WO2020051650A1

WO2020051650A1 - System and method for automated phenology analysis

Info

Publication number: WO2020051650A1
Application number: PCT/AU2019/050991
Authority: WO
Inventors: Kim George KHOR
Original assignee: SnapTrap Pty Ltd
Priority date: 2018-09-14
Filing date: 2019-09-16
Publication date: 2020-03-19

Abstract

A system for automated phenology analysis is described comprising an insect detection module configured to detect and generate insect data relating to insects within a detection area; a plurality of sensors and/or sources configured to detect or receive contextual data regarding current and/or future environmental conditions and data representing an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species and either or both of the contextual data and the insect data, the model predicting the future lifecycle stages of the at least one target insect species; and wherein based on the analysis model, a course of action is performable. The system is, particularly, suitable for the automated control of insect species taking into account environmental conditions, in the form of contextual data, and by capturing insect data in a particular area, such as a trap.

Description

System and method for automated phenology analysis

Technical Field

[0001 ] The present disclosure relates to a method and a system for automated phenology analysis and, in some embodiments, automated action as a result of the automated phenology analysis. It has particular, though not exclusive, application to the protection of horticultural crops from insect damage.

Background

[0002] Worldwide, loss of crops and crop damage, by insects is estimated at roughly 12%+, or US$2 trillion. Fruit flies alone cost Australian horticulture approximately US$300 million per year. Crop damage by insects traditionally forces the reliance on chemical control methods, typically insecticides, by the fruit industry. However, economic, environmental and social consequences of traditional controls are necessitating the exploration of alternative options.

[0003] Methods for reducing the insect population can be classified as: attractive elimination techniques; biological elimination techniques; or chemical elimination techniques. Attractive elimination involves the process of eradicating insect pests via luring the insects into a trap. Traps may be broadly defined as including electrocution grids, adhesive traps, UV light traps, flying insect airflow traps, and terrestrial and aquatic arthropod traps.

[0004] Biological elimination techniques include the sterile insect technique which is a biological control method that assists in reducing broad spectrum chemical treatments. Typically fruit flies are reared in the millions, sterilised, and then the males released into target areas. Subsequent mating results in infertile egg laying. However to be effective, there needs to be a high ratio of sterile to wild males and the sterile males must be applied over vast areas of land.

[0005] Chemical elimination techniques to minimise crop loss include the controlled discharge of pheromones or parapheromones. Such substances are typically synthesized airborne molecules that activate a behavioural response in an insect to cause mating disruption, aggregation or dispersion. Other chemical elimination techniques include insecticides, such as spraying over a broad area, and can also include an attraction technique, such as spraying both a lure and an insecticide at a perimeter of a crop. [0006] Any discussion of documents, acts, materials, devices, articles and the like in this specification is included solely for the purpose of providing a context for the present invention. It is not suggested or represented that any of these matters formed part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed in Australia or elsewhere before the priority date of each claim of this application.

[0007] It is to be understood that, throughout the description and claims of the specification, the word "comprise" and variations of the word, such as "comprising" and "comprises", is not intended to exclude other additives, components, integers or steps.

Summary

[0008] According to a first aspect of the present invention there is provided, a device, comprising: a memory to store instructions; and a processor to execute the instructions to: receive, from a suitable insect detection module, insect data regarding detected insects receive contextual data regarding current and/or future environmental conditions; retrieve data relating to an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species and either or both of the contextual data and the insect data, the model predicting the future lifecycle stages of the at least one target insect species; determine, based on the analysis model, a course of action.

[0009] In some embodiments, the phenology model includes data relating to growing degree days (GDD).

[0010] In some embodiments, the processor further execute the instructions to determine if the contextual data and/or the insect data correlates to the progression of the analysis model, and if the contextual data and/or the insect data does not correlate to the progression of the analysis model then optimising the analysis model to produce a refined analysis model using the contextual data and/or the insect data.

[001 1 ] In some embodiments, the insect data includes one or more of the following: number; lifecycle stage (maturity); species; subspecies; cohort group; gender; nourishment status and behavioural characteristics , behaviour characteristics including mating behaviour; foraging; and/or feeding behaviour.

[0012] In some embodiments, the contextual data includes data concerning one or more of the following: weather forecast; temperature; humidity; air pressure; solar radiation; leaf wetness; photosynthetically-active radiation (PAR); wind; oxygen; carbon dioxide; other gas or chemical sensors; sap flow; light; plant water potential; leaf area meters; canopy analyzers; dendrometers; radiometers; spectrometers; fluorescence; reflectance; plant hydraulic conductance; photosynthesis; plant temperature; root scanning; particulate; and/or infra-red radiation.

[0013] In some embodiments, once a course of action has been determined, the processor is operable to execute instructions to initiate the course of action.

[0014] In some embodiments, the course of action comprises a control technique selected from one or more of the following: initiating a control technique of one or more insect species; initiating a control technique for a product.

[0015] In some embodiments, the control technique of the or each insect species is selected from one or more of the following: pheromone discharge; parasitoids discharge, parasitoid and predator lure substance discharge, semiochemical discharge, parasitoid bacterial symbiont discharge, entomopathogen discharge, baits (lure + toxicant) discharge; and insecticide discharge.

[0016] In some embodiments, the initiation of the course of action is at an optimum time, calculated according to the analysis model and, preferably, in the future.

[0017] In some embodiments, the processor executes instructions to operate a module associated with the control technique, preferably, a discharge module.

[0018] According to a second aspect of the present invention, there is provided, a method, comprising: receiving from a suitable insect detection module, insect data regarding detected insects; receiving contextual data regarding current and/or future environmental conditions; retrieving data relating to an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species; and determining, based on the analysis model, a course of action.

[0019] In some embodiments, the phenology model includes data relating to growing degree days (GDD).

[0020] In some embodiments, the second aspect further comprising: determining if the contextual data and/or the insect data correlates to the progression of the analysis model, and if the contextual data and/or the insect data does not correlate to the progression of the analysis model then optimising the analysis model to produce a refined analysis model using the contextual data and/or the insect data.

[0021 ] In some embodiments, the insect data includes one or more of the following: number; lifecycle stage (maturity); species; subspecies; cohort group; gender; nourishment status and behavioural characteristics , behaviour characteristics including mating behaviour; foraging; and/or feeding behaviour.

[0022] In some embodiments, the contextual data includes data concerning one or more of the following: weather forecast; temperature; humidity; air pressure; solar radiation; leaf wetness; photosynthetically-active radiation (PAR); wind; oxygen; carbon dioxide; other gas or chemical sensors; sap flow; light; plant water potential; leaf area meters; canopy analyzers; dendrometers; radiometers; spectrometers; fluorescence; reflectance; plant hydraulic conductance; photosynthesis; plant temperature; root scanning; particulate; and/or infra-red radiation.

[0023] In some embodiments, the second aspect further comprising initiating the course of action, once it has been determined.

[0024] In some embodiments, the course of action comprises a control technique selected from one or more of the following: initiating a control technique of one or more insect species; initiating a control technique for a product.

[0025] In some embodiments, the control technique of the or each insect species is selected from one or more of the following: pheromone discharge; parasitoids discharge, parasitoid and predator lure substance discharge, semiochemical discharge, parasitoid bacterial symbiont discharge, entomopathogen discharge, baits (lure + toxicant) discharge; and insecticide discharge.

[0026] In some embodiments, the second aspect further comprising initiating the control technique at an optimum time, calculated according to the analysis model and, preferably, in the future.

[0027] In some embodiments, further comprising operating a module associated with the course of action, preferably, a discharge module.

[0028] According to a third aspect of the present invention there is provided a system for automated phenology analysis comprising: an insect detection module configured to detect and generate insect data relating to insects in the detection area; a plurality of sensors and/or sources configured to detect or receive contextual data regarding current and/or future environmental conditions and data representing an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species and either or both of the contextual data and the insect data, the model predicting the future lifecycle stages of the at least one target insect species; and wherein based on the analysis model, a course of action is performable.

[0029] In some embodiments, the phenology model includes data relating to growing degree days (GDD).

[0030] In some embodiments, the system determines if the contextual data and/or the insect data correlates to the progression of the analysis model, and if the contextual data and/or the insect data does not correlate to the progression of the analysis model then optimising the analysis model to produce a refined analysis model using the contextual data and/or the insect data.

[0031 ] In some embodiments, the insect data includes one or more of the following: number; lifecycle stage (maturity); species; subspecies; cohort group; gender; nourishment status and behavioural characteristics , behaviour characteristics including mating behaviour; foraging; and/or feeding behaviour.

[0032] In some embodiments, the contextual data includes data concerning one or more of the following: weather forecast; temperature; humidity; air pressure; solar radiation; leaf wetness; photosynthetically-active radiation (PAR); wind; oxygen; carbon dioxide; other gas or chemical sensors; sap flow; light; plant water potential; leaf area meters; canopy analyzers; dendrometers; radiometers; spectrometers; fluorescence; reflectance; plant hydraulic conductance; photosynthesis; plant temperature; root scanning; particulate; and/or infra-red radiation.

[0033] In some embodiments, once a course of action has been determined, the system initiates the course of action.

[0034] In some embodiments, the course of action comprises a control technique selected from one or more of the following: initiating a control technique of one or more insect species; initiating a control technique for a product.

[0035] In some embodiments, the control technique of the or each insect species is selected from one or more of the following: pheromone discharge; parasitoids discharge, parasitoid and predator lure substance discharge, semiochemical discharge, parasitoid bacterial symbiont discharge, entomopathogen discharge, baits (lure + toxicant) discharge; and insecticide discharge.

[0036] In some embodiments, the initiation of the course of action is at an optimum time, calculated according to the analysis model and, preferably, in the future.

[0037] In some embodiments, the system operates a module associated with the course of action, preferably, a discharge module.

[0038] In some embodiments, including in all aspects of the invention, the insect detection module comprises one or more of the following: a camera; wing-beat detection module; and/or a bio-impedance detection module.

[0039] In some embodiments, including in all aspects of the invention, a discharge module is operable to discharge a control techniques, such as pheromone dispersal .

[0040] In some embodiments, including in all aspects of the invention, a taggant illumination and/or detection system is included as part of the insect detection module, for the purposes of identifying tagged insects in the insect data.

[0041 ] In some embodiments, including in all aspects of the invention, an insect trap is included, at least a portion of which includes the detection area of the insect detection module.

[0042] In some embodiments, including in all aspects of the invention, the trap may further include a fan configured to distribute a scent.

[0043] In some embodiments, including in all aspects of the invention, the insect trap may further include a trap door to drop carcasses into a specimen container, or to the ground and a vacuum to suck the carcasses from the specimen container. Still further, the insect trap may include means to enable volumetric or mass measurement of the carcasses. For instance the insect trap housing, or specimen container, may include a sightglass or dipstick and a weight measurement sensor. Furthermore, the insect trap may include a automated genetic analyser, which acts to process insect remains and provide a genetic identification.

[0044] The insect trap may further include a recharge system to replenish the lure and/or toxicant. Still further, the insect trap housing may include a sensor configured to measure the amount of lure or toxicant in the active area of the trap. [0045] Embodiments of the invention have particular application to the protection of horticultural/agriculture crops from insect damage. Furthermore, the invention has application to provide audit trail evidence to support market access negotiations for horticulture produce marketers and to control techniques for horticultural/agriculture crops based on the prevalence of certain insect species, including, for example, beneficial insect species.

Brief Description of the Drawings

[0046] In order that the present invention may be more clearly ascertained, embodiments will now be described, by way of example, with reference to the accompanying drawings in which:

[0047] Figure 1 is a schematic diagram illustrating one embodiment of a system for automated phenology analysis incorporating a singular master device;

[0048] Figure 2 is a flow chart of an example process for causing a course of action in relation to control of a target insect species;

[0049] Figure 3 illustrates the modelling of the growing degree days over which is mapped the stages of an insects lifecycle; and

[0050] Figure 4 is a schematic diagram of an embodiment of a wide area analysis depicting several multiple master/slave devices.

Detailed Description

[0051 ] With reference to figure 1 , there is shown a schematic diagram of one embodiment of a system 100 to enable automated phenology analysis, which incorporates a singular master device. The system includes an insect trap 105 having an opening (not shown) through which insects can enter the interior of the trap. The housing is configured to house a lure 1 10, a disabler 1 15, in the form of a toxicant designed to kill insects so that the insects remains in situ to be observed, an insect detection module in the form of a camera 120 to detect characteristics of trapped insects, a plurality of sensing elements 125, a charging device 130, energy storage element 135, a communications module 140, a trigger sensor 160, and a first device, or more commonly referred to as a master device in the form of a control module 145 in data communication with the communications module 140, the output of the camera 120, the output of each of the sensing elements 125, and the trigger sensor 160. A discharge module 165 is operable by the control module 145. The discharge module 165, in this embodiment, is a parasitoid discharge module, releasing parasitoids to control the population of the target insect species. It should be appreciated that the discharge module 165 can be replaced with any suitable action that may be autonomously operated by the control module 145.

[0052] The disabler 1 15, in the above embodiment, is described as being a toxicant, which is a chemical disabler, but it should be recognised that any appropriate disabler may be used, such as other chemical disablers and/or physical disablers, such as an adhesive pad.

[0053] As in the example described above, the insect detection modules can be a camera, and, particularly a digital camera capable of providing digital images locally or remotely for analysis. However, the insect detection module can also include alternatively or additionally, one or more of the following: a wing-beat detection module, such as an acoustic sensor suitable for such purpose; or a bio-impedance detection module. In addition, the insect detection module can be trigger sensors, such as trigger sensors 160, or are supplemented by trigger sensors. Examples of suitable trigger sensors are infra-red beam, ultrasound or optical sensor-trip types.

[0054] The lure 1 10 is able to be any one of the many known lures designed to lure nominated insect types into the trap 105. It will be appreciated that the type of lure will depend upon the desired insect to be trapped. The disabler 1 15 may be a chemical or a mechanical mechanism configure to disable insects. The lure 1 10 and disabler 1 15 are optional components of the system 100, as the trap 105 may not require a lure and the insect detection module may not require the insects to be disabled, although these are likely to be preferred components for most insect species. In particular, if an insect species that is being monitored is a beneficial insect, that is an insect which is beneficial to the growth or reproduction of a crop, or other product (for example, honey), a lure can be used without a disabler, such that the beneficial insect is not harmed. However, in some instances, even beneficial insects may be disabled such that a "sample" of the beneficial insect population can be monitored and used as part of the input to the system.

[0055] As used herein, the term "module" may also refer to any of an application- specific integrated circuit ("ASIC"), an electronic circuit, a processor (shared, dedicated, or group) that executes one or more software or firmware programs, a combinational logic circuit, and/or other suitable components that provide the described functionality.

[0056] The charging device 130, which typically is in the form of solar cell or an inductive charging device, can be mounted to or be an integral part of the insect trap 105, or in some embodiments, can be placed in an advantageous position for sunlight away from the trap 105. Solar cell(s) can be used to create electricity from light whereas an inductive charging device uses electromagnetic induction to create electrical current to charge energy storage element 135. In the latter case, an external charging station would create an alternate magnetic field and would be positioned near the coils of inductive charging device to send electromagnetic energy to the inductive charging device thereby inducing an electrical current within the coils of inductive charging device. Charging device 130 is connected to a rechargeable electrical energy storage element 135 by a line. Energy storage element 135 is typically in the form of a battery. However, energy storage element 135 can also be of other types, such as a capacitive energy storage element. Charging device 130 and energy storage element 135 constitute a power source. Energy storage element 135 is connected by power lines to each sensing element 125, a controller 140 and the camera 120.

[0057] In this example, sensing elements 125 include sensors that sense contextual data regarding environmental conditions, such as temperature, humidity, air pressure, solar radiation, leaf wetness, solar radiation, photosynthetically-active radiation (PAR), location, orientation and wind. Additional sensors may include sensors to measure other contextual data regarding environmental conditions such as oxygen, carbon dioxide or other gas or chemical sensors, sap flow, light, plant water potential, leaf area meters, canopy analyzers, dendrometers, radiometers, spectrometers, fluorescence and reflectance sensors, plant hydraulic conductance meters, photosynthesis, plant temperature sensors, root scanning, infra-red radiation etc. Such sensors can be implemented by any suitable means known in the art.

[0058] In addition, via the communications module, contextual data regarding the environmental conditions can be received from other sources. For example, it may be sufficient to know the measured temperature, air pressure or humidity from a remote source rather than the immediately local data for that information. The communications module can also receive predicted contextual data, such as weather forecasts (rain, wind, temperature, air pressure, hours of sun, etc.).

[0059] Separate models or calculations can also be utilised by the control module

145 to make future predictions about one or more contextual data types to generate contextual data of future environmental conditions.

[0060] Control module 145 includes a microprocessor 150 with several different peripherals such as memory 155, and input and output devices (not shown). Peripherals can include but are not limited to: USB (Universal Serial Bus), I2C (l-squared-C) computer bus, ADC (Analog to Digital Converter), DAC (Digital to Analog Converter), Timers, Pulse Width Modulators, Flash Memory, RAM Memory, EEPROM (Electrically Erasable Programmable Read Only Memory), Bluetooth interface, Ethernet interface. An example of such a control module could be the Texas Instruments TMS320LF28XX family of microcontrollers.

[0061 ] Control module 145 further includes memory 155 which is a readable, writeable memory and may be non-volatile. Programming of the control module 145 may be altered in the field, by erasing and rewriting the program memory via wireless download. The control module 145 functions to process the contextual data captured from each of the sensing elements 125, insect data captured from the insect detection module 120, and contextual, external derived, data received via the communications module 140. Data from the sensing elements or other sources, the insect detection module and the communications module may also be stored in the memory 155 for later retrieval.

[0062] The control module 145 can retrieve from memory 155, or via the communications module 140, a phenology model of a target insect species, such as data relating to the growing degree days (GDD) of that target insect species. In some embodiments, more than one target insect species is configured and the system will either have a separate phenology model for each of the target insect species or, where appropriate, the model will be applicable to each of the target insect species. Alternatively, the control module 145 manages data, particularly data from sensors 125, via the communication module 140 to a remote system which comprises the model or models of the target species and the relevant phenology model.

[0063] In particular, the control module 145 may, via the communications module

145, be operable to receive an analysis model, or a refined analysis model, from a remote system. That is, the control model 145 provides local data, such as insect data and contextual data, to a remote system, such as a remote computer and/or processor, and the remote system applies the local data, after any necessary pre-processing, such as image analysis or the like, to an analysis model of a target insect species. The output of the analysis model being communicated back to the control module 145, at least when necessary for an action to be taken by the control module 145. For example, if the output of the analysis model is such that it is an optimum time for a control technique for a target insect species which is a pest, then the control module 145 may receive communications which result in the control module operating a discharge module, such as releasing a pheromone or a parasitoid. [0064] Alternatively or additionally, the output of the analysis model may provide an alert to an end user, either via the control module 145, such as by operating an indicating light, or by other means, such as by email or text message, indicating that conditions are optimum for an action to be taken, such as a control technique.

[0065] Control module 145 may in addition manage battery charging and also conservation of power by powering down peripherals, and even entering a low power mode (sleep mode) and only exit from the low power mode (wake up) at either certain intervals to check if signals are present, or may wake up due to a signal being presented to peripherals which are capable of waking the controller from a sleep state.

[0066] The communication module 140 may include at least one of a wireless connection; a wired connection; a TCP/IP connection; a removable memory device (e.g. flash card or USB stick); a cellular modem; a WiFi adapter; and a Bluetooth adapter; active and passive RFID.

[0067] The communication module 140 includes an interface for bi-directional data communication to transfer information between the memory 155 of the control module 145, a remote database and an end user computer. Information from the database may then be used for monitoring and report generation. Bi-directional communication enables a user to configure the system 100, for instance to enter the GPS location of the trap 105 deployed in the field, and further enables external data to be uploaded.

[0068] A plurality of trigger sensors 160 are used at the openings of the trap 105 to detect objects entering the trap 105. The control module 145 is further operable to evaluate the data from the trigger sensors 160 to estimate the efficiency of the trap 105 and the number of target insects that escape the trap 105.

[0069] Although not shown, a UV lamp is also present in the trap 105. Sterile insects (“SIT” flies, for example) may be "manufactured" and suppress reproduction success of populations of wild insects. When they are manufactured they are doped with a tag g ant, such as a UV responsive dye or an isotope. In this embodiment, the UV lamp in the trap 105 allows the sterile insects to be identified separately to the wild insect population. The sterile insects, through their identification over the wild insects, can then be tracked geographically and separate in the data to provide more accurate results and models.

[0070] In this way, the insect detection module can provide insect data which differentiates between naturally occurring wild insects and synthetically occurring manufactured insects. This modification, providing a taggant illumination and/or detection system as part of the insect detection module, for the purposes of identifying tagged insects, is applicable to all embodiments.

[0071 ] It should be appreciated that, although the embodiment above, and other embodiments described herein, provide for the use of an insect trap, an insect trap is not essential for all embodiments. The insect detection module may be capable of operating on insects which are not trapped but may be within a particular volume of space, such as a RADAR or LIDAR detector and the system may operate on that basis. Where there is reference to use of a trap in this specification, it should be understood that the same embodiment may operate without a trap.

[0072] The system 100 may be configured to attach time and date stamps to all data being recorded, including both insect data and contextual data.

Methodology

[0073] In use, some embodiments, such as the embodiment described above, have the insect trap 105 configured with a lure and a disabler in the form of a toxicant. In addition or alternatively, the insect trap 105 is capable of operating a control technique, such as an insect disabler discharge (for example by spraying) or release of pheromones. In other embodiments control techniques may include parasitoids, parasitoid and predator lure substances, other semiochemicals, parasitoid bacterial symbionts, entomopathogens, and “attract & kill” baits (lure + toxicant). Each of these may be delivered by various physical media including aerial spray vapour or gas, surface spray fluid, a gel or a similar semi-solid, a solid material that degrades to release the active content, or a cartridge of “emergence habitat”. In most embodiments, these control techniques will act on a pest insect population in a manner such that the population is reduced. Some embodiments may, however, be such that the control technique promotes the insect population or an activity of the insect population, particularly where the target insect species is considered a beneficial insect. For example, bees generally seek either sugar or protein and when bees are fed sugar, they are more likely to seek protein. Accordingly, providing a control technique where sugar is fed to bees causes the bees to be more effective pollinators because they move further into the flower seeking the protein in the pollen, hence carry more pollen around.

[0074] With reference to Figure 2, process 200 includes the control module 145, or other system sources, providing data comprised of phenology model data 210, which may be a standard phenology model. Standard phenology model data typically is composed of hourly temperature data and the mathematical technique of degree-days (DD), otherwise referred to as growing -degree-days (GDD). Alternatively, the phenology model may be derived specifically for the specific target insect species. For example, a phenology model may predict that a target insect species will peak at a pre-determined period of time after detection of a particular population of another insect species in certain contextual conditions. This phenology model can be applied to the system 100. The output of one of an example phenology model will be described later when referring to Figure 3.

[0075] Contextual data from external sources 220, i.e meteorological forecast data and local sources 230, i.e from the various sensing elements 125 and insect data from the insect detection module 120, is provided to the system 200. The data is then pre-processed 240 so that the data is in a suitable format that to refine the phenology model. This may include an initial step of filtering the data to remove extraneous or misleading data, providing image analysis of images, averaging data or other similar processes. The system 200, in this embodiment, uses machine learning techniques to analyse the data and, in the first instance, provide an analysis model 250. The machine learning techniques may include, for example, artificial neural networks, Bayesian statistics, learning automata, decision trees, linear or quadratic classifiers, or the like. Once created, the analysis model, or refined analysis model, is stored in a data structure . As data is continuously being generated, the system 200 is operable to further refine the analysis model based on the receipt of new data (shown by the feedback loop from the analysis model store 260 to the external source data 220. Refinement may occur at pre-defined regular intervals or continuously. For example, it may appropriate to refine the analysis model on a daily or weekly basis.

[0076] Once an analysis model is available, it may be processed either at pre defined regular intervals or continuously. For example, it may be appropriate to apply the inputs to the analysis model every 5 minutes or every 30 minutes. When the analysis model is processed a course of action 280 is generated by the model. The course of action may be to do nothing at the current time, take a specific action at the current time, or take a specific action at a future time. The course of action may be a recommended action, especially if the action is not within the ability of the system 200 to action or an automated action initiated by the system 200. For example, the action may be a control technique for a insect specific, as discussed elsewhere in this specification, a control technique for a target crop (such as watering), or an alert to a third party (such as another system or a user of the system 200).

[0077] It should be appreciated that the system 200 is capable of being operated locally to an appropriate insect trap, such as by the control module 145, or remotely in, for example, a remote computer or cloud based system. Where the system 200 is operated wholly remotely, the control module 145 acts as a communication conduit between the local contextual and insect data and any actions it is operable to take, and the system 200. The system 200 is also capable of being operable across both a local system, where the insect trap is based, and a remote computer, with each system having particular tasks to undertake and communicating with the other appropriately. For example, refinement of the analysis model may take place remotely but all other steps are executed locally.

[0078] Figure 3 illustrates a phenology model being based on a growing degree days model over which is mapped the stages of an insects lifecycle. In this case, data has been obtained for a particular trap identified as MD01 , 305, and the data has been collected over a five month window 310. Users of this particular model are able to select a weather station 315, and such users may be able to obtain weather forecast data to extrapolate the progression of the model into the future, hence providing a predictive calculation as to the timing of the next lifecycle event. Users may also toggle between obtaining data from locally positioned physical instrumentation or alternatively choose to download data from one or more online sources.

[0079] This model relates to the Andrew Jessup model which depicts four stages in the development threshold of an insect's lifecycle. Stage 1 , 320, is representative of the stage bounded by teneral adulthood and oviposition (exclusion from the pupa to depositing eggs, in fruit for instance). Stage 2, 325, is representative of that stage bounded by oviposition and egg hatching (referred to as the first instar), when the larvae start feeding on the fruit. Stage 3, 330, is representative of the first instar and pupation which consists of a number of sub-stages in which the larvae increase in size and successively shed their shells. Stage 4, 335, is representative of the stage from pupa (the final larval stage) to teneral adult. In this stage, and using the example given, the pupae leave the fruit and fall to the ground where they then burrow to take refuge beneath the surface. The cycle then starts again, when the teneral adults emerge from the pupal form. In the graph depicted, stage 1 is fleeting (of the order of a day due to the magnitude of the accumulated heat).

[0080] The dashed line 340, represents the growing degree days, or the magnitude of heat being accumulated during over time. According to the Andrew Jessup model, the daily temperature curve can be described as a temperature based insect dose-response curve. According to this reasoning, the area under the temperature curve over time gives a measure of how far the insect has developed over time.

[0081 ] With regard to the mapping, label Ί 7' (24 July) represents the interface between stages 1 and 2, label '81 ' represents the interface between stages 2 and 3, label 224' represents the interface between stages 3 and 4 and label '82' represents the interface between stages 4 and 1 . The interfaces between the respective stages, represent periods in which the insect is at its most vulnerable. For instance the interface depicted at Ί 7' is the time when the eggs are laid. If the eggs are laid on the surface of a leaf, the system 200 can seize the opportunity to release parasitoids or predators to coincide with this occurrence. In the case where eggs are deposited inside fruit, an insecticide may be released or sprayed, or an alarm indicator provided for manual insecticide application. In contrast, at the interface depicted at '224' the pupae are falling to the ground. The timing is now ripe for the release of chickens to consume the pupae. At the interface‘82’, teneral adults should emerge with the need to forage for food, and this behavioural requirement may be exploited with baits that mimic a food source. . However, the analysis model in this case provides that, even though the original growing degree days model predicts sufficient growth, mating conditions were not met, based on contextual data such as cloud cover and air pressure gradient. Hence the analysis model is such that progression of the original growing degree days model accumulation is paused at this point, until mating conditions are met and suitable actions in relation to mating is delayed accordingly. Calculation of the growing degree days may be calculated by numerous techniques, such as, but not limited to calculating a rolling average of temperatures, finite element analysis or integration.

[0082] The data from the sensing elements 125 is recorded at hourly intervals, though it should be appreciated that the frequency at which data is recorded may be modified depending on the objective. Data from the insect detection module 120 is generally taken at a longer interval, for instance two hourly during daytime and six hourly during night.

Control work decision information

[0083] As mentioned, each lifecycle stage presents vulnerability in the insect that can be exploited for control work. The control module 145 is operable to determine, using the model and at least a subset of the contextual data, a recommended course of action, in respect of which control technique is relevant for the lifecycle stage. Furthermore, the control module is operable to accurately identify the appropriate timing for executing the control work. Furthermore, using the model and at least a subset of the sensor and external data, the control module 145 is operable to determine suggestions for alternatives of control techniques, any additional supplies that may be necessary as a result of the alternative technique, and the frequency of application of the alternative control technique.

[0084] In addition, and utilising the meteorological forecasting information, the control module 145 is operable to determine a forecast as to when advancement to a subsequent lifecycle stage is likely to occur. This provides significant advantages in workflow planning and waste reduction for operators.

Reporting

[0085] The control module 145 is operable to generate a summary and detailed reporting of observations and analyses. The generated report may be transmitted using the communication module 140 to a remote database or end user computer in order to for instance inform immediate work practice decisions, inform policy decisions, economic and budget planning for primary producers, grower associations, marketers and trade negotiators, facilitate compliance reporting and alerts (biosecurity and export protocol compliance), and to enable control work execution.

[0086] Once the control module 145 has identified the appropriate timing for control work, and the type of control work, the control module 145 may be further configured to autonomously execute the control work.

Control techniques

[0087] The control work may for instance be the discharge of pheromones or parapheromones which are used to activate a behavioural response in the insect. Generally they are airborne and affect an olfactory response in the insect. The three basic types of pheromone and the response they invoke are: a) Mating: the insects are grouped by gender, and one gender (generally the female) emits a mating pheromone to provide the other gender the ability to locate a viable mating partner. Discharging a synthesized mating pheromone, en masse, distracts the responsive insect from the true source of the naturally emitted mating pheromone, and destroys its ability to locate the viable mating partner insect. This is an established technique known as“mating disruption” or“mating confusion” b) Aggregation: the insects are led to believe that their counterparts are signalling for them to gather because a beneficial circumstance exists, such as food availability. c) Dispersion or Stress-response: the insects are led to believe that they should disperse and leave the proximal area because their counterparts are signalling the presence of danger or unfavourable conditions.

[0088] There are other types outside of these three main types, and species-specific controls techniques might exploit another type of pheromone. [0089] Other control techniques can be executed, such as the use of baiting (i.e attract & kill). In various physical circumstances different techniques might be chosen. In a greenhouse where the atmosphere is quite controllable, lowering the humidity and increasing temperature can be effected to suppress an insect behaviour.

[0090] Whilst the foregoing has referred to the autonomous application of control techniques, it should be appreciated that control techniques may be applied semi- autonomously i.e., requiring human approval. Human approval may be used for assurance, and for improvement of methods or algorithms. For assurance, manual observations might be used to confirm the accuracy of the system, in matters such as timing or frequency. For improvement, system decisions would be compared with manual observations, to develop improved or new logic. Some situations might exclude fully autonomous decisions for a regulation or safety issue.

Slave Device

[0091 ] The system may further comprise a slave device, wherein the slave device includes a memory to store instructions, a bi-directional communications module to enable communication of instructions from the first device (otherwise referred to as the master device), and a processor in communication with the communications module and operable to execute the instructions to: (i) receive data from an environmental sensor, and (ii) discharge a payload at a specified time and frequency.

[0092] In the instance where the control technique is the discharge of pheromones, the slave device is equipped with an ampule of the pheromone payload, a discharge device comprised of a pump and atomizer (or vapouriser, or mister) and a communications module in the form of a radio to enable bi-directional data communication between the master device and the slave device. The slave device is also equipped with environmental sensors that inform the correct deployment of the payload. These include wind direction sensors and solar radiation sensors. This sensor data is reported back to the master device via the communications radio, along with information recording discharge actions taken.

[0093] Data from the wind sensor will assist in determining the physical vectoring of the vapour mist when discharged, as the mist will be carried by the wind. If the wind direction is unfavourable, the processor will halt the discharge of the payload until the wind direction is favourable. A solar radiation sensor is desirable in this application since pheromone molecules are often photo-sensitive. That is, they will degrade prematurely in disadvantageous sunlight conditions. Obtaining solar radiation data assist in the processor determining whether the discharge should proceed or be withheld until conditions are favourable.

[0094] Similarly, baits that include toxicants are often photosensitive and their efficacy is affected by wind. The sensing of conditions by the slave device allows the system to optimize the discharge timing for favourable conditions.

[0095] The design of the network incorporating a master device and a slave device is optimally configured to accommodate and optimise local conditions. Knowledge of the target insect, the prevailing winds, water sources and food sources is beneficial to the design. For example a baiting method for a particular insect might be best situated in a food source that is positioned away from the crops to be protected, and upwind from the crop to be protected. Hence the attractant lure will be naturally carried downwind towards the protected crop, indicating that the desired location is upwind, and away from the protected crop. The food source will reinforce the desirability of the alternative location to the insect. Once the insect arrives at the alternative location it will ingest (or otherwise take up) the toxicant and expire.

[0096] Information from each of the sensing elements is stored in the memory of the control module 145. Again, this information can also be transmitted using the communication module, such as a cellular modem (or other wired or wireless communications module), providing data to a remote database or end user computer such as a cell phone, Smartphone, laptop, PC, server, or other computing device.

[0097] Whilst the system has been described in relation to a singular master/slave device, it should be appreciated that multiple units could be deployed to enable a wide area analysis as generally depicted in Figure 4. The system illustrated in Figure 4 includes area analysis is a hub 510, one or more master devices which may exist in isolation 520, and one or more master devices 525 each integrated with a respective slave device 530. Additionally, there may be one or more slave devices 535 which exist in isolation and which include sensors. Each of the master devices 520, 525 and slave devices 535 are configured for bidirectional communication with the cloud based hub 510 via communication channels, such as radio communications. Further included in the greater system are user devices 540 such as mobile phones and tablets which may be configured to carry out review, analysis and control functionalities. It is envisioned that data obtained from the master and/or slave devices may in certain circumstances be sent to external systems 500, for instance systems associated with farm management, biosecurity and government agencies. For example if an insect is detected in a trap, the system may respond by raising a“red flag” which would be sent to the supply chain as a compliance action relating to a trade agreement. As another example, when a lifecycle stage is approaching, the system may be configured to send an alert to the farm management software that assists with planning intervention work to coincide with the lifecycle change.

[0098] The hub 510 includes a central repository which houses data related to growing degree days, further, the hub 510 may have functionality to carry out data analysis, processing and reporting functions.

[0099] The processor of each of the master devices 525 creates a predictive model using data obtained from the hub. The hub 510 is operable to send instructions via the communication channels to the slave devices 535, or 530 via main devices 525 to take appropriate action. For instance, one action may be for any particular slave device to spray pheromone at pre-determined times, another option may be to pause, or advance the predictive model as a result of an identified local variance in the insect's behaviour.

[00100] Thus it should be appreciated that automated decisions may be made at the master device, slave device, and/or hub and where exactly decisions are made will depend on the situations Some situations will be complex, and some simple, and this complexity will have bearing on which part of the system will make the decision.

Machine learning

[00101 ] Machine learning is also able to be used to determine patterns in the incidence of insect appearance over time. The control module's processor 145 may be configured, such as through machine learning of images, where a camera is used as the insect detection module, to recognize the target insect, determines age and gender, and maps the intervals amongst insect appearances, amongst other aspects of the insect. The system can record counts of insects and provides for numerical analyses such as overall “pest pressure” over time.

[00102] These local observations then contribute to wide-area analyses of pest pressure and behaviour patterns. Localised variations (or adaptions) of the single species may be observed in multiple locations. This correlation of variations is used to further refine the reasoning and factors for predicting local variations and adaptions by that species.

[00103] Whilst in some examples use of the insect detection module 120 may be enough to determine the type, timing and number of insects at a given point in time, human observation and record data entry may optionally, or additionally be performed to record the timing, type and number of insects. In the case of the latter, record data will be uploaded to the control module.

[00104] The system 100 is able to use the meteorological forecast data to estimate the occurrence of future lifecycle stages. Moreover, data from the insect detection module 120 provides behavioural factors. For example, a particular insect may perform mating behaviour only under certain conditions, such as threshold temperature at dusk. If these conditions are not met then the progression of the model must be paused until the dusk mating conditions are met. By way of a further example, the foraging for food may be affected by the extent of solar radiation and so the progression of the model must be paused until the foraging conditions are met.

Pest pressure

[00105] Quantum numbers of particular insects over a period of time indicates the prevalence (pest pressure) imposed by that insect’s presence. A measure of the pest pressure informs the control work decisions. In conditions of higher counts, an intervention might be performed more times or with more intensity.

[00106] In addition, the system is capable of identifying attrition of insects once they appear in the trap. Instances of attrition may for instance be a result of ants consuming the carcasses of trapped insects or due to the action of wind vortices. The system is configured to track such sources of attrition in order to refine the observation of pest pressure. This is achieved using the insect detection module and/or the trigger sensors.

Claims

The claims defining the invention are as follows:

1 . A device, comprising: a memory to store instructions; and a processor to execute the instructions to:

receive, from a suitable insect detection module, insect data regarding detected insects

receive contextual data regarding current and/or future environmental conditions; retrieve data relating to an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species and either or both of the contextual data and the insect data, the model predicting the future lifecycle stages of the at least one target insect species; determine, based on the analysis model, a course of action.

2. The device of claim 1 , wherein the phenology model includes data relating to growing degree days (GDD).

3. The device of claim 1 or 2, where the processor further execute the instructions to determine if the contextual data and/or the insect data correlates to the progression of the analysis model, and if the contextual data and/or the insect data does not correlate to the progression of the analysis model then optimising the analysis model to produce a refined analysis model using the contextual data and/or the insect data.

4. The device of any one of claims 1 to 3, wherein the insect data includes one or more of the following: number; lifecycle stage (maturity); species; subspecies; cohort group; gender; nourishment status and behavioural characteristics , behaviour characteristics including mating behaviour; foraging; and/or feeding behaviour.

5. The device of any one of claims 1 to 4, wherein the contextual data includes data concerning one or more of the following: weather forecast; temperature; humidity; air pressure; solar radiation; leaf wetness; photosynthetically-active radiation (PAR); wind; oxygen; carbon dioxide; other gas or chemical sensors; sap flow; light; plant water potential; leaf area meters; canopy analyzers; dendrometers; radiometers; spectrometers; fluorescence; reflectance; plant hydraulic conductance; photosynthesis; plant temperature; root scanning; particulate; and/or infra-red radiation.

6. The device of any one of claims 1 to 5, wherein once a course of action has been determined, the processor is operable to execute instructions to initiate the course of action.

7. The device of any one of claims 1 to 6, wherein the course of action comprises a control technique selected from one or more of the following: initiating a control technique of one or more insect species; initiating a control technique for a product.

8. The device of claim 7, wherein the control technique of the or each insect species is selected from one or more of the following: pheromone discharge; parasitoids discharge, parasitoid and predator lure substance discharge, semiochemical discharge, parasitoid bacterial symbiont discharge, entomopathogen discharge, baits (lure + toxicant) discharge; and insecticide discharge.

9. The device of any one of claims 6 to 8, wherein the initiation of the course of action is at an optimum time, calculated according to the analysis model and, preferably, in the future.

10. The device of claim 9, wherein the processor executes instructions to operate a module associated with the control technique, preferably, a discharge module.

1 1 . A method, comprising:

receiving from a suitable insect detection module, insect data regarding detected insects;

receiving contextual data regarding current and/or future environmental conditions; retrieving data relating to an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species; and

determining, based on the analysis model, a course of action.

12. A method according to claim 1 1 , wherein the phenology model includes data relating to growing degree days (GDD).

13. A method according to claim 1 1 or 12, further comprising:

determining if the contextual data and/or the insect data correlates to the progression of the analysis model, and if the contextual data and/or the insect data does not correlate to the progression of the analysis model then optimising the analysis model to produce a refined analysis model using the contextual data and/or the insect data.

14. A method according to any one of claims 1 1 to 13, wherein the insect data includes one or more of the following: number; lifecycle stage; maturity; species; subspecies; cohort group; gender; nourishment status and behavioural characteristics , behaviour characteristics including mating behaviour; foraging; and/or feeding behaviour.

15. A method according to any one of claims 1 1 to 14, wherein the contextual data includes data concerning one or more of the following: weather forecast; temperature; humidity; air pressure; solar radiation; leaf wetness; photosynthetically-active radiation (PAR); wind; oxygen; carbon dioxide; other gas or chemical sensors; sap flow; light; plant water potential; leaf area meters; canopy analyzers; dendrometers; radiometers; spectrometers; fluorescence; reflectance; plant hydraulic conductance; photosynthesis; plant temperature; root scanning; particulate; and/or infra-red radiation.

16. A method according to any one of claims 1 1 to 15, further comprising initiating the course of action, once it has been determined.

17. A method according to any one of claims 1 1 to 16, wherein the course of action comprises a control technique selected from one or more of the following: initiating a control technique of one or more insect species; initiating a control technique for a product.

18. A method according to claim 17, wherein the control technique of the or each insect species is selected from one or more of the following: pheromone discharge; parasitoids discharge, parasitoid and predator lure substance discharge, semiochemical discharge, parasitoid bacterial symbiont discharge, entomopathogen discharge, baits (lure + toxicant) discharge; and insecticide discharge.

19. A method according to any one of claims 16 to 18, further comprising initiating the control technique at an optimum time, calculated according to the analysis model and, preferably, in the future.

20. A method according to claim 19, further comprising operating a module associated with the course of action, preferably, a discharge module.

21 . A system for automated phenology analysis comprising:

an insect detection module configured to detect and generate insect data relating to insects within a detection area;

a plurality of sensors and/or sources configured to detect or receive contextual data regarding current and/or future environmental conditions and

data representing an analysis model of the lifecycle stages of at least one target insect species, the analysis model being based on and/or having inputs including data relating to a phenology model of the at least one target insect species and either or both of the contextual data and the insect data, the model predicting the future lifecycle stages of the at least one target insect species; and

wherein based on the analysis model, a course of action is performable.

22. A system as claimed in claim 21 , wherein the phenology model includes data relating to growing degree days (GDD).

23. A system as claimed in claim 21 or 22, wherein the system determines if the contextual data and/or the insect data correlates to the progression of the analysis model, and if the contextual data and/or the insect data does not correlate to the progression of the analysis model then optimising the analysis model to produce a refined analysis model using the contextual data and/or the insect data.

24. A system according to any one of claims 21 to 23, wherein the insect data includes one or more of the following: number; lifecycle stage (maturity); species; subspecies; cohort group; gender; nourishment status and behavioural characteristics , behaviour characteristics including mating behaviour; foraging; and/or feeding behaviour.

25. A system according to any one of claims 21 to 24, wherein the contextual data includes data concerning one or more of the following: weather forecast; temperature; humidity; air pressure; solar radiation; leaf wetness; photosynthetically-active radiation (PAR); wind; oxygen; carbon dioxide; other gas or chemical sensors; sap flow; light; plant water potential; leaf area meters; canopy analyzers; dendrometers; radiometers; spectrometers; fluorescence; reflectance; plant hydraulic conductance; photosynthesis; plant temperature; root scanning; particulate; and/or infra-red radiation.

26. A system according to any one of claims 21 to 25, wherein once a course of action has been determined, the system initiates the course of action.

27. A system according to any one of claims 21 to 26, wherein the course of action comprises a control technique selected from one or more of the following: initiating a control technique of one or more insect species; initiating a control technique for a product.

28. A system according to claim 27, wherein the control technique of the or each insect species is selected from one or more of the following: pheromone discharge; parasitoids discharge, parasitoid and predator lure substance discharge, semiochemical discharge, parasitoid bacterial symbiont discharge, entomopathogen discharge, baits (lure + toxicant) discharge; and insecticide discharge.

29. A system according to any one of claims 26 to 28, wherein the initiation of the course of action is at an optimum time, calculated according to the analysis model and, preferably, in the future.

30. A system according to claim 29, wherein the system operates a module associated with the course of action, preferably, a discharge module.

31 . A system according to any one of claims 21 to 30, wherein the insect detection module comprises one or more of the following: a camera; wing-beat detection module; and/or a bio-impedance detection module.

32. A system according to any one of claims 21 to 31 further comprising an insect trap at least a portion of which includes the detection area of the insect detection module.