Information processing system, information processing method, and program

Abstract

To predict the distribution of a diffusion substance at low cost and with high accuracy.SOLUTION: The prediction distribution generation part generates a prediction distribution which is a future distribution of the diffusion material by performing data assimilation of a diffusion equation including a characteristic parameter which is a parameter representing a characteristic of increase or attenuation of the diffusion material by using sensor data related to the diffusion material. The present technology can be applied to, for example, an information processing system including sensor devices installed at a plurality of points, an input / output device including an input unit that receives an input from a user and a display unit that displays a prediction distribution, and an estimation device that generates the prediction distribution.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present technology relates to an information processing system, an information processing method, and a program, and more particularly to an information processing system, an information processing method, and a program that enable the distribution of a diffused substance to be predicted at low cost and with high accuracy. [Background technology]

[0002] Yellow rust fungi spread through the air and cause yellow rust when they attach to plants. Once yellow rust occurs, it is known to cause devastating damage to fields, making it necessary to predict its arrival. However, the occurrence of yellow rust fungi does not have a seasonal cycle, and there are many unknowns about the path the fungi take to reach fields, making it difficult to predict its arrival.

[0003] Possible methods for predicting the arrival of pathogens such as yellow rust include methods using sensor data and simulations. However, because the range of spread of pathogens is on the order of hundreds to thousands of kilometers, prediction accuracy depends on the number of sensors, and it is difficult to make highly accurate predictions at a realistic cost using sensor data.

[0004] One possible approach using simulations is to use a pathogen diffusion model linked to meteorological data in combination with a pathogen biological model (which estimates proliferation, death, and infection). However, simulations using a pathogen diffusion model linked to meteorological data are highly susceptible to the influence of boundary conditions representing the calculation range, such as latitude, longitude, and altitude, as well as initial conditions such as the initial distribution of the pathogen, making it difficult to ensure reliable accuracy over a prediction period of one week or so. Furthermore, while pathogen biological models are generated based on the specific characteristics of the pathogen, it is difficult to verify the accuracy of the pathogen biological model. Therefore, it has not been possible to predict the arrival of pathogens with high accuracy using simulations.

[0005] Meanwhile, methods for predicting the diffusion of substances released into the atmosphere include, for example, a method for predicting the diffusion of radioactive materials (see, for example, Patent Document 1) and a method for predicting dust concentration distribution (see, for example, Patent Document 2). In the method described in Patent Document 1, the diffusion of radioactive materials released into the atmosphere from a known diffusion source is predicted based on dose information obtained from a dosimeter. In the method described in Patent Document 2, a local wind field is reanalyzed based on numerical weather prediction parameters, and the dust concentration distribution is predicted by atmospheric advection-diffusion simulation from the results and the results of a dust generation amount prediction using a dust generation model.

[0006] An airborne substance source detection device has also been devised that detects the source of particulate airborne substances by time inverse analysis (see, for example, Patent Document 3).A sequential control program has also been devised that predicts observed values ​​of a mathematical model related to microbial fermentation and updates multiple parameters included in the mathematical model using the distribution of predicted values ​​and actual observed values ​​(see, for example, Patent Document 4). [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-7605 [Patent Document 2] Japanese Patent Application Laid-Open No. 2009-109317 [Patent Document 3] Japanese Patent Application Laid-Open No. 2013-185939 [Patent Document 4] Japanese Patent Application Publication No. 2019-198251 Summary of the Invention [Problem to be solved by the invention]

[0008] However, no method has been devised to predict the distribution of diffusing substances such as pathogens at low cost and with high accuracy.

[0009] The present technology has been made in view of such circumstances, and makes it possible to predict the distribution of diffused substances at low cost and with high accuracy. [Means for solving the problem]

[0010] An information processing system according to a first aspect of the present technology is an information processing system including a predictive distribution generation unit that uses sensor data relating to a diffusing substance to data assimilate a diffusion equation including characteristic parameters that are parameters that represent the characteristics of the increase or attenuation of the diffusing substance, thereby generating a predictive distribution, which is the future distribution of the diffusing substance.

[0011] An information processing method according to a first aspect of the present technology is an information processing method including an information processing system using sensor data relating to a diffusing substance to generate a predictive distribution, which is the future distribution of the diffusing substance, by data assimilation of a diffusion equation including characteristic parameters, which are parameters that represent the characteristics of the increase or attenuation of the diffusing substance.

[0012] The program of the first aspect of the present technology is a program for causing a computer to function as a predictive distribution generation unit that generates a predictive distribution, which is the future distribution of a diffusing substance, by using sensor data related to the diffusing substance to data assimilate a diffusion equation including characteristic parameters, which are parameters that represent the characteristics of the increase or attenuation of the diffusing substance.

[0013] In a first aspect of the present technology, sensor data relating to a diffusing substance is used to assimilate a diffusion equation including characteristic parameters that represent the characteristics of the increase or attenuation of the diffusing substance, thereby generating a predictive distribution that is the future distribution of the diffusing substance.

[0014] An information processing system according to a second aspect of the present technology is an information processing system including: a sensor device that acquires sensor data related to a diffusing substance; and an information processing device that includes a predictive distribution generation unit that generates a predictive distribution, which is the future distribution of the diffusing substance, by using the sensor data acquired by the sensor device to data assimilate a diffusion equation including characteristic parameters that are parameters that represent the characteristics of the increase or attenuation of the diffusing substance.

[0015] In a second aspect of the present technology, a sensor device that acquires sensor data related to a diffusive substance and an information processing device that includes a predictive distribution generation unit that generates a predictive distribution, which is the future distribution of the diffusive substance, by using the sensor data acquired by the sensor device to data assimilate a diffusion equation including characteristic parameters that are parameters that represent the characteristics of increase or attenuation of the diffusive substance. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a block diagram showing a configuration example of an embodiment of an information processing system to which the present technology is applied. [Figure 2] FIG. 10 is a diagram showing an example of a diffusion source distribution result screen. [Figure 3] FIG. 10 is a diagram illustrating an example of a predicted distribution result screen. [Figure 4] 10 is a flowchart illustrating a diffusion source distribution generation process. [Figure 5] 10 is a flowchart illustrating a predictive distribution generation process. [Figure 6] FIG. 2 is a block diagram illustrating an example of the hardware configuration of a computer. DETAILED DESCRIPTION OF THE INVENTION

[0017] Hereinafter, modes for carrying out the present technology (hereinafter referred to as embodiments) will be described in the following order. 1. One embodiment (information processing system) 2. Computer

[0018] <1. One embodiment> <Example of information processing system configuration> FIG. 1 is a block diagram showing an example of the configuration of an embodiment of an information processing system to which the present technology is applied.

[0019] The information processing system 10 in Fig. 1 is composed of a sensor device 11, an estimation device 12, and an input / output device 13. The information processing system 10 estimates future or past concentration distribution of a diffusing substance in three-dimensional space by data assimilation of a diffusion equation or an inverse calculation (backward trajectory) of the diffusion equation using sensor data related to the diffusing substance acquired by the sensor device 11.

[0020] Specifically, the sensor devices 11 are installed at multiple locations. The sensor devices 11 acquire quantitative data representing the amount of the diffused substance reaching the target location, such as data representing the concentration of the diffused substance, as sensor data related to the diffused substance. The sensor devices 11 may also acquire data representing physical quantities such as temperature and wind speed as sensor data. The sensor devices 11 transmit the sensor data to the estimation device 12.

[0021] The estimation device 12 (information processing device) includes a communication unit 31, a database 32, a diffusion source distribution generation unit 33, a predicted distribution generation unit , and an information generation unit .

[0022] The communication unit 31 receives sensor data transmitted from the sensor device 11 and supplies it to the database 32 for storage. The communication unit 31 receives input information, which is information input by a user, transmitted from the input / output device 13. The input information includes an estimation mode, which is the operation mode of the estimation device 12, information related to the estimation mode, various requests, etc. The estimation modes include a diffusion source distribution estimation mode that estimates a diffusion source distribution, which is the past concentration distribution of a diffusing substance, and a predicted distribution estimation mode that estimates a predicted distribution, which is the future concentration distribution of a diffusing substance.

[0023] Information related to the diffusion source distribution estimation mode includes target point information, diffusion source distribution analysis region information, diffusion source distribution analysis period, diffusion material information, and additional sensor information. The target point information is information indicating a target point, which is a point of interest for the user, such as a point where the arrival path of the diffusion material is to be estimated. The diffusion source distribution analysis region information is information indicating the region of the diffusion source distribution to be estimated. The diffusion source distribution analysis period is a period from the present to a past time corresponding to the diffusion source distribution to be estimated. The diffusion material information is information indicating the type of diffusion material. Examples of types of diffusion material include various pathogens such as organic yellow rust pathogens, and particulate materials such as soot, pollen, and radioactive materials. The additional sensor information is information indicating the position of a sensor device 11 that is newly added as necessary as the calculation progresses, and whose acquired sensor data is used for data assimilation in the inverse calculation of the diffusion equation.

[0024] Information related to the predictive distribution estimation mode includes target point information, diffusion source distribution information, predictive distribution analysis region information, predictive distribution analysis period, diffusion material information, output format information, etc. Diffusion source distribution information is information that instructs the estimation of diffusion source distribution or information that indicates a known diffusion source distribution. Predictive distribution analysis region information is information that indicates the region of the predictive distribution to be estimated. The predictive distribution analysis period is the period from the present to a future time that corresponds to the predictive distribution to be estimated. Output format information is information that indicates the output format of the predictive distribution, information related to that output format, etc.

[0025] The output format of the predictive distribution may be, for example, a format that outputs the predictive distribution itself as predictive distribution information regarding the predictive distribution, a format that outputs warning information, or a format that outputs at least one of the arrival time and amount of a diffused substance at a target point based on the predictive distribution. The warning information is information that warns that the amount of a diffused substance arriving at a target point based on the predictive distribution is equal to or greater than a threshold. Note that if the diffused substance is a pathogen, the output format of the predictive distribution may be a format that outputs, as predictive distribution information regarding the predictive distribution, the amount of pesticide that corresponds to the amount of the pathogen that will arrive at the target point based on the predictive distribution, necessary to avoid damage from the pathogen. Information regarding the output format for outputting warning information may include a threshold value used to determine whether or not to issue a warning.

[0026] When the estimation mode in the input information is the diffusion source distribution estimation mode, the communication unit 31 supplies information about the diffusion source distribution estimation mode in the input information to the diffusion source distribution generation unit 33. On the other hand, when the estimation mode is the predictive distribution estimation mode, the communication unit 31 supplies information about the predictive distribution estimation mode in the input information to the predictive distribution generation unit 34. The communication unit 31 supplies the estimation mode in the input information to the information generation unit 35. The communication unit 31 transmits information for displaying various screens supplied from the information generation unit 35 to the input / output device 13.

[0027] The communication unit 31 also acquires weather data such as ds093.0 provided by the National Centers for Environmental Prediction (NCEP) or the National Center for Atmospheric Research (NCAR) via the network. The weather data is composed of data indicating physical quantities such as wind speed, temperature, and humidity at each location, and data (maps) indicating topography. The communication unit 31 supplies the weather data to the database 32 for storage.

[0028] The database 32 stores the sensor data and meteorological data supplied from the communication unit 31. The database 32 stores, for each type of diffusing substance, a diffusion equation (mathematical model) linked to the meteorological data, including characteristic parameters that are parameters that represent the characteristics of the increase (growth) or decay (aging or death) of the diffusing substance.

[0029] For example, pathogens bind to aerosols when suspended in the atmosphere, becoming substances with unknown growth or decay characteristics that then diffuse and deposit. Therefore, the diffusion equation for pathogens is a diffusion equation that incorporates meteorological data using e-folding models, SIR models, etc., that include physical property data related to the physical properties of aerosols as parameters, as well as unknown characteristic parameters.

[0030] Examples of physical property data include data described in "The Asian emissions inventory from the INTEX-B (Intercontinental Chemical Transport Experiment-Phase B) project," NASA (The National Aeronautics and Space Administration), 2006. The e-folding model models the decay of pathogens bound to aerosols over a predetermined period equivalent to the half-life of radioactive materials, and includes unknown characteristic parameters. The SIR model models the increase or decrease of pathogens bound to aerosols after their arrival, and includes unknown characteristic parameters. Note that if the period for estimating the concentration distribution of the diffusing substance is short or if the influence of the characteristics of increase or decay is minor, the diffusion equation for pathogens, etc., does not need to include characteristic parameters.

[0031] The diffusion equation for pollen, which diffuses independently without combining with aerosols and has known characteristics of growth, aging, death, etc., is a diffusion equation linked to meteorological data, including characteristic parameters that represent those characteristics. The diffusion equation for radioactive materials with known half-lives is a diffusion equation linked to meteorological data, including characteristic parameters that represent the half-life as a decay characteristic.

[0032] The diffusion source distribution generation unit 33 reads out, from the database 32, a diffusion equation corresponding to the type represented by the diffusion material information, as the diffusion equation to be used this time, based on the diffusion material information among the input information supplied from the communication unit 31. The diffusion source distribution generation unit 33 reads out, from the database 32, sensor data acquired by the sensor device 11 at a predetermined position during a predetermined period, to be used for data assimilation in the inverse calculation of the diffusion equation to be used, based on the diffusion source distribution analysis region information, the diffusion source distribution analysis period, and the additional sensor information.

[0033] The diffusion source distribution generation unit 33 also reads meteorological data for a specified region for a specified period from the database 32 based on the diffusion source distribution analysis region information, the diffusion source distribution analysis period, and the additional sensor information. The diffusion source distribution generation unit 33 generates a diffusion source distribution by assimilating the inverse calculation of the used diffusion equation using the read sensor data and meteorological data. Specifically, the diffusion source distribution generation unit 33 calculates the concentration distribution one time before by inversely calculating the used diffusion equation using meteorological data, and then performs data assimilation using sensor data from each point in the concentration distribution one time before, repeating this process sequentially within the diffusion source distribution analysis period. This allows for highly accurate estimation of the trajectory of the diffusing substance reaching the target point, ultimately generating a highly accurate diffusion source distribution. If the diffusing substance is a pathogen or the like, the diffusion source distribution generation unit 33 estimates not only the diffusion source distribution but also its characteristic parameters through data assimilation. The diffusion source distribution generation unit 33 supplies the diffusion source distribution to the predicted distribution generation unit 34 and the information generation unit 35.

[0034] Based on the diffusion material information included in the input information supplied from the communication unit 31, the predicted distribution generation unit 34 reads out a diffusion equation corresponding to the diffusion material information from the database 32 as a diffusion equation to be used, similar to the diffusion source distribution generation unit 33. When the diffusion source distribution information included in the input information indicates a known diffusion source distribution, the predicted distribution generation unit 34 sets the diffusion source distribution as an initial condition for the diffusion equation to be used. On the other hand, when the diffusion source distribution information indicates an instruction to estimate a diffusion source distribution, the predicted distribution generation unit 34 supplies the information to the information generation unit 35. The predicted distribution generation unit 34 sets the resulting diffusion source distribution supplied from the diffusion source distribution generation unit 33 as the initial condition for the diffusion equation to be used.

[0035] Based on the predictive distribution analysis region information and the predictive distribution analysis period from the input information, the predictive distribution generation unit 34 reads out sensor data acquired during a predetermined period by the sensor device 11 installed at a predetermined location from the database 32. The sensor data is used for data assimilation of the diffusion equation used. Based on the predictive distribution analysis region information and the predictive distribution analysis period, the predictive distribution generation unit 34 also reads out meteorological data for a predetermined period in a predetermined region from the database 32. The meteorological data is used for data assimilation of the diffusion equation used.

[0036] The predictive distribution generation unit 34 uses the read sensor data and meteorological data to generate a predictive distribution for the predictive distribution analysis period by data assimilation of the used diffusion equation with the diffusion source distribution as the initial condition. At this time, if the type of diffusing substance is a pathogen or the like, the predictive distribution generation unit 34 estimates not only the predictive distribution but also the characteristic parameters through data assimilation. The predictive distribution generation unit 34 supplies the output format information and the predictive distribution from the input information to the information generation unit 35.

[0037] The information generation unit 35 generates information for displaying various screens and supplies it to the communication unit 31. For example, when the estimation mode supplied from the communication unit 31 is the diffusion source distribution estimation mode, or when diffusion source distribution information instructing estimation of the diffusion source distribution is supplied from the predicted distribution generation unit 34, the information generation unit 35 generates diffusion source distribution setting screen information. The diffusion source distribution setting screen information is information for displaying a diffusion source distribution setting screen as a UI (User Interface) for setting information related to the diffusion source distribution estimation mode. The information generation unit 35 supplies the diffusion source distribution setting screen information to the communication unit 31.

[0038] On the other hand, when the estimation mode is the predictive distribution estimation mode, the information generation unit 35 generates predictive distribution setting screen information for displaying a predictive distribution setting screen as a UI for setting information related to the predictive distribution estimation mode, and supplies the generated information to the communication unit 31.

[0039] The information generation unit 35 generates diffusion source distribution result screen information for displaying a diffusion source distribution result screen including the diffusion source distribution supplied from the diffusion source distribution generation unit 33 as an analysis result, and supplies the information to the communication unit 31. The information generation unit 35 generates predictive distribution information based on the output format information and predictive distribution supplied from the predictive distribution generation unit 34. The information generation unit 35 generates predictive distribution result screen information for displaying a predictive distribution result screen including the predictive distribution information as an analysis result, and supplies the information to the communication unit 31.

[0040] The input / output device 13 is configured by, for example, a smartphone, a personal computer, or the like. The input / output device 13 includes an input unit 41, a display unit 42, a control unit 43, and a communication unit 44. The input unit 41 receives input from a user and generates input information. The input unit 41 supplies the input information to the control unit 43.

[0041] The display unit 42 displays a diffusion source distribution setting screen, a predicted distribution setting screen, a diffusion source distribution result screen, or a predicted distribution result screen under the control of the control unit 43.

[0042] The control unit 43 controls each unit of the input / output device 13. For example, the control unit 43 controls the communication unit 44 to transmit input information supplied from the input unit 41 to the estimation device 12. The control unit 43 controls the communication unit 44 to receive diffusion source distribution setting screen information, predicted distribution setting screen information, diffusion source distribution result screen information, and predicted distribution result screen information transmitted from the communication unit 31. The control unit 43 causes the display unit 42 to display a diffusion source distribution setting screen based on the diffusion source distribution setting screen information. The control unit 43 causes the display unit 42 to display a predicted distribution setting screen based on the predicted distribution setting screen information. The control unit 43 causes the display unit 42 to display a diffusion source distribution result screen based on the diffusion source distribution result screen information. The control unit 43 causes the display unit 42 to display a predicted distribution result screen based on the predicted distribution result screen information.

[0043] The communication unit 44 communicates with the estimation device 12 under the control of the control unit 43 .

[0044] <Example of diffusion source distribution results screen> FIG. 2 is a diagram showing an example of a diffusion source distribution result screen displayed on the display unit 42.

[0045] First, when the user inputs the position of the target point, an estimation mode selection screen consisting of the estimation mode selection unit 61 shown in Fig. 2 is displayed on the display unit 42. The estimation mode selection unit 61 is operated by the user when inputting the diffusion source distribution estimation mode or the predicted distribution estimation mode as the estimation mode. In the example of Fig. 2, the user inputs the diffusion source distribution estimation mode as the estimation mode by checking the check box to the left of "Diffusion Source Distribution Estimation."

[0046] In this case, a diffusion source distribution setting screen consisting of an estimation mode selection section 61, a diffusion source distribution analysis region input section 62, a diffusion source distribution analysis period input section 63, a diffusion material information input section 64, and an additional sensor information input section 65 is displayed.

[0047] The diffusion source distribution analysis region input unit 62 is operated by the user when inputting diffusion source distribution analysis region information. The user inputs, as the diffusion source distribution analysis region information, for example, the longitude, latitude, and altitude of a predetermined position such as the center of the diffusion source distribution analysis region, and the size of the diffusion source distribution analysis region. The diffusion source distribution analysis region is, for example, the region surrounding the target point.

[0048] The diffusion source distribution analysis period input unit 63 is operated by the user when inputting the diffusion source distribution analysis period. For example, the user inputs the period from the current date of July 20, 2010, to July 16, 2010, four days prior, as the diffusion source distribution analysis period. The diffusion material information input unit 64 is operated by the user when inputting diffusion material information. The user selects a type corresponding to the desired diffusion source distribution from the types of diffusion materials corresponding to the diffusion equations stored in the database 32, which are displayed by a pull-down menu in the diffusion material information input unit 64, thereby inputting diffusion material information indicating the type.

[0049] The additional sensor information input unit 65 is operated by the user when inputting additional sensor information. As the additional sensor information, the user inputs, for example, the latitude and longitude of a sensor device 11 that is newly added as needed as the calculation progresses and whose acquired sensor data is used for data assimilation in the inverse calculation of the diffusion equation.

[0050] When a diffusion source distribution is generated based on the input information on the diffusion source distribution setting screen, a diffusion source distribution result screen 60 shown in Fig. 2 is displayed on the display unit 42. The diffusion source distribution result screen 60 is obtained by adding a diffusion source distribution display unit 66 to the diffusion source distribution setting screen, which displays the diffusion source distribution generated based on the input information on the diffusion source distribution setting screen as an analysis result.

[0051] In the diffusion source distribution display section 66 in Fig. 2, the diffusion source distribution is displayed in association with a map of the diffusion source distribution analysis area. Information indicating the position of the target point (triangle marks in the example of Fig. 2) and information indicating the position of the sensor device 11 (circle marks in the example of Fig. 2) are also superimposed on this map.

[0052] <Example of predictive distribution results screen> FIG. 3 is a diagram showing an example of a predicted distribution result screen displayed on the display unit 42.

[0053] In FIG. 3, parts corresponding to those in FIG. 2 are given the same reference numerals.

[0054] First, when the user inputs the position of the target point, an estimation mode selection screen consisting of an estimation mode selection unit 61 is displayed on the display unit 42. In the example of Fig. 3, the user inputs the predictive distribution estimation mode as the estimation mode by checking the check box to the left of "predictive distribution estimation."

[0055] In this case, a predicted distribution setting screen consisting of an estimation mode selection unit 61, a diffusion source distribution information input unit 81, a predicted distribution analysis area input unit 82, a predicted distribution analysis period input unit 83, a diffusion material information input unit 84, and an output format information input unit 85 is displayed on the display unit 42.

[0056] The diffusion source distribution information input unit 81 is operated by the user when inputting whether or not the diffusion source distribution is known. If the diffusion source distribution is known, the user inputs that the diffusion source distribution is known in the diffusion source distribution information input unit 81, and inputs information indicating the diffusion source distribution as diffusion source distribution information. The information indicating the diffusion source distribution is composed of, for example, the latitude and longitude of each position in the diffusion source distribution and the concentration of the diffusion material at that position. Examples of cases in which the diffusion source distribution is known include cases in which the diffusion material is radioactive material released by a nuclear power plant accident, soot emitted from a factory, etc.

[0057] On the other hand, if the diffusion source distribution is unknown, the user inputs information instructing the estimation of the diffusion source distribution as diffusion source distribution information. In this case, a message prompting the user to estimate the diffusion source distribution is displayed, and if the user accepts, the diffusion source distribution setting screen is displayed. As a result, the diffusion source distribution result screen of FIG. 2 is displayed, and if the user inputs an instruction such as "OK", the display screen returns to the predicted distribution setting screen.

[0058] The predictive distribution analysis area input unit 82 is operated by the user when inputting predictive distribution analysis area information. The user inputs, as predictive distribution analysis area information, for example, the longitude, latitude, and altitude of a predetermined position such as the center of the predictive distribution analysis area, and the size of the predictive distribution analysis area. The predictive distribution analysis area is, for example, an area surrounding a target point.

[0059] The predictive distribution analysis period input unit 83 is operated by the user when inputting the predictive distribution analysis period. The diffusing material information input unit 84 is operated by the user when inputting the diffusing material information. The user inputs the diffusing material information indicating the desired type by selecting the type corresponding to the desired predictive distribution from the types of diffusing materials corresponding to the diffusion equations stored in the database 32, which are displayed in a pull-down menu in the diffusing material information input unit 84. The output format information input unit 85 is operated by the user when inputting the output format information.

[0060] When a predictive distribution is generated based on the input information on the predictive distribution setting screen, the display unit 42 displays a predictive distribution result screen 80 shown in Fig. 3. The predictive distribution result screen 80 is obtained by adding a predictive distribution information display unit 86 to the predictive distribution setting screen, which displays the predictive distribution information generated based on the input information on the predictive distribution setting screen as an analysis result.

[0061] In the example of Fig. 3, output format information indicating the format in which the predictive distribution itself is to be output is input in an output format information input unit 85. Accordingly, a predictive distribution 91, which is a moving image of the predictive distribution analysis period, a time axis bar 92, a play button 93, a rewind button 94, and a fast-forward button 95 are displayed in the predictive distribution information display unit 86. In the example of Fig. 3, the predictive distribution 91 is displayed in association with a map of the predictive distribution analysis area. Information indicating the position of the target point (a triangular mark in the example of Fig. 3) is also superimposed on this map.

[0062] The time axis bar 92 represents the predictive distribution analysis period. When specifying a playback position in the predictive distribution 91, the user operates the position on the time axis bar 92 that corresponds to the playback position. The play button 93 is operated by the user when playing back the predictive distribution 91. The rewind button 94 is operated by the user when rewinding the predictive distribution 91. The fast-forward button 95 is operated by the user when fast-forwarding the predictive distribution 91.

[0063] In addition, when the diffusible substance is a pathogen or the like, the diffusion source distribution result screen 60 in Fig. 2 and the predicted distribution result screen 80 in Fig. 3 may include notification information notifying the characteristics represented by the estimated characteristic parameters. In this case, the information generation unit 35 generates the notification information based on the characteristic parameters estimated by the diffusion source distribution generation unit 33 and the predicted distribution generation unit 34. Then, the information generation unit 35 generates diffusion source distribution result screen information on the diffusion source distribution result screen and predicted distribution result screen information on the predicted distribution result screen, which include the notification information. This notification information allows the user to know the lifespan (lifetime) of the pathogen.

[0064] <Explanation of diffusion source distribution generation process> Fig. 4 is a flowchart illustrating a diffusion source distribution generation process by the diffusion source distribution generation unit 33 in Fig. 1. This diffusion source distribution generation process is started, for example, when a user sets information related to a diffusion source distribution estimation mode on a diffusion source distribution setting screen.

[0065] In step S11 of FIG. 4, the diffusion source distribution generating unit 33 reads out the diffusion equation to be used from the database 32 based on the diffusion material information transmitted from the input / output device 13 and supplied via the communication unit 31.

[0066] In step S12, the diffusion source distribution generating unit 33 sets the count value N to 1. In step S13, the diffusion source distribution generating unit 33 determines the time of the diffusion source distribution analysis period transmitted from the input / output device 13 and supplied via the communication unit 31 as time t p Set to.

[0067] In step S14, the diffusion source distribution generation unit 33 sets the initial condition I, the terrain information r, and the boundary condition z of the diffusion equation to be used that were read in step S11. Specifically, the diffusion source distribution generation unit 33 sets the region of diffusion source distribution represented by the diffusion source distribution analysis region information transmitted from the input / output device 13 and supplied via the communication unit 31 as the initial condition I. The diffusion source distribution generation unit 33 reads terrain data of the vicinity of the region of diffusion source distribution from the database 32, and sets the terrain information r based on the terrain data. This terrain information r is a value corresponding to terrain such as vegetation, snow surface, sea surface, etc., and is a model coefficient used in the diffusion equation. The diffusion source distribution generation unit 33 sets the model coefficient, etc. as the boundary condition z.

[0068] In step S15, the diffusion source distribution generation unit 33 estimates the concentration distribution of the diffusing substance NΔt time before the current time by performing an inverse calculation of the diffusion equation to which the initial condition I, boundary condition z, and topographical information r were set in step S12, where Δt is the estimation interval of the concentration distribution of the diffusing substance.

[0069] In step S16, the diffusion source distribution generation unit 33 performs data assimilation of the inverse calculation of the diffusion equation to be used, using the sensor data and meteorological data corresponding to the diffusion source distribution analysis region information, the diffusion source distribution analysis period, and the additional sensor information stored in the database 32. As a result, for example, the initial condition I of the diffusion equation is updated to an intermediate value between the concentration distribution of the diffusing substance estimated by the processing of step S15 and the actual measured value of the concentration distribution of the diffusing substance corresponding to the sensor data. At this time, the unknown characteristic parameter f included in the diffusion equation is also updated.

[0070] In step S17, the diffusion source distribution generating unit 33 calculates the diffusion source distribution NΔt as the time t p In step S17, it is determined whether NΔt is equal to or greater than time t p If it is determined that this is not the case, that is, if the concentration distribution of the diffusion material for the entire diffusion source distribution analysis period has not yet been estimated, the process proceeds to step S18.

[0071] In step S18, the diffusion source distribution generating unit 33 increments the count value N by 1. Then, the process returns to step S15, and the subsequent processes are repeated.

[0072] On the other hand, in step S17, NΔt is set to time t p If it is determined that this is the case, that is, if it is determined that the concentration distribution of the diffusing substance has been estimated for the entire diffusion source distribution analysis period, the process proceeds to step S19. In step S19, the diffusion source distribution generating unit 33 outputs the concentration distribution of the diffusing substance last estimated in the process of step S15 as the diffusion source distribution to the predicted distribution generating unit 34 and the information generating unit 35, and the diffusion source distribution generating process ends.

[0073] <Explanation of the predictive distribution generation process> Fig. 5 is a flowchart illustrating the predictive distribution generation process performed by the predictive distribution generator 34 in Fig. 1. This predictive distribution generation process is started, for example, when the user sets information related to the predictive distribution estimation mode on the predictive distribution setting screen.

[0074] In step S31 of FIG. 5, the predicted distribution generating unit 34 reads out the diffusion equation to be used from the database 32 based on the diffusing material information transmitted from the input / output device 13 and supplied via the communication unit 31.

[0075] In step S32, the predictive distribution generation unit 34 sets the count value M to 1. In step S33, the predictive distribution generation unit 34 sets the time of the predictive distribution analysis period transmitted from the input / output device 13 and supplied via the communication unit 31 as time t f Set to.

[0076] In step S34, the predictive distribution generation unit 34 sets the initial condition I, the topographical information r, and the boundary condition z of the diffusion equation to be used that were read in step S31. Specifically, the predictive distribution generation unit 34 sets the diffusion source distribution indicated by the diffusion source distribution information transmitted from the input / output device 13 and supplied via the communication unit 31, or the diffusion source distribution output by the processing of step S19 in FIG. 4, as the initial condition I. The predictive distribution generation unit 34 reads out topographical data in the vicinity of the region of the predictive distribution from the database 32, and sets the topographical information r based on the topographical data, similar to the diffusion source distribution generation unit 33. The predictive distribution generation unit 34 sets the model coefficients, etc., as the boundary condition z.

[0077] In step S35, the predicted distribution generating unit 34 estimates the concentration distribution of the diffusing substance MΔt hours from the current time by calculating the used diffusion equation in which the initial condition I, topographical information r, and boundary condition z are set in step S32.

[0078] In step S36, the predictive distribution generator 34 performs data assimilation of the diffusion equation to be used using the sensor data and meteorological data corresponding to the predictive distribution analysis region information and the predictive distribution analysis period stored in the database 32. As a result, for example, the initial condition I of the diffusion equation is updated to an intermediate value between the concentration distribution of the diffusing substance estimated by the process of step S35 and the actual measured value of the concentration distribution of the diffusing substance corresponding to the sensor data. At this time, the unknown characteristic parameter f included in the diffusion equation is also updated.

[0079] In step S37, the predictive distribution generator 34 calculates the time interval MΔt between the time t f In step S37, it is determined whether MΔt is equal to or greater than the time t f If it is determined that this is not the case, that is, if the concentration distribution of the diffusive substance has not yet been estimated for the entire predicted distribution analysis period, the process proceeds to step S38.

[0080] In step S38, the predictive distribution generating unit 34 increments the count value M by 1. Then, the process returns to step S35, and the subsequent processes are repeated.

[0081] On the other hand, in step S37, MΔt is set to time t f If it is determined that this is the case, i.e., if it is determined that the concentration distribution of the diffusible substance has been estimated for the entire predictive distribution analysis period, the process proceeds to step S39. In step S39, the predictive distribution generation unit 34 outputs the concentration distribution of the diffusible substance for the predictive distribution analysis period, which was sequentially estimated in the process of step S35, as a predicted distribution to the information generation unit 35, and also outputs output format information to the information generation unit 35. This output format information is transmitted from the input / output device 13 and supplied via the communication unit 31. After the process of step S39, the predictive distribution generation process ends.

[0082] As described above, the predictive distribution generator 34 generates a predictive distribution by data assimilation of a diffusion equation including characteristic parameters using sensor data related to diffusing substances. Therefore, a highly accurate predictive distribution can be easily generated at low cost, with accuracy not significantly dependent on boundary conditions or the resolution (level of detail) of meteorological data. Specifically, the predictive distribution generator 34 performs data assimilation of the diffusion equation, taking into account the uncertainty of the predicted values ​​calculated using the diffusion equation and the observed values, such as sensor data. This data assimilation allows the reliability of the calculated values ​​to be evaluated using sensor data, so the predictive distribution generator 34 can predict the concentration distribution of diffusing substances at low cost and with high accuracy.

[0083] As a result, for example, the information generation unit 35 can generate highly accurate warning information as predicted distribution information based on this predicted distribution. This allows, for example, farmers to predict the arrival of pathogens above a threshold level before the pathogens arrive and take measures to minimize damage, such as spraying pesticides. The information generation unit 35 can also generate at least one of the arrival time and amount of a diffused substance at a target location as predicted distribution information based on this predicted distribution. This allows, for example, if the diffused substance is a pathogen, farmers to spray a pesticide in an amount corresponding to the amount of pathogens that has arrived at the appropriate location at the appropriate time. This allows farmers to produce safe crops using the minimum amount of pesticide necessary at low cost.

[0084] The diffusion source distribution generator 33 generates a diffusion source distribution by data assimilation of the inverse calculation of the diffusion equation using sensor data related to the diffusing substance. Therefore, the predicted distribution generator 34 can set this diffusion source distribution as the initial condition of the diffusion equation, for example. Therefore, a more accurate concentration distribution in three-dimensional space can be set as the initial condition compared to when the initial condition is set to a concentration distribution, which is a point distribution of the concentration of the diffusing substance represented by sensor data at each position.

[0085] The diffusion equation does not need to be linked to meteorological data, and may also use information on the structure of the region and atmospheric layers.

[0086] At least some of the sensor devices 11 may be installed in mobile objects such as automobiles and drones. In this case, the estimation device 12 feeds back the locations of the sensor devices 11 installed in the mobile objects. Specifically, for example, the information generation unit 35 generates, based on the predicted distribution (diffusion source distribution), location information indicating one or more positions in three-dimensional space where many of the diffusing substances are suspended, i.e., where the sensitivity of the diffusing substances is high, as location information indicating the locations of the sensor devices 11. Then, the information generation unit 35 transmits this location information to the input / output device 13 via the communication unit 31, thereby displaying it on the display unit 42.

[0087] The user moves the mobile body on which the sensor device 11 is installed to the position indicated by this location information. For example, if the concentration of the diffused substance at high altitudes is high in the predicted distribution (diffusion source distribution), the user moves the mobile body on which the sensor device 11 is installed so that the altitude of the mobile body increases based on the location information indicating the high altitude position. As a result, the accuracy of the predicted distribution (diffusion source distribution) can be improved.

[0088] The estimation device 12 may feed back the functions of the sensor device 11, such as the type of sensor data, instead of the location information, or may feed back both the location information and the functions of the sensor device 11.

[0089] <2. Computer> The series of processes of the above-described estimation device 12 and input / output device 13 can be executed by hardware or software. When the series of processes are executed by software, the programs constituting the software are installed in a computer. Here, the computer includes a computer incorporated in dedicated hardware, and a general-purpose personal computer, for example, that can execute various functions by installing various programs.

[0090] FIG. 6 is a block diagram showing an example of the hardware configuration of a computer that executes a series of processes of the above-described estimation device 12 and input / output device 13 by a program.

[0091] In the computer, a CPU (Central Processing Unit) 901, a ROM (Read Only Memory) 902, and a RAM (Random Access Memory) 903 are interconnected by a bus 904.

[0092] An input / output interface 905 is also connected to the bus 904. An input unit 906, an output unit 907, a storage unit 908, a communication unit 909, and a drive 910 are connected to the input / output interface 905.

[0093] The input unit 906 includes a keyboard, a mouse, a microphone, etc. The input unit 906 corresponds to, for example, the input unit 41. The output unit 907 includes a display, a speaker, etc. The output unit 907 corresponds to, for example, the display unit 42. The storage unit 908 includes a hard disk, a non-volatile memory, etc., and corresponds to, for example, the database 32. The communication unit 909 includes a network interface, etc. The drive 910 drives removable media 911 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

[0094] In the computer configured as above, the CPU 901 performs the above-described series of processes by, for example, loading a program stored in the storage unit 908 into the RAM 903 via the input / output interface 905 and the bus 904 and executing the program. For example, the CPU 901 functions as the control unit 43 or the communication unit 44, or as the communication unit 31, the diffusion source distribution generation unit 33, the predicted distribution generation unit 34, or the information generation unit 35.

[0095] The program executed by the computer (CPU 901) can be provided by being recorded on removable media 911 such as package media, for example. The program can also be provided via wired or wireless transmission media such as a local area network, the Internet, or digital satellite broadcasting.

[0096] In a computer, the program can be installed in the storage unit 908 via the input / output interface 905 by inserting the removable medium 911 into the drive 910. The program can also be received by the communication unit 909 via a wired or wireless transmission medium and installed in the storage unit 908. Alternatively, the program can be installed in the ROM 902 or the storage unit 908 in advance.

[0097] The program executed by the computer may be a program that processes in chronological order according to the order described in this specification, or may be a program that processes in parallel or at the required timing, such as when called.

[0098] In this specification, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all the components are contained in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device with multiple modules housed in a single housing, are both systems.

[0099] The embodiments of the present technology are not limited to the above-described embodiments, and various modifications are possible without departing from the spirit of the present technology.

[0100] For example, this technology can be configured as cloud computing, in which a single function is shared and processed collaboratively by multiple devices via a network.

[0101] Furthermore, each step described in the above flowchart can be executed by one device, or can be shared and executed by multiple devices.

[0102] Furthermore, when one step includes multiple processes, the multiple processes included in that one step can be executed by one device or can be shared and executed by multiple devices.

[0103] The effects described in this specification are merely examples and are not limiting, and other effects may also be present.

[0104] The present technology can take the following configurations. (1) a predictive distribution generating unit that generates a predictive distribution, which is a future distribution of the diffused material, by data assimilating a diffusion equation including characteristic parameters, which are parameters that represent the characteristics of the increase or attenuation of the diffused material, using sensor data related to the diffused material; An information processing system comprising: (2) The diffusion equation is coupled with meteorological data It was configured as The information processing system according to (1) above. (3) A region or period input section for inputting the region or period of the predicted distribution Further equipped The information processing system according to (1) or (2). (4) A diffusion material information input section for inputting the type of diffusion material Further equipped The information processing system according to any one of (1) to (3) above. (5) The diffusion equation differs for each type of diffusing material, The predictive distribution generating unit generates the predictive distribution by using the sensor data to data assimilate the diffusion equation corresponding to the type of the diffusive material input by the diffusive material information input unit. It was configured as The information processing system according to (4) above. (6) a display unit that displays predictive distribution information regarding the predictive distribution generated by the predictive distribution generating unit; Further equipped The information processing system according to any one of (1) to (5) above. (7) The predictive distribution information is the predictive distribution It was configured as The information processing system according to (6) above. (8) The predicted distribution information is warning information that warns that the amount of the diffused material reaching a predetermined point based on the predicted distribution is equal to or greater than a threshold. It was configured as The information processing system according to (6) above. (9) The predicted distribution information is at least one of the arrival time and the arrival amount of the diffusing substance at a predetermined point based on the predicted distribution. It was configured as The information processing system according to (6) above. (10) The predictive distribution generating unit also estimates the characteristic parameters by data assimilation of the diffusion equation. It was configured as The information processing system according to any one of (1) to (9) above. (11) a notification information generation unit that generates notification information for notifying the characteristic based on the characteristic parameters estimated by the predictive distribution generation unit; Further equipped The information processing system according to (10) above. (12) a diffusion source distribution generation unit that generates a diffusion source distribution, which is a distribution of the diffusion material in the past, by data assimilation of an inverse calculation of the diffusion equation using the sensor data; Furthermore, The predictive distribution generating unit generates the predictive distribution by data assimilation of the diffusion equation using the diffusion source distribution generated by the diffusion source distribution generating unit as an initial condition. It was configured as The information processing system according to any one of (1) or (3) to (11) above. (13) A diffusion source distribution input section for inputting a diffusion source distribution, which is a distribution of the diffusion material in the past. Furthermore, The predictive distribution generation unit generates the predictive distribution by data assimilation of the diffusion equation using the diffusion source distribution input by the diffusion source distribution input unit as an initial condition. It was configured as The information processing system according to any one of (1) to (11) above. (14) the diffusing substance is a pathogen; The diffusion equation also includes as parameters the physical property data of the aerosol to which the pathogen binds when suspended. It was configured as The information processing system according to any one of (1) to (13) above. (15) a placement information generation unit that generates placement information indicating placement of the sensor devices that acquire the sensor data based on the predictive distribution; Further equipped The information processing system according to any one of (1) to (14). (16) The information processing system Using sensor data related to the diffusing material, a diffusion equation including characteristic parameters that are parameters that represent the characteristics of the increase or attenuation of the diffusing material is data assimilated to generate a predictive distribution that is a distribution of the diffusing material in the future. An information processing method including: (17) Computer, a predictive distribution generating unit that generates a predictive distribution, which is a future distribution of the diffused material, by data assimilating a diffusion equation including characteristic parameters, which are parameters that represent the characteristics of the increase or attenuation of the diffused material, using sensor data related to the diffused material; A program to function as a (18) a sensor device for acquiring sensor data relating to a diffusing material; an information processing device including a predictive distribution generation unit that generates a predictive distribution, which is a future distribution of the diffused substance, by data assimilating a diffusion equation including characteristic parameters that are parameters that represent the characteristics of increase or attenuation of the diffused substance using the sensor data acquired by the sensor device; An information processing system comprising: [Explanation of symbols]

[0105] 10 Information processing system, 11 Sensor device, 12 Estimation device, 33 Diffusion source distribution generation unit, 34 Prediction distribution generation unit, 35 Information generation unit, 42 Display unit, 81 Diffusion source distribution information input unit, 82 Prediction distribution analysis area input unit, 83 Prediction distribution analysis period input unit, 84 Diffusion material information input unit

Claims

1. a predictive distribution generating unit that generates a predictive distribution, which is a future distribution of the diffused material, by data assimilating a diffusion equation including characteristic parameters, which are parameters that represent the characteristics of the increase or attenuation of the diffused material, using sensor data related to the diffused material; An information processing system comprising:

2. The diffusion equation is coupled with meteorological data It was configured as The information processing system according to claim 1 .

3. A region or period input section for inputting the region or period of the predicted distribution Further equipped The information processing system according to claim 1 .

4. A diffusion material information input section for inputting the type of diffusion material Further equipped The information processing system according to claim 1 .

5. The diffusion equation differs for each type of diffusing material, The predictive distribution generating unit generates the predictive distribution by using the sensor data to data assimilate the diffusion equation corresponding to the type of the diffusive material input by the diffusive material information input unit. It was configured as The information processing system according to claim 4 .

6. a display unit that displays predictive distribution information regarding the predictive distribution generated by the predictive distribution generating unit; Further equipped The information processing system according to claim 1 .

7. The predictive distribution information is the predictive distribution It was configured as The information processing system according to claim 6.

8. The predicted distribution information is warning information that warns that the amount of the diffused material reaching a predetermined point based on the predicted distribution is equal to or greater than a threshold. It was configured as The information processing system according to claim 6.

9. The predicted distribution information is at least one of the arrival time and the arrival amount of the diffusing substance at a predetermined point based on the predicted distribution. It was configured as The information processing system according to claim 6.

10. The predictive distribution generating unit also estimates the characteristic parameters by data assimilation of the diffusion equation. It was configured as The information processing system according to claim 1 .

11. a notification information generation unit that generates notification information for notifying the characteristic based on the characteristic parameters estimated by the predictive distribution generation unit; Further equipped The information processing system according to claim 10.

12. a diffusion source distribution generation unit that generates a diffusion source distribution, which is a distribution of the diffusion material in the past, by data assimilation of an inverse calculation of the diffusion equation using the sensor data; Furthermore, The predictive distribution generating unit generates the predictive distribution by data assimilation of the diffusion equation using the diffusion source distribution generated by the diffusion source distribution generating unit as an initial condition. It was configured as The information processing system according to claim 1 .

13. A diffusion source distribution input section for inputting a diffusion source distribution, which is a distribution of the diffusion material in the past. Furthermore, The predictive distribution generation unit generates the predictive distribution by data assimilation of the diffusion equation using the diffusion source distribution input by the diffusion source distribution input unit as an initial condition. It was configured as The information processing system according to claim 1 .

14. the diffusing substance is a pathogen; The diffusion equation also includes as parameters the physical property data of the aerosol to which the pathogen binds when suspended. It was configured as The information processing system according to claim 1 .

15. a placement information generation unit that generates placement information indicating placement of the sensor devices that acquire the sensor data based on the predictive distribution; Further equipped The information processing system according to claim 1 .

16. The information processing system Using sensor data related to the diffusing material, a diffusion equation including characteristic parameters that are parameters that represent the characteristics of the increase or attenuation of the diffusing material is data assimilated to generate a predictive distribution that is a distribution of the diffusing material in the future. An information processing method including:

17. Computer, a predictive distribution generating unit that generates a predictive distribution, which is a future distribution of the diffused material, by data assimilating a diffusion equation including characteristic parameters, which are parameters that represent the characteristics of the increase or attenuation of the diffused material, using sensor data related to the diffused material; A program to function as a

18. a sensor device for acquiring sensor data relating to a diffusing material; an information processing device including a predictive distribution generation unit that generates a predictive distribution, which is a future distribution of the diffused substance, by data assimilating a diffusion equation including characteristic parameters that are parameters that represent the characteristics of increase or attenuation of the diffused substance using the sensor data acquired by the sensor device; An information processing system comprising:

Images