuatable devices 14, 16 via the cables 20.
  • the micro-controller 30 may interact with a farmer app on the mobile electronic device 22 via the cloud 28 to manage irrigation scheduling for the irrigation area 8 (e.g. on a farm).
  • Data from the sensors 12a, 12b, 12c and 12d is sent from the micro-controller 30 of irrigation system 2 to the IPC 250b using a messaging protocol.
  • Irrigation schedules may be sent from the IPC 250b to the micro-controller 30 using a messaging protocol.
  • Management data for the irrigation area 8 may be sent from a farmer app on the mobile electronic device 22 to the micro-controller 30 using a messaging protocol. Management data may be sent by multiple users from multiple farms or other sites.
  • the management data may include the site, crop and local information 152 for multiple farms or other sites and may make up or be included in crop and farm management data schema.
  • the management data may include instructions for the interruption of irrigation schedules and the selection of an irrigation mode (manual or automated).
  • User feedback information (e.g. farmer feedback information) requesting changes to the automated schedules may be sent from the farmer app on the mobile electronic device 22 to the micro-controller 30 using a messaging protocol.
  • machine learning may be used to modify irrigation models and/or schedules based on input, for example user input. Any suitable type of machine learning process may be used. In some cases a reinforcement learning process may be used. Any suitable process that adjusts a model based on input data to provide improved outcome may be used. Any suitable metric for determining outcome may be used, including but not limited to irrigation level, crop yield, crop growth or success rate, or user input.
  • the process may be supervised, partly supervised or unsupervised in some embodiments.
  • each actuatable device 14, 16 may be provided with power independently of the irrigation controller 10 and each actuatable device 14, 16 may be configured for wired or wireless communications with the irrigation controller 10 so that the irrigation controller 10 may communicate with the actuatable devices 14, 16 so as to control the actuatable devices 14, 16 e.g. turn on and off, or open and close the actuatable devices 14, 16.
  • the irrigation controller 10 may be located, at least in part, remotely from the irrigation apparatus 4 and the sensors 12.
  • the irrigation controller 10 may be configured for communication with at least some of the sensors 12 and the actuatable devices 14, 16 of the irrigation apparatus 4 via the cloud 28.

Landscapes

  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Water Supply & Treatment (AREA)
  • Environmental Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Soil Sciences (AREA)
  • Feedback Control In General (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

La présente invention concerne un dispositif de commande permettant de commander un appareil d'irrigation, le dispositif de commande comprenant une mémoire stockant un modèle d'irrigation et une ressource de traitement, la ressource de traitement étant conçue pour : déterminer un premier programme d'irrigation pour l'appareil d'irrigation pendant une première période sur la base du modèle d'irrigation ; commander l'appareil d'irrigation de telle sorte que l'appareil d'irrigation fournit de l'eau à un ou plusieurs emplacements à l'intérieur d'une zone d'irrigation pendant la première période selon le premier programme d'irrigation ; recevoir des données d'entrée associées à la première période, comprenant au moins des données d'entrée d'utilisateur qui représentent une évaluation par un utilisateur de l'irrigation fournie dans la zone d'irrigation pendant la première période ; ajuster le modèle d'irrigation sur la base, au moins en partie, des données d'entrée reçues associées à la première période pour obtenir un modèle d'irrigation ajusté ; et déterminer un second programme d'irrigation pour l'appareil d'irrigation pendant une seconde période consécutive à la première période sur la base du modèle d'irrigation ajusté.
PCT/EP2020/082742 2019-11-22 2020-11-19 Dispositif de commande, système et procédé d'irrigation WO2021099497A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
GB2208397.6A GB2605321A (en) 2019-11-22 2020-11-19 Irrigation controller, system and method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GBGB1917082.8A GB201917082D0 (en) 2019-11-22 2019-11-22 Irrigation controller, system and method
GB1917082.8 2019-11-22

Publications (1)

Publication Number Publication Date
WO2021099497A1 true WO2021099497A1 (fr) 2021-05-27

Family

ID=69137235

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2020/082742 WO2021099497A1 (fr) 2019-11-22 2020-11-19 Dispositif de commande, système et procédé d'irrigation

Country Status (2)

Country Link
GB (2) GB201917082D0 (fr)
WO (1) WO2021099497A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022198274A1 (fr) * 2021-03-25 2022-09-29 Central Queensland University Système et procédé de distribution d'eau
EP4353076A1 (fr) * 2022-10-12 2024-04-17 Concare GmbH Système d'irrigation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2196085A2 (fr) * 2008-12-10 2010-06-16 Rain Bird Corporation Contrôleur d'irrigation à réglage automatique doté d'un capteur de température et de pluviosité
US20140236868A1 (en) * 2013-02-15 2014-08-21 Banyan Water, Inc. System and method for automated, range-based irrigation
US20170131435A1 (en) 2015-11-05 2017-05-11 Heriot-Watt University Localized weather prediction
WO2018081853A1 (fr) * 2016-11-02 2018-05-11 The Yield Technology Solutions Pty Ltd Commande de zones de production agricoles
WO2018217170A1 (fr) * 2017-05-24 2018-11-29 Remote Grid Pte. Ltd. Système, procédé et appareil de gestion de ressource agricole

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2196085A2 (fr) * 2008-12-10 2010-06-16 Rain Bird Corporation Contrôleur d'irrigation à réglage automatique doté d'un capteur de température et de pluviosité
US20140236868A1 (en) * 2013-02-15 2014-08-21 Banyan Water, Inc. System and method for automated, range-based irrigation
US20170131435A1 (en) 2015-11-05 2017-05-11 Heriot-Watt University Localized weather prediction
WO2018081853A1 (fr) * 2016-11-02 2018-05-11 The Yield Technology Solutions Pty Ltd Commande de zones de production agricoles
WO2018217170A1 (fr) * 2017-05-24 2018-11-29 Remote Grid Pte. Ltd. Système, procédé et appareil de gestion de ressource agricole

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
GUO CEN ET AL: "A Data-Driven Real-Time Irrigation Control Method Based on Model Predictive Control", 2018 IEEE CONFERENCE ON DECISION AND CONTROL (CDC), IEEE, 17 December 2018 (2018-12-17), pages 2599 - 2604, XP033503684, DOI: 10.1109/CDC.2018.8619549 *
JAUME CASADESS ET AL: "A general algorithm for automated scheduling of drip irrigation in tree crops", COMPUTERS AND ELECTRONICS IN AGRICULTURE, ELSEVIER, AMSTERDAM, NL, vol. 83, 8 January 2012 (2012-01-08), pages 11 - 20, XP028472565, ISSN: 0168-1699, [retrieved on 20120114], DOI: 10.1016/J.COMPAG.2012.01.005 *
WALTER, I.A.ALLEN, R.G.ELLIOTT, R.JENSEN, M.E.ITENFISU, D.MECHAM, B.HOWELL, T.A.SNYDER, R.BROWN, P.ECHINGS, S.: "ASCE's standardized reference evapotranspiration equation", WATERSHED MANAGEMENT AND OPERATIONS MANAGEMENT, 2000, pages 1 - 11

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022198274A1 (fr) * 2021-03-25 2022-09-29 Central Queensland University Système et procédé de distribution d'eau
EP4353076A1 (fr) * 2022-10-12 2024-04-17 Concare GmbH Système d'irrigation

Also Published As

Publication number Publication date
GB201917082D0 (en) 2020-01-08
GB2605321A (en) 2022-09-28
GB202208397D0 (en) 2022-07-20

Similar Documents

Publication Publication Date Title
US11341591B2 (en) Environmental management zone modeling and analysis
AU2018317494B2 (en) Method and system for water distribution and soil moisture determination
EP3179319B1 (fr) Procédé de planification d'irrigation et système pour sa mise en oeuvre
Taghvaeian et al. Irrigation scheduling for agriculture in the United States: The progress made and the path forward
Wang et al. Development of a closed-loop irrigation system for sugarcane farms using the Internet of Things
US20140236868A1 (en) System and method for automated, range-based irrigation
Wang et al. Web-based decision support system for canal irrigation management
KR101723173B1 (ko) Ict 기반의 농산물 관리 및 생산 시스템
WO2021099497A1 (fr) Dispositif de commande, système et procédé d'irrigation
CN104521699A (zh) 田间智能灌溉在线控制管理方法
KR20190013073A (ko) 일체형 스마트 작물 관리 모듈
US20180039243A1 (en) Systems and Methods For Site-Based Irrigation Control
EP3986107A1 (fr) Système pour optimiser l'utilisation de l'eau dans l'irrigation sur la base du calcul prédictif du potentiel hydrique du sol
US20230139920A1 (en) Method for estimating precipitation distribution for a geographical region
KR102374864B1 (ko) 영농형 태양광 발전 시스템의 관리 장치, 방법 및 시스템
Winkler et al. Plug-and-play irrigation control at scale
CN113168599A (zh) 植物的栽培管理系统以及植物的栽培管理装置
Morchid et al. High-technology agriculture system to enhance food security: A concept of smart irrigation system using Internet of Things and cloud computing
Zapata et al. Field test of an automatic controller for solid-set sprinkler irrigation
Chen et al. Genotype× environment interactions for phenological adaptation in narrow-leafed lupin: a simulation study with a parameter optimized model
Subathra et al. Automated intelligent wireless drip irrigation using ANN techniques
Coates et al. Wireless sensor network for precision irrigation control in horticultural crops
Sun et al. A review of smart irrigation decision support systems
Pajany et al. Enhancing Irrigation Efficiency with AI-Based Instinctive Irrigation System (IIS) in Wireless Sensor Networks
JP2021056572A (ja) 作物生産支援プログラム、作物生産支援方法および作物生産支援装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20810954

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 202208397

Country of ref document: GB

Kind code of ref document: A

Free format text: PCT FILING DATE = 20201119

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20810954

Country of ref document: EP

Kind code of ref document: A1