JP6846782B2 - Analysis data pipeline system from environmental sensing data acquisition to advanced analysis - Google Patents

Description

The present invention relates to an analysis data pipeline system from outdoor environmental sensing data acquisition to advanced analysis such as in the fields of agricultural prediction and disaster prediction.

In the agricultural field of Japan, the utilization of information and communication technology (IT: Information Technology) is advancing as a powerful means for improving the efficiency of production control, mainly for large-scale management bodies. In particular, in recent years, the amount of agricultural environment data has been growing due to the spread of IoT (Internet of Things), which connects all things to the Internet and promotes social change. For example, the utilization of big data is rapidly increasing in the agricultural field as well. With progress, the agricultural IT industry is expanding rapidly.

On the other hand, as Japan's agricultural tailors are aging, they are facing various issues such as a serious labor shortage, and they are facing various issues such as labor saving and yield by utilizing IT and artificial intelligence (AI).・ It is important to dramatically improve productivity such as quality improvement and to develop human resources.

Therefore, along with these changes, there is a rapidly increasing need for ensuring the interoperability and portability of so-called agricultural information, such as sharing and comparing data between different agricultural IT systems. In addition, if interoperability and portability of agricultural information are ensured, it is expected that big data analysis of information obtained from agricultural IT systems will lead to the creation of new services and new businesses.

Under these circumstances, the Government of Japan has also established a framework for the utilization of agricultural big data as a guideline, and at the same time, points to keep in mind when the excellent know-how of Japanese agriculture, which is attracting attention from overseas, is utilized as big data. Was established ahead of the rest of the world (see, for example, Non-Patent Document 1). This guideline has been standardized for the purpose of ensuring interoperability and portability of agricultural information, and data exchange is performed by presenting a reference model for the interface for exchanging data with domestic agricultural IT systems. This is to facilitate the process of. This will ensure data interoperability and portability through the promotion of data linkage between systems, and will facilitate data linkage with fields other than agriculture. In addition, by using the same guidelines from research and development to production sites and even in stores, we will promote the effects of collaboration between industry, academia and government in Japan's agricultural IT industry, and export IT-enabled agricultural equipment and solutions. Agricultural IT business can be expected to develop.

By the way, regarding the use of agriculture-related information, for example, observation systems and observation stations that do not require complicated and complicated work such as installation of field sensors in the field and data collection / disclosure are disclosed (for example, patent documents). 1).

Japanese Unexamined Patent Publication No. 2010-49560

"<H28-GL2> Individual guidelines for data items of environmental information used in agricultural IT systems (full-scale operation version), Kiyoshi Honda (draft guidelines): Prime Minister's Office, Policy Council, Advanced Information and Communication Network Society Promotion Strategy Headquarters (IT) General Strategy Headquarters), New Strategy Promotion Expert Committee Subcommittee "[Search on December 10, 2017], Internet <URL: http://www.kantei.go.jp/jp/singi/it2/senmon_bunka/nougyou .html ＞

However, even if it is said that "the interoperability and portability of agricultural information is ensured and the information obtained from the agricultural IT system is analyzed by big data", this is not easy to do in reality. The reason will be described below.

The first reason is that analysis of big data obtained from agricultural IT systems, such as predictive analytics such as crop yield prediction and weather prediction, requires a high level of diverse agricultural expertise.

Traditionally, farmers have always needed forecast information related to various agricultural environments. For example, when is the optimal planting date to respond to climate change, when to plant the maximum yield, what is the impact of El Nino, and how changing the amount of fertilizer will affect the yield. When will the flowers come out, when should they be planted quickly using expensive high-performance agricultural machinery, when the flowers will bloom (which can predict when they will bear fruit, pests, pesticide countermeasures, etc.), and when will the warehouse management company be? Is rice collected from farmers?

However, no matter how much big data is related to the agricultural environment, the fact is that accurate information analysis cannot be performed without advanced expertise. In particular, the construction of a data processing pipeline from the acquisition of agricultural environment data to analysis has become extremely important to ensure the efficiency of analysis, and it will lead to an increase in the value of the data. Is not easy to build.

The second reason is that at present, regarding the observation data provided by each sensor, the server, access means, metadata, format, measurement unit, each name, etc. are not unified at all, and interoperability of agricultural information The property and portability have not been ensured.

For example, as shown in FIG. 26, in the conventional agricultural data analysis system 100, an observation station (field sensor 101) having a plurality of sensors is installed in a field and connected to a network. This field sensor 101 is equipped with various sensors that collect meteorological data such as temperature, humidity, rainfall, solar radiation, and air volume, and a camera sensor that observes the growth status of crops, and publishes the data observed by the WWW protocol. are doing. As a result, it is assumed that the weather data of the remote area where the field sensor is placed can be referenced and collected using a Web browser.

However, as a practical matter, data such as the server, access means, metadata, format, unit of measurement, and name of each field sensor 101 are not unified at all, and are different for each sensor and each company. Therefore, in order to acquire the sensor information in the terminal device 102, it is necessary to install software corresponding to each sensor 101, format conversion work for reading the sensor information, and the like, which requires a great deal of time and effort for the engineer. .. Under such circumstances, it is difficult to ensure the interoperability and portability of agricultural environment data, which in turn makes it difficult to acquire and analyze agricultural environment data.

The present invention has been made in view of the above problems, and while ensuring and integrating interoperability of agricultural environment data, which has been disjointed in the past, analysis data from acquisition of environmental sensing data to advanced analysis in the agricultural field. The purpose is to provide a pipelined system.

In order to achieve the above object, the present invention is an analysis data pipeline system that performs from outdoor environment sensing data acquisition to advanced analysis, and the analysis data pipeline system was observed using an observation sensor. An observation data distribution server that distributes environmental sensing data, an analysis server that automatically acquires environmental sensing data from the observation data distribution server when requested, and performs a predetermined analysis, and a predetermined application. The observation data distribution server includes a terminal device that acquires analysis results from the analysis server and displays the analysis results when a user requests advanced analysis using the application. The analysis server and the terminal device are connected via a web network, and the observation data distribution server, the analysis server, and the terminal device are chained based on an API (Application Programming Interface) to obtain the environment sensing data. There line pipelined processing up advanced analysis, the the observed data distribution server and the observation sensor, server, access means, metadata, format, units of measurement, at least one of the data of the name common The feature is that the API of is defined.

In this analysis data pipeline system, the observation data distribution server uses at least an international standard Web API that provides SOS (Sensor Observation Service), which is a service for distributing the measurement results of observation data by the observation sensor. It is preferable that the data distribution server creates a command based on the API and transmits the command to the observation sensor, and the observation sensor returns the measurement result to the observation data distribution server when the command is received. ..

In this analysis data pipelined system, there are a plurality of the analysis servers, and the terminal device, the analysis server, and the observation data distribution server call a predefined API in a chained manner to perform pipeline processing. It is preferable to perform the process from the acquisition of the environmental sensing data to the advanced analysis.

In this analysis data pipeline system, the outdoor environment sensing data is data related to agriculture, and the analysis server calculates a future weather scenario based on at least the past weather data acquired from the observation data distribution server. It is preferable to provide a meteorological generator API server that provides an API for generating, and a crop model API server that provides an API for creating a future crop scenario based on the meteorological scenario.

In this analysis data pipeline system, the crop model API server is linked with the weather generator API server using API, and the weather generator API server is SOS, which is an international standard Web API that provides the SOS using API. It is preferable to chain with the API server.

In this analysis data pipeline system, the terminal device transmits a crop scenario creation command based on the API to the crop model API server when a user requests a crop scenario using the application. When the crop model API server receives the crop scenario creation command, the crop model API server issues an API-based meteorological scenario creation command to the meteorological generator API server, and the meteorological generator API server issues the meteorological scenario creation command. Is received, the observation data distribution server is instructed to create meteorological data based on the API, and when the observation data distribution server receives the meteorological data creation command, the latitude and longitude information is used. Based on this, it is preferable to return the environment sensing data to the weather generator API server.

In this analysis data pipeline system, when the meteorological generator API server receives the meteorological data, the meteorological generator API server creates a meteorological scenario and returns it to the crop model API server, and the crop model API server receives the meteorological data. When a scenario is received, a crop scenario is created using this meteorological scenario and returned to the terminal device, and the terminal device analyzes the crop scenario and displays the result as a GUI (Graphical User Interface). Is preferable.

In this analysis data pipelined system, the analysis server further includes a pest information service API server that provides an API for generating information about pests, and a soil profile service that provides an API for generating information about soil. It is preferable to have an API server.

In this analysis data pipeline system, the terminal device is via an input unit that receives input from a user, an application execution unit that executes the application for performing analysis based on the environment sensing data, and the input unit. When there is a request for advanced analysis from the user, an API request generation unit that generates an analysis request command based on the API for the analysis server, a transmission / reception unit that transmits the request command to the analysis server, and the analysis. The analysis server includes an arithmetic unit that performs an operation to display the analysis result on a web browser when the analysis result is acquired from the server, and the analysis server receives the analysis request command and receives the observation data distribution server. An API processing unit that generates a data request command based on the API, a transmission / reception unit that receives the analysis request command from the terminal device and transmits the data request command to the observation data distribution server, and the observation data distribution. It is preferable to include a program execution unit that executes a program that analyzes the environment sensing data when the environment sensing data is acquired from the server.

In order to solve the above problems, the present invention is an analysis program used for an analysis data pipeline system that performs from acquisition of outdoor environment sensing data to advanced analysis, and the analysis data pipeline system uses an observation sensor. An observation data distribution server that distributes the environment sensing data observed using it, and an analysis server that automatically acquires environment sensing data from the observation data distribution server and performs a predetermined analysis when requested. , The observation is provided with a terminal device that executes a predetermined application and acquires an analysis result from the analysis server and displays the analysis result when a user requests advanced analysis using the application. The data distribution server, the analysis server, and the terminal device are connected via a web network, and the observation data distribution server, the analysis server, and the terminal device are linked based on an API (Application Programming Interface). look including the step of performing a pipelined processing to advanced analysis environmental sensing data acquisition, the the observed data distribution server and the observation sensor, server, access means, metadata, format, units of measurement, data of the name At least one of them is characterized in that a common API is defined.

The present invention can be realized not only as such an analysis data pipeline system, but also as an analysis method using a characteristic means provided in the apparatus constituting such an analysis data pipeline system as a step. It can be realized or as a program that causes a computer to execute those steps. Needless to say, such a program can be distributed via a recording medium such as USB or a transmission medium such as the Internet.

The analysis data pipeline system according to the present invention automatically acquires environment sensing data from the SOS API server that distributes the environment sensing data observed using the observation sensor and the SOS API server when requested. Then, when a crop model API server and a weather generator server, which are analysis servers that perform a predetermined analysis, and a predetermined application are executed, and a user requests advanced analysis using the application, the analysis server sends the data. It is equipped with a terminal device that acquires analysis results and displays them. Then, by calling these APIs in a chain reaction, it is possible to realize from acquisition of environmental sensing data to advanced analysis in the agricultural field.

It is a system block diagram of the analysis data pipeline system which concerns on embodiment of this invention. It is a figure which shows an example of the web browser screen at the time of using the application of the terminal apparatus provided in the analysis data pipeline system (a) and (b). It is a figure which shows an example of the web browser screen at the time of using the application of the terminal device. It is explanatory drawing of (a) to (b) the analysis data pipeline system of the same as above. Same as above It is a figure explaining the data of NARO 1km mesh used for the meteorological analysis of the analysis data pipeline system. Same as above This is a diagram for explaining the API used for the SOS API server that constitutes the analysis data pipeline system. It is a figure for demonstrating (a) and (b) the API used for the SOS API server of the same as above. It is a functional block diagram of the analysis data pipeline system of the same as above. It is a sequence diagram of the analysis data pipeline system of the same as above. It is a figure which shows the example of the API (especially when the crop scenario is requested) used for the analysis data pipeline system of the same as (a) and (b). It is a figure which shows the example of the API (especially when the crop scenario is requested) used for the analysis data pipeline system of the same as above. It is a figure which shows the example of the API (especially when the meteorological scenario is requested) used for the analysis data pipeline system of (a) and (b). Same as above It is a figure which shows an example of API (especially when a meteorological scenario is requested) used for an analysis data pipeline system. It is a figure which shows the example of the API (especially the meteorological data request and the simulation result) used for the analysis data pipeline system. It is a figure which shows an example of the screen display (GUI) of the terminal apparatus as above. (A) An explanatory diagram of an analysis procedure of the same analysis data pipeline system, and (b) a diagram showing a display example of a prediction result in the same terminal device. Same as above The figure shows an example of Python used to create an API used for an analysis data pipeline system. (A) The figure shows an example of the SDK used for creating the API used for the analysis data pipelined system of the same as above, and (b) is a diagram showing an example of Python used for creating the API used for the analysis data pipelined system of the same as above. It is the flowchart which shows the operation procedure of the terminal apparatus as above. Same as above It is a flowchart which shows the operation procedure of the crop model API server provided in the analysis data pipeline system. Same as above It is a flowchart which shows the operation procedure of the meteorological generator API server provided in the analysis data pipeline system. Same as above It is a flowchart showing the operation procedure of the SOS API server provided in the analysis data pipeline system. It is explanatory drawing of the analysis data pipeline system which concerns on the modification 1 of embodiment. It is explanatory drawing of the analysis data pipeline system of the same as above. It is a sequence diagram of the analysis data pipeline system which concerns on the modification 2 of embodiment. It is a system configuration diagram of the conventional agricultural information acquisition system.

(Embodiment)
An analysis data pipeline system (hereinafter referred to as a pipeline system) from the acquisition of outdoor environment sensing data to advanced analysis according to the embodiment of the present invention will be described with reference to the drawings. As a web server, this pipeline system mainly pipelines from acquisition to analysis of agricultural environment data.

<Overall system diagram>
As shown in FIG. 1, the pipeline system 1 according to the present embodiment includes a terminal device 10, a crop model API (Application Programming Interface) server 20, a weather generator API server 30, and an SOS (Sensor Observation Service) API server 40. , And the weather sensor 50 are connected and configured via a wide area network such as the Internet.

An API (Application Programming Interface) is a procedure for calling and using a function of a certain software, data to be managed, etc. from another software. There are various types of APIs, which are indispensable when building IT systems, but here we assume data linkage between different agricultural IT systems. Data linkage between agricultural IT systems does not necessarily require open APIs or fixed procedures, but when linking with multiple agricultural IT systems via WEB services is assumed, it is necessary in advance. It is desirable to develop an API with fixed procedures.

The terminal device 10 is a terminal such as a personal computer or a smartphone of a user contractor of the pipeline system 1, installs and executes a dedicated application, and displays the result on a screen. This dedicated application is a visualization web application of agricultural environment forecast information, and is an agricultural environment forecast information (for example, for example, as shown in screens 2, 4a to 4d of FIGS. 2 and 3) for an application contractor such as a farmer. Display a GUI (Graphical User Interface) as a web browser for farming / weather scenarios. More specifically, in this dedicated application, for example, the terminal device 10 acquires a crop scenario from the crop model Web API server 20 via a network, and based on this crop scenario, a GUI that makes it easy to understand future crop prediction information. Display with.

The dedicated application program installed on the terminal device 10 is, for example, an application created by a software development company requested by an agricultural cooperative, and is provided to a contractor such as an agricultural manager or an agricultural corporation. In the present invention, by preparing a crop model API and a weather generator API for a software development company, the software development company can easily obtain agricultural environment information that was difficult to analyze in-house because the data existed separately in the past. Obtainable. Therefore, by using the Web API according to the present embodiment, the software development company can create various application programs for agricultural environment prediction, which was impossible in the past.

The GUI depends on the user's choice of using the pipeline, and the GUI design is basically left to the user. Display past weather data and display detailed forecast information in a unique display method. The present invention is characterized in that the data that is the basis of the display can be easily obtained by using the pipelined system 1.

When the crop model API server (analysis server) 20 receives a crop scenario creation command for crop prediction in the form of an Http request from the terminal device 10, the crop model API server (analysis server) 20 acquires the weather scenario from the weather generator Web API server 30 and obtains this weather scenario. Provides an API for simulating crop growth (Crop SIM) based on. A typical simulation performed by the crop model API server 20 predicts 100 relationships (crop scenarios) between the planting date and the yield prediction based on, for example, 100 weather scenarios, varieties, and location information. However, there are multiple crop model API servers 20, and the ones that can output each are different. In addition to crop scenarios, leaf area index (LAI, Leaf Area Index), biomass, evapotranspiration, nitrogen movement, soil moisture, etc. can be calculated daily. Calculate and provide.

The Web API is one of the types of procedures and rules (API) for calling and using the functions provided by a computer program from another external program, and is a web technology such as Http (Hypertext Transfer Protocol). It was constructed using. When a request content is sent from another terminal device in the Http request format via the Internet to a server that publishes a predetermined function defined by the Web API, this processing result is the request source in the Http response format. It will be returned to the terminal device. The format of this transmitted / received data varies depending on the API, but XML (Extensible Markup Language), HTML (HyperText Markup Language), JSON (JavaScript (registered trademark) Object Notation), and various image file formats that are often used on the web are available. Mainly used. By using such a Web API, there is a possibility that various new services can be provided.

The meteorological generator (WEGN) API server (analysis server) 30 provides an API for creating a future meteorological scenario based on past meteorological data.

The SOS (Sensor Observation Service) API server (observation data distribution server) 40 is an international standard Web API, and regarding observation data (environmental sensing data) by the weather sensor 50, etc., metadata of the usage / installation status of the weather sensor 50 and It is an API that provides a service that delivers measurement results. Here, SOS is a service that distributes metadata and measurement results of sensor specifications and installation status regarding observation data by sensors and the like. Since a common data format and API are defined, it is possible to ensure interoperability for data linkage between sensors.

Further, the SOS API server 40 is improved in efficiency by using, for example, CloudSense, NARO Agent, etc. as a platform. In this embodiment, it has interoperability mainly using the NARO 1 km mesh and the SOS API of the local sensor, and when a meteorological data request request comes from the meteorological generator API server 30, the past corresponding to the latitude and longitude. It is an API that gives weather data.

The meteorological sensor 50 is, for example, a field sensor equipped with various sensors, and includes a small and low power consumption computer that implements the SOS protocol. This SOS protocol is a communication protocol for sensors for transmitting configurations and data related to sensors. Based on this SOS protocol, a communication session between the SOS API server 40 and the weather sensor 50 can be established to send and receive sensor ML files related to observation data.

The various conventional environmental databases (see FIG. 4) provided to the SOS API server 40 are diverse, for example, NARO soil layer DB, Global HC27, WISE, NARO 1km Mesh, Meteo Crop DB, water level sensor, field. There are valves, forecast information history, etc. Further, for example, FIG. 5 provides current and past meteorological data of the 1 km mesh of the Japanese topography with respect to the NARO 1 km Mesh data.

The meteorological sensor 50 is, for example, a field sensor installed in a field, and this field sensor observes various sensors that collect meteorological data such as temperature, humidity, rainfall, solar radiation, and air volume, and the growth status of crops. Equipped with a camera sensor and the like.

That is, the analysis data pipeline system 1 has an observation data distribution server (SOS API server 40) that distributes environmental sensing data observed using the observation sensor (meteorological sensor 50), and when requested, the analysis data pipeline system 1. An analysis server (crop model API server 20 and a weather generator API server 30) that automatically acquires environment sensing data from an observation data distribution server and performs a predetermined analysis, executes a predetermined application, and executes the application. It is provided with a terminal device 10 that acquires an analysis result from an analysis server and displays the analysis result when a user requests advanced analysis. The observation data distribution server, analysis server, and terminal device 10 are connected via a web network and chained based on the API to create a pipeline process from acquisition of environmental sensing data related to outdoor agriculture and disasters to advanced analysis. Is to do.

<Procedure for SOS API server 40 to acquire meteorological data>
Here, the procedure for the SOS API server 40 to acquire meteorological data will be described with reference to FIGS. 6 and 7. In the pipeline system 1 according to the present embodiment, in particular, data such as the outdoor weather sensor 50 is used to ensure interoperability and compatibility of observation data that were previously disjointed, and SOS. It is possible for the API server 40 to acquire environment data. Here, three APIs in the SOS API server 40 will be illustrated.

The interface related to data linkage of SOS environment information includes "(1) API for acquiring meta information of service usage conditions" and "(2) API for acquiring meta information of sensor specifications and measurement conditions". There are three APIs, "(3) API for acquiring measurement result meta information and transaction data".

(1) API for acquiring meta information of service usage conditions
Use GetCapabilities, the SOS API. The item of meta information of service usage conditions is, for example, as described in "3.2.1 Service usage conditions" of the data item guideline of environmental information at the policy meeting mentioned above. If you send the xml created in GetCapabilities format as input to the API GetCapabilities, you will get the xml created in Capabilities format as output. FIG. 6 shows an image of the API in which the weather generator API server 30 acquires meta information.

(2) API for acquiring meta information of sensor specifications and measurement conditions
Use DescribeSensor, which is an API of SOS. The items of the meta information of the sensor specifications and the measurement conditions are as described in, for example, "3.2.2 Sensor specifications and measurement conditions" of the data item guideline of the environmental information at the policy meeting mentioned above. If you send the xml created in DescribeSensor format as input to the API DescribeSensor, you will get the xml created in Sensor Model Language format as output. FIG. 7A shows an image of an API in which the SOS API server 40 acquires meta information of sensor specifications and measurement conditions.

(3) API for acquiring meta information and transaction data of measurement results
Use GetObservation, the API of SOS. The items of meta information of the measurement results are as described in "3.2.3 Measurement results" of the data item guideline of environmental information at the policy meeting mentioned above. If you send the xml created in the GetObservation format as input to the API GetObservation, you will get the xml created in the Observations & Measurements format as the output. FIG. 7 (b) shows an image of an API in which the SOS API server 40 acquires meta information and transaction data of the measurement results of the weather sensor 50.

In this way, by using these APIs (GetCapabilities, DescribeSensor, GetObservation), the SOS API server 40 can use the past when there is a request from the weather generator API server 30 based on latitude / longitude, date / time information, and the like. The measured meteorological data can be immediately returned to the meteorological generator API server 30 side.

Further, in this way, between the SOS API server 40 and the weather sensor 50, data such as the server, access means, metadata, format, measurement unit, and each name of each sensor are standardized based on the above-mentioned guidelines. And a common API is defined. For this reason, the client can accurately understand the characteristics of the data, and it is possible to ensure interoperability in the data linkage between the sensors.

<Functional block diagram>
Next, a functional block diagram of each processing unit of the pipelined system 1 according to the present embodiment will be described with reference to FIG.

As described above, the terminal device 10 is a terminal such as a personal computer, and executes an input unit 11 such as a keyboard that accepts input from a user, for example, a dedicated application for performing analysis based on environmental sensing data related to agriculture on a Web browser. When there is a request for analysis of environmental data from the user via the application execution unit 12 and the input unit 11, the API request generation unit 13 for generating the crop scenario generation command for the crop model API server 20, the request command ( A transmission / reception unit 14 that transmits a crop scenario generation command) to the crop model API server 20, a calculation unit 15 that performs GUI display processing based on the analysis result when an analysis result is acquired from the crop model API server 20, and a storage unit 16. To be equipped.

More specifically, the application execution unit 12 executes a Web browser used to display the execution result of the application on the screen. The API request generation unit 13 makes a processing request (Http request) via the Http protocol to the program running on the crop model API server 20 in response to the input from the user via the input unit 11. The data format at the time of the Http request transmitted from the transmission / reception unit 12 is, for example, XML.

The transmission / reception unit 14 receives the execution result on the crop model API server 20 based on the request command as a response (Http response) from the crop model API server 20 side. The calculation unit 15 interprets the XML received from the crop model API server 20 side, calculates the result of the crop scenario, and performs a process of displaying it on a display unit such as a liquid crystal display using a GUI in a form that is easy for the user to understand.

The storage unit 16 is a memory, and stores, for example, a contractor ID for executing an application, a password, and user information thereof.

The crop model API server 20 generates XML and returns it in accordance with the Http request from the terminal device 10. In addition, data is searched, calculated, and generated using the designated Web API (Crop Model API), and this is converted into a data format such as XML and returned to the terminal device 10. Specifically, the crop model API server 20 is, for example, a computer such as a server or a workstation, and when an analysis request command is received from the terminal device 10, a request based on the API is made to the weather generator API server 30. An API processing unit 21 that generates a command (meteorological scenario creation command), a transmission / reception unit 22 that receives a request command from the terminal device 10 and sends a data request command to the weather generator API server 30, and a weather scenario from the weather generator API server 30. Is provided with a program execution unit 23 that executes a crop scenario creation process when the above is acquired.

More specifically, the API processing unit 21 sets an API developed independently, and performs processing for interpreting and executing the API. The transmission / reception unit 22 transmits / receives data to / from the terminal device 10 and the weather generator API server 30 based on the Http protocol. The program execution unit 23 executes the application program. Here, the application program is a program (Crop Model) that realizes the processing to be executed by the worker.

The weather generator API server 30 also includes an API processing unit 31, a transmission / reception unit 32, and a program execution unit 33 (weather simulation) having the same functions as the crop model API server 20.

<Overall sequence diagram>
Next, the entire sequence of the pipelined system 1 according to the present embodiment will be described with reference to FIG. Here, it is assumed that the contractor of the application of the terminal device 10 requests to create a crop scenario for a predetermined crop such as rice via a web browser.

First, a contractor who uses the application of the terminal device 10 such as a farmer launches a dedicated application of the terminal device 10. Specifically, for example, the farmer can log in to the system by registering the latitude / longitude, name, and login information. When you log in, past weather data, short-term forecast data, harvest forecast results for each field owned by pre-registered farmers, actual measurement values, etc. are displayed. In addition, when the details of each field are opened, the distribution of the yield, the planting date, and the predicted value of the harvest date is displayed. When the details (profile) of the registered login information is opened, the user is provided with functions such as weather information prediction data, reference to a list of registered fields, and registration / deletion thereof. Here, the user of the terminal device 10 specifies specific information and requests the creation of a crop scenario in the future (step S101).

Using the GUI as shown in FIGS. 2 and 3, the contractor specifies and inputs information such as the scenario he / she wants to know, that is, location information, date / time information, variety information, and the desired number of crop scenarios. And press the crop scenario creation button. Then, the application of the terminal device 10 transmits a crop scenario creation command to the crop model API server 20 side (calls the crop model API) in the form of an HTTP request (step S102). The data format in this crop scenario creation directive is described in a general-purpose language such as XML.

At this time, examples of the API transmitted from the terminal device 10 are shown in FIGS. 10 and 11. The API generally consists of HTTP request messages and structural definitions of messages such as XML or JSON format. FIG. 10A shows a URI pattern and an Http method corresponding to each API, and FIG. 10B shows details of the API when issuing a crop scenario creation command. FIG. 11 shows the input parameters and output formats in the Crop SIM request (SIMRIM). Using the API that provides the growth model SIMRIM, simulations are performed for each meteorological scenario and results are obtained.

Next, the crop model API server 20 that has received this crop scenario creation command interprets the API in this crop scenario creation command so as to cooperate with each other. Then, the crop model API server 20 transmits a meteorological scenario creation command specifying the latitude / longitude information, the date and time, the number of meteorological scenarios created according to the number of crop scenarios, etc. to the meteorological generator API server 30 in the HTTP request format (WEGN API). (Call) (step S103). The data format in this meteorological scenario creation command is described in a general-purpose language such as XML. Examples of this data format are shown in FIGS. 12 and 13. FIG. 12A shows a URI pattern and an Http method corresponding to each API, and FIGS. 12B and 13 show details of the API when issuing a weather scenario creation command.

Then, the meteorological generator API server 30 that has received this meteorological scenario creation command issues a meteorological data request command (weather data creation command) in HTTP request format in order to acquire past meteorological data based on latitude and longitude. It is transmitted to the server 40 side (step S104). The data format in this meteorological data request directive is described in a general-purpose language such as XML. An example of this data format is shown in FIG.

Next, the SOS API server 40 uses the specified latitude and longitude information to accumulate past 1 km mesh (NARO 1 km mesh meteorological information) and meteorological data based on MeteoCropDB (daily average temperature, daily maximum temperature). , Daily minimum temperature, daily precipitation, etc.), the latest weather data from the weather sensor 50 up to yesterday, which is updated from time to time, etc., is returned to the weather generator API server 30 in HTTP response format, data format XML (step). S105). Then, the meteorological generator API server 30 obtains the xml created in the Observations & Measurements format in which the meteorological data is described from the SOS API server 40.

Then, the meteorological generator API server 30 generates a specified number (for example, 100 ways) of future meteorological scenarios based on the meteorological data received from the SOS API server 40 (step S106). At this time, the generated meteorological scenario results are shown in FIG. Further, the generated meteorological scenario is programmed to automatically respond to the crop model API server 20 side in the HTTP response format and data format XML (step S107). At this time, if necessary, format conversion processing and data storage processing for the memory unit are performed.

In this way, the meteorological generator API server 30 acquires the regional weather / field sensor (data) from the SOS API server 40, and for example, the meteorological data as a normal value based on the past data and the current year until the current day. Create a weather scenario with a weather generator program based on data trends. In addition, real-time updates of weather scenarios can be performed.

In the description of the present embodiment, the detailed algorithm for creating the weather scenario in the weather generator API server 30 will not be described. In summary, the program executed by the meteorological generator API server 30 will generate a meteorological scenario considering short-term forecasts in NARO's national 1km mesh meteorological information, and combine various meteorological data obtained from the SOS API server 40. Generates a forecast scenario that minimizes weather uncertainty. In addition, the Japan Meteorological Agency provides medium- to long-term forecasts (seasonal forecasts: 1-month forecast, 3-month forecast, warm weather forecast, cold weather forecast, etc.) as weather forecasts. Meteorological scenarios can be corrected and adjusted, enabling more accurate simulations.

Next, the crop model API server 20 acquires 100 different weather scenarios from the weather generator API server 30 and creates 100 different crop scenarios (step S108). As a result, the crop model API server 20 is used to execute the crop simulation for each created weather scenario, and the result is obtained. It can also be executed with separately prepared weather information.

Then, the crop model API server 20 returns the generated crop scenario to the terminal device 10 in the form of HTTP response and data format XML (step S109). At this time, the crop model API server 20 performs format conversion processing and data storage processing for the memory unit, if necessary.

In this way, the meteorological generator API server 30 automatically accesses the SOS API server 40 when given latitude and longitude, acquires past meteorological data, and generates a specified number of scenarios (for example, 100). There is also a method of giving past weather data prepared separately. Further, the crop model (CropSIM) API server 20 executes the simulation when the weather scenario prepared by the weather generator API server 30 is given. It can also be executed in the weather prepared separately. Therefore, by calling these APIs in a chain reaction, it is possible to construct an analysis data pipeline system 1 from environment sensing data acquisition to advanced analysis.

In the past, these programs had to be prepared and executed as independent programs. However, in this embodiment, the pipeline from the weather generator API server 30 to the SOS API server 40 is already prepared, and by calling the provided API twice, CropSIM → WGEN → SOS can be easily pipelined. Can be built as.

In addition, in the conventional single program, the user had to check and convert the format of the input data one by one, but in the API according to the present invention, the data format conversion work is prepared and executed by the internal API. , Work efficiency can be greatly improved.

Finally, the application of the terminal device 10 analyzes the received 100 crop scenarios and, for example, as shown in FIG. 16B, 100 yields per planting day, Box Plt, above and below a certain yield. Probability presentation, easy-to-understand operation to farmers / agricultural corporations, and provision with display GUI (step S110). FIG. 15 shows an example of the result of the crop scenario displayed as the GUI of the terminal device 10. By using GUI, the distribution of heading / ripening time and yield can be visualized. Note that FIG. 16A shows the overall flow of the pipeline system 1, which can be realized by utilizing the Web API.

As a result, the user of the terminal device 10 can use the GUI of the prototype of the tool (application) that visualizes the heading / ripening time and the distribution of the yield, quantify / visualize the influence of the uncertainty of the weather on the yield, and when sowing / seeding. Is it possible to expect high yield and stability (gain) when transplanting, when to increase the probability of low yield and instability (risk) to sow and transplant, whether to hurry to transplant to multiple fields, and it may take time. Information such as whether or not can be visualized and acquired. As a result, farmers and agricultural corporations can judge the balance between high yield, low risk, and work schedule, and make a planting plan that minimizes risk based on the setting scenario (variety, planting time, fertilization plan, etc.). Is possible. If one-third of Japanese rice farmers introduce it, the production value will increase by 3% and an economic effect of about 14 billion yen is expected.

In other words, the contractor of the terminal device 10 simply searches for the agricultural environment information that he / she wants to open a web browser, and uses the pipelined Web API to obtain more accurate weather information and based on a highly specialized program. The crop simulation results can be obtained. Then, based on the information, it is possible to recommend farm work schedules such as sowing, fertilization, and harvesting, and to make harvest prediction simulations for each. In other words, crop forecasting in the agricultural field is pipelined based on the Web API. Specifically, in this pipeline system 1, the crop model API server 20 is linked to the meteorological generator API server 30, and the meteorological generator API server 30 is also an international standard SOS (Sensor Observation Service) API server 40. Is chained to.

As described above, the pipeline system 1 according to the present embodiment is the development of a decision-making assisting tool for reflecting the uncertainty of the weather in farming. In the past, it was almost impossible for farmers to produce their own results because crop simulation predictions such as when to plant and how much yield can be expected and when flowers will bloom require a high degree of expertise. It was. However, by using the analysis data pipeline system 1 according to the present embodiment, the farmer can predict extremely advanced agriculture-related information very easily by simply launching a Web browser on the terminal device 10 and searching. Will be able to be obtained.

<SDK>
The API can be called from any language, but the code shown in FIGS. 10 to 14 may be used to call the API, or by preparing an SDK (Software Development Kit) as shown in FIGS. 17 and 18. The API is easier to use and the readability of the program is improved. The SDK needs to be prepared for each language, but the above code is made into a library as a function and is provided by the inventor. Figure 18 (b) shows an example in python. Using the SDK, you can execute the API in one line after importing the library in one line (from agroapi_sdk import *). Users can also prepare their own convenient SDK for themselves. FIG. 18A shows an example of using the SDK when requesting a crop scenario or the like.

<Flow of each processing unit>
FIG. 19 is a flowchart showing an operation procedure in the terminal device 10. First, the application execution unit 12 of the terminal device 10 determines whether or not there is a start process from the user (S201), and if there is (Yes in S201), starts a dedicated application and displays it on the screen (S202). ). Then, when there is a request for creating a crop scenario from the user (Yes in S203), a request creation command is transmitted to the crop model API server 20 in the format of the Http request and the data format of the XML file (S204). Next, when the crop scenario is received from the crop model API server 20 (Yes in S205), the calculation unit 15 performs calculation processing and graphic processing and displays it as a user-friendly GUI (S206).

FIG. 20 is a flowchart showing an operation procedure in the crop model API server 20. First, the API processing unit 21 of the crop model API server 20 determines whether or not there is a command to create a crop scenario, and if there is (Yes in S301), in the format of the Http request and the data format of the XML file. , Sends a weather scenario creation command to the weather generator API server 30 (S302). Next, when a weather scenario is received from the weather generator API server 30 (Yes in S303), a crop scenario is created in the program execution unit 22 (S304), and when format conversion is required (Yes in S305). After the format conversion, the crop scenario is transmitted to the terminal device 10 in the Http response format and the XML file data format (S307). On the other hand, when the format conversion is unnecessary (No in S305), the crop scenario is transmitted to the terminal device 10 in the Http response format and the XML file data format without the format conversion (S306).

FIG. 21 is a flowchart showing an operation procedure in the weather generator API server 30 according to the present embodiment. First, the API processing unit 31 of the meteorological generator API server 30 determines whether or not there is a command to create a meteorological scenario, and if there is (Yes in S401), in the format of the Http request and the data format of the XML file. , Sends a weather data request command to the SOS API server 40 (S402). Next, when the weather data is received from the SOS API server 40 (Yes in S403), the program execution unit 32 creates a weather scenario (S404), and when format conversion is required (Yes in S405), the format is used. After conversion, the weather scenario is sent to the crop model API server 20 in the Http response format and XML file data format (S407). On the other hand, when the format conversion is not required (No in S305), the weather scenario is transmitted to the crop model API server 20 in the Http response format and the XML file data format without the format conversion (S406).

FIG. 22 is a flowchart showing an operation procedure in the SOS API server 40 according to the present embodiment. First, the SOS API server 40 periodically transmits a get observation based on the API for obtaining observation data from the meteorological sensor 50 to the meteorological sensor 50 side (S501) as a loop process (S501). Is acquired and accumulated (S502), and the loop processing is terminated (S504). Further, the SOS API server 40 determines whether or not there is a weather data request command, and if there is (Yes in S505), the SOS API server 40 is in the Http request format and the XML file data format for the weather generator API server 30. And send the weather data (S506).

<Effect>
As described above, the analysis data pipeline system 1 from the acquisition of the environment sensing data to the advanced analysis according to the present embodiment contributes to the agricultural environment data through the standard AIP and the sensor observation service (SOS). Compatibility and interoperability are ensured using the cloud sense information platform. That is, the client of the terminal device 10 can access the metadata and the data of the mixed sensor system by using a single API from the terminal device 10. The information in the weather sensor 50 ensures compatibility and interoperability of agricultural information based on the above guidelines, and as a result, the client can accurately understand the characteristics of the data.

In addition, the API used in the crop model API server 20 and the weather generator API server 30 that execute analysis such as weather analysis and crop simulation obtains data from the SOS API server 40, and the analysis result of the environment sensing data is output to the terminal device 10. Can be provided to users of. Then, the user of the terminal device 10 can describe and specify a script (search information) that calls the API to build the processing pipeline by starting the dedicated application. In this way, the user of the application of the terminal device 10 selects a single API that falls like a waterfall for a plurality of APIs for realizing the requested analysis, and forms a chain of processing. Can be executed.

As a result, the pipelined system 1 can support a compatible agricultural environment data platform, multiple APIs for analysis, and a simple and flexible processing pipeline. As a result, scientists and engineers can be freed from the tedious format conversion action and installation of complicated programs, and can contribute to the operation of the analysis results of agricultural environment information.

In other words, the pipeline system 1 according to the present embodiment uses (1) interoperability of conventionally disjointed agricultural environment data (agro-environmental data), particularly using data from outdoor environment sensors. Ensuring interoperability and making it possible to acquire environmental data, (2) implementing advanced weather analysis, advanced crop prediction system, etc. as a Web API, (3) as a result, a data pipeline on the net It can be easily constructed and strongly supports the creation of advanced predictive analysis programs for agricultural information.

As a result, the pipeline system 1 according to the present embodiment secures the interoperability of the agricultural environment data, which has been disjointed in the past, and integrates and analyzes the agricultural environment data. In addition, it can provide environment sensing data acquisition to advanced crop / weather analysis as a web server for application contractors.

(Modification example 1)
A modified example 1 of the present embodiment will be described with reference to FIGS. 23 and 24. The same components as those of the pipelined system 1 according to the above embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.

In the pipelined system 1 according to the first modification, the pest and disease information service API (Pest and Disease API) server that provides an API for generating information on pests in addition to the API of the above-described embodiment as an API. 60. A soil profile service API (Soil Profile API) server 70 that provides an API for generating information about soil has been added. In this configuration, the same effects as those of the above embodiment can be obtained, and more diverse agricultural environment information / prediction can be provided to the contractor of the application of the terminal device 10 by using more diverse APIs.

(Modification 2)
A modified example 2 of the present embodiment will be described with reference to FIG. In the pipeline system 1 according to the second modification, the application contract user of the terminal device 10 wants to know only the future weather forecast.

First, the contractor of the terminal device 10 uses the GUI to specify and input information such as location information, date and time information, and the number of weather scenarios that he / she wants to know, and presses a weather scenario creation button or the like. Press (step S601). Then, the application of the terminal device 10 transmits a weather scenario creation command to the weather generator API server 30 (calls the WEGN API) in the form of an HTTP request (step S602).

Then, the weather generator API server 30 that receives the weather scenario creation command transmits the weather data request command to the SOS API server 40 side in the HTTP request format in order to acquire the past weather data based on the latitude and longitude. (Step S603).

Next, the SOS API server 40 returns the weather data to the weather generator API server 30 in the form of an HTTP response and the data format XML based on the designated latitude / longitude information (step S604). Then, the meteorological generator API server 30 obtains the xml created in the Observations & Measurements format in which the meteorological data is described from the SOS API server 40.

Then, the meteorological generator API server 30 generates a specified number of future meteorological scenarios based on the meteorological data received from the SOS API server 40 (step S605). Further, the generated meteorological scenario is automatically returned to the terminal device 10 in the HTTP response format and data format XML (step S606). At this time, if necessary, format conversion processing and data storage processing for the memory unit are performed. Finally, the application of the terminal device 10 analyzes the received 100 weather scenarios and provides the result to the user with an easy-to-understand GUI (step S607). As a result, the user of the terminal device 10 can assemble an optimum work schedule or the like based on the predicted weather scenario. In this way, by using the Web API, the user of the terminal device 10 simply inputs the search conditions into the web browser, and the application selects the desired API to acquire, analyze, and display the results regarding the agricultural environment. It can be carried out.

The present invention is not limited to the configuration of each of the above embodiments, and various modifications can be made without changing the gist of the invention. Although the pipeline processing has been described above, each API can be individually called via the network, and the API calling procedure of the above embodiment is an example and is not fixed. .. Furthermore, it goes without saying that there are many other API models for disaster information such as heavy rains and typhoons in fields other than agricultural environment information, and advanced predictive analysis can be performed by replacing them.

1 Analysis data pipeline system 10 Terminal equipment 11 Input unit 12 Application execution unit 13 API request generation unit 14 Transmission / reception unit 15 Calculation unit 16 Storage unit 20 Crop model API server (analysis server)
21 API processing unit 22 Program execution unit 23 Transmission / reception unit 30 Meteorological generator API server (analysis server)
40 SOS API server (observation data distribution server)
50 Meteorological sensor (observation sensor)
60 Pest Information Service API Server 70 Soil Profile Service API Server

Claims (10)

An analysis data pipeline system that performs everything from outdoor environment sensing data acquisition to advanced analysis.
The analysis data pipeline system is
An observation data distribution server that distributes environmental sensing data observed using observation sensors,
When there is a request, an analysis server that automatically acquires environmental sensing data from the observation data distribution server and performs a predetermined analysis, and
It is provided with a terminal device that executes a predetermined application and acquires an analysis result from the analysis server and displays the analysis result when a user requests advanced analysis using the application.
The observation data distribution server, the analysis server, and the terminal device are connected via a web network.
By chaining the observation data distribution server, the analysis server, and the terminal device based on an API (Application Programming Interface), the pipeline processing from the acquisition of the environment sensing data to the advanced analysis is performed.
The observation data distribution server and the observation sensor are characterized in that at least one of a server, an access means, metadata, a format, a unit of measurement, and data of each name has a common API defined. Analysis data pipeline system. The observation data distribution server uses at least an international standard Web API that provides SOS (Sensor Observation Service), which is a service that distributes the measurement results of observation data by the observation sensor.
This observation data distribution server creates a command based on the API and sends the command to the observation sensor.
The analysis data pipeline system according to claim 1, wherein the observation sensor returns a measurement result to the observation data distribution server when the command is received. There are a plurality of the analysis servers.
The terminal device, the analysis server, and the observation data distribution server perform pipeline processing by calling a predefined API in a chained manner, and perform from the acquisition of the environment sensing data to the advanced analysis. 2. The analysis data pipeline system according to claim 2. The outdoor environment sensing data is data related to agriculture, and is
The analysis server includes at least a meteorological generator API server that provides an API for generating a future meteorological scenario based on past meteorological data acquired from the observation data distribution server, and future crops based on the meteorological scenario. The analysis data pipeline system according to claim 2 or 3, further comprising a crop model API server that provides an API for creating a scenario. The crop model API server is linked with the weather generator API server using the API.
The analysis data pipeline system according to claim 4, wherein the weather generator API server is linked to an SOS API server, which is an international standard Web API that provides the SOS using an API. When the user requests a crop scenario using the application, the terminal device transmits a crop scenario creation command based on the API to the crop model API server.
When the crop model API server receives the crop scenario creation command, the crop model API server issues a weather scenario creation command based on the API to the weather generator API server.
When the meteorological generator API server receives the meteorological scenario creation command, the meteorological generator API server issues a meteorological data creation command based on the API to the observation data distribution server.
5. The observation data distribution server is characterized in that when it receives a command to create the meteorological data, it returns environmental sensing data to the meteorological generator API server based on the latitude / longitude information. Described analysis data pipeline system. When the weather generator API server receives the weather data, it creates a weather scenario and returns it to the crop model API server.
When the crop model API server receives the weather scenario, it creates a crop scenario using this weather scenario and returns it to the terminal device.
The analysis data pipeline system according to claim 6, wherein the terminal device analyzes the crop scenario and displays the result in a GUI (Graphical User Interface). The analysis server further includes a pest information service API server that provides an API for generating information about pests, and a soil profile service API server that provides an API for generating information about soil. The analysis data pipeline system according to claim 4, which is characterized. The terminal device is
An input unit that accepts input from the user,
An application execution unit that executes the application for performing analysis based on the environment sensing data, and an application execution unit.
When there is a request for advanced analysis from the user via the input unit, an API request generation unit that generates an analysis request command based on the API for the analysis server, and an API request generation unit.
A transmission / reception unit that transmits the request command to the analysis server, and
It is provided with an arithmetic unit that performs an operation to display the analysis result on a web browser when the analysis result is acquired from the analysis server.
The analysis server
When the analysis request command is received, the API processing unit that generates the data request command based on the API to the observation data distribution server, and
A transmission / reception unit that receives the analysis request command from the terminal device and transmits the data request command to the observation data distribution server.
Any one of claims 1 to 3, further comprising a program execution unit that executes a program that analyzes the environment sensing data when the environment sensing data is acquired from the observation data distribution server. The analysis data pipeline system described in Section. An analysis program used in an analysis data pipeline system that performs everything from outdoor environment sensing data acquisition to advanced analysis.
The analysis data pipeline system is
An observation data distribution server that distributes environmental sensing data observed using observation sensors,
When there is a request, an analysis server that automatically acquires environmental sensing data from the observation data distribution server and performs a predetermined analysis, and
It is provided with a terminal device that executes a predetermined application and acquires an analysis result from the analysis server and displays the analysis result when a user requests advanced analysis using the application.
The observation data distribution server, the analysis server, and the terminal device are connected via a web network.
The observation data distribution server, the analysis server, and the terminal device are linked based on an API (Application Programming Interface) to include a step of performing a pipeline process from the acquisition of the environment sensing data to the advanced analysis.
The observation data distribution server and the observation sensor are characterized in that at least one of a server, an access means, metadata, a format, a unit of measurement, and data of each name has a common API defined. Analysis program to be used in the data pipeline system.

Images