WO2016191893A1

WO2016191893A1 - Real-time interactive monitoring system for precision agriculture

Info

Publication number: WO2016191893A1
Application number: PCT/CL2016/050027
Authority: WO
Inventors: Sergio HERNÁNDEZ
Original assignee: Universidad Católica Del Maule
Priority date: 2015-06-02
Filing date: 2016-06-01
Publication date: 2016-12-08
Also published as: CL2017003059A1

Abstract

The present invention relates to a real-time interactive monitoring system for precision agriculture for obtaining the real conditions of a plot or piece of land. The system comprises: a plurality of sensors situated in various parts of the piece of agricultural land and connected to a central node by means of a network of nodes; a data processing system for working with large volumes of data; at least one database located in at least one server; and a web interface that allows the collected data to be viewed.

Description

Real-time Interactive Monitoring System for Agriculture of

Precision

SCOPE

The present invention relates to a real-time interactive monitoring system and procedure for precision agriculture to obtain the real conditions of a property or land through the arrangement of a series of sensors connected through a network of nodes with a central node, to send the information that will be verified in real time

BACKGROUND The quality and volume of agricultural production are largely related to growing conditions. Fluctuations in the climate, especially in temperature and humidity can negatively affect the quality and quantity of some crops such as grapes or berries. Other climatic and environmental influences can decrease or increase the potential of crops, therefore wind, soil moisture, solar radiation and other factors combine to influence the final result of crop quality. For example, for the cultivation of the grapevine it is considered that the variety of the grape, the type of soil, the terrain and the climate are the determining factors of the quality, being the climatic conditions very relevant for being the most dangerous for the optimization of the crop

Real-time monitoring of agroclimatic variables has been developed by different universities and government institutions throughout the world. An example of the use of this technology worldwide is the agro-climatic network of the state of Washington (http: // www. Weath er \ wsu. Ed u /), which combines the information at regional level with information at property level. At the national level, the Agroclima network (hjtj3: //www.ag _i rQ ^ provides information at the regional level.

However, the additional information provided by wireless sensor networks cannot be captured by regional monitoring methods and are very useful in precision agriculture applications. In addition there is a limited number of weather stations, located only in some specific places, which serve only as a reference for a certain climatic variable of agricultural interest and do not represent the particular reality of the property.

Similar to Agroclima, it is the Agromet Network (http://aqromet.inia.cl). Like the Agroclima network, this Network belongs to INIA (Institute of Agricultural Research of Chile) and provides information at regional level. This Network delivers the information very similar to that provided by the system presented in this document. On the Agromet website it is possible to acquire the current and historical data of the stations in the form of a table and graphs. Although this system presents the data in a fairly simple and attractive way, it only presents it at the regional level and only serves as a reference for a property located near a weather station. In addition, the variables that this system delivers are limited to atmospheric variables, and it does not provide information on soil data such as those presented by the system of the present invention.

At the national level is the company Epssilon, which is dedicated to the sale and installation of wireless sensor networks for the monitoring of various environments, within which the agricultural one is located, which they measure are measures at the property level, so They are very useful. A big difference existing among the service presented by this company in comparison to the system of the present invention, is that they are mainly dedicated to the installation of the equipment (Wireless sensor network) and do not provide the model to remotely monitor, nor provide summaries of the data collected inside the property.

Internationally we can find the FieIdScript system of the Monsanto company. This system is fed with data from meteorological stations, soil composition samples and historical productivity records, in order to determine the variability of productivity for the same seed variety and make recommendations to users, based on similar agroclimatological conditions that may have occurred. achieved high standards of productivity by concept of correct application of nutrients or other agronomic practices. Although this system provides particular data of the farmer's land, the focus of this system is placed on equipment that allows to enhance seed productivity, this is achieved through soil studies that indicate which are the areas with the highest production of the farm .

Precision agriculture is a term from agronomy, which refers to the use of technologies such as sensors, satellites, etc. to make measurements on the premises in order to make decisions, improve productivity and avoid risk factors.

Wireless sensor networks are a widely used tool for monitoring various environmental phenomena. This type of networks has had a great boom due to the great benefits provided in areas such as the precision farming. Its operation is based on a collection of sensors (nodes) located in a certain area, which send the data to a central node. The central node is responsible for storing information on all nodes.

Particularly in the area of agriculture these technologies have had great acceptance, because they provide quite interesting benefits compared to traditional weather stations. Among the benefits provided by wireless sensor networks we can highlight that these are low cost, scalable and with high fault tolerance. In addition to being able to acquire more of these, it is possible to monitor large estates and obtain abundant amounts of information with which it is possible to develop decision-making based on the variables that are being measured.

The processing of historical data and real-time agroclimatic data requires great processing power that can only be achieved through modern parallel and distributed processing techniques.

Document CN103228063 refers to a network of wireless sensor nodes for an agricultural installation. A network system of nodes with wireless sensors for an agricultural installation is described, based on a network of ZigBee wireless sensors, using a combination of advantages and features of wireless sensor network technology over a crop growing environment in a greenhouse .

Another solution is that mentioned in document CN202143105 (U), which mentions an agricultural information system, comprising at least one sensor used for the collection of environmental information in an agricultural sector; at least one node of a wireless sensor network in connection with the sensor; a Gateway or portal of the wireless sensor network that is used to receive wireless information sent by the wireless sensor node network.

Document CN102664949 discloses a pre-alarm system in a projected agricultural environment, with heterogeneously distributed wireless sensor networks. The system comprises a plurality of wireless sensor nodes, a plurality of wireless router nodes, a central wireless control node and a global system for mobile communications (GSM) short message alarm module. One or more of the wireless sensor nodes that connect to the plurality of wireless router nodes which are connected to one or more of the wireless router nodes based on the principle of the nearest router.

Another solution, described in document CN203070064, relates to a wireless sensor network monitoring node that contains a mobile processor module and the power module. The wireless processor module is connected to a control module in order to carry out data acquisition and is connected to a control module in order to control the equipment.

Document n ^Q CN201533413 (U) describes a Gateway node for a wireless monitoring system in an agricultural environment. The Gateway node comprises a processor module, a short distance wireless communication module for bidirectional communication with the nodes of a lower sensor and a wireless long distance communication module that is connected to a superior communication server. It also comprises a power supply module that is connected respectively to the processor module, the wireless short distance communication module and the wireless long distance communication module. The wireless long distance communication module is a GPRS wireless communication mechanism, the short distance wireless communication module and the GPRS wireless communication module are both connected to the processor module.

Document CN203100772 (U) describes an intelligent sampling node with positioning. The sampling node is used to sense the temperature and humidity in the crop growth environment based on precise spatial positions. A ZigBee communications module is used to transmit data with other sensor nodes, acquiring various information on farmland and uploading it to an agricultural information center, in order to perform remote monitoring of information on the farmland environment.

Document CN101 155090 (A) describes an agricultural irrigation water saving system based on a wireless sensor network, consisting of a sensor node, an entrance door, a wireless management center and an irrigation device. It also discloses a water irrigation system based on a wireless sensor network, using a wireless sensor node and a gateway. The above allows the moisture state of the plant to be controlled in real time. The monitoring data is processed by a team of a management center and a rational and efficient water saving irrigation project is arranged based on different characteristics of the plant. SUMMARY OF THE INVENTION

The present invention relates to a system and procedure to support decisions for the technical staff of vineyards and different types of orchards. The system relates atmospheric and climatological data with information about plants and soils. The design and construction of the technology required to collect data and the development of mathematical models for information processing have been the focus of research and scientific collaboration between Chile and New Zealand. The system is related to the implementation of a wireless sensor architecture on land in a given region in order to monitor and model the state of the agroclimatic variables that affect the variability of agricultural production.

The interactive real-time monitoring system of the present invention allows agri-climatological data to be collected every 10 minutes from different points of a property and sent to a central server that performs the following calculations:

Evapotranspiration

Dew point temperature and frost probability. Degrees day

The server has a geospatial processing system that allows you to make temporary space queries and return complex relationships between the variables measured in the field and the variables calculated. This level of processing requires parallel and distributed computing that can be scaled by adding new monitoring sites and sampling points that feed the central Sistem. For this purpose, the use of NoSQL databases is preferred, which allows not only to reach this level of scalability but also to perform spatio-temporal queries for the calculation of agroclimatic variables through data mining techniques.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 consists of an operating scheme of the MapReduce computational model for estimating the average daily temperature, according to an embodiment of the invention.

Figure 2 consists of a scheme of components of the precision agriculture system according to an embodiment of the invention.

Figure 3 consists of a view of the web portal interface of the system according to an embodiment of the invention.

Figure 4 consists of a diagram of the climatological variables used according to an embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a system (100) and interactive real-time monitoring procedure for precision agriculture to obtain the real conditions of a property or land through the provision of a plurality of sensors associated with their respective nodes (200 ) connected through a network of nodes (300) with a central node (400), to send the information that will be verified in real time. In particular, the system (100) of the present invention has the following characteristics:

Accessibility: The data obtained is accessible from any site through an interface (500) that allows reading and writing from mobile devices, web clients or embedded hardware devices located on the premises.

Robustness: The system (100) is prepared for failures of at least one central server (600) and its replicas. At the same time, it is also possible that there are communication failures between the data acquisition devices, therefore it is also required to store data locally on each site in order to upload the data when communication is restored.

Scalability: The hardware and software that manages the system (100) scale linearly by adding new properties and monitoring points on each property.

· Simplicity: The system (100) provides the user interface (500) simple and easy to use for non-technical users.

The real-time monitoring system (100) comprises a data processing system for working with large volumes of data. In a preferred embodiment of the invention, this model is a distributed computing model (MapReduce) used by Google ™. An example of the use of this model is shown in Figure 1.

Under this paradigm, large volumes of data can be indexed in parallel through distributed mapping operations (MAP) and whose results are consolidated through aggregation operations (REDUCE). For example, in the case of the present invention, it is possible to distribute the evapotranspiration calculation by the following steps:

- the indexes (day) of the hourly data obtained in the nodes (200) are distributed among a cluster of servers (600) and the task of calculating daily averages is divided.

- the reference evapotranspiration calculation is performed (according to Penman-Monteith equation)

the user is shown graphically via the interface (500).

Traditionally this operation is performed on the server (600) through operations with the SQL (Structured Query Language), which performs consolidation operations such as averages, sums, maximums and minimums. However, the cost of making an aggregate query over an extended period of time means that these types of operations are not easily scalable, especially for a large number of monitoring points that deliver agroclimatic variables in real time.

Currently, the model is available as part of the open Hadoop project and in other NoSQL database products, such as MongoDB and CouchDB. It is also possible to use storage and parallel computing services in the cloud, such as Amazon EC2 from where these jobs can be run. These types of technologies are quite useful in cases such as this application, where large volumes of information are generated daily.

The main features of the real-time monitoring system (100) are the following:

Accessibility, simplicity, robustness and scalability. Interface (500) type RestFUL (Representational State Transfer) for the collection and consultation of agroclimatological data in real time.

Model of agroclimatic information consultations (evapotranspiration and day degrees) using the model.

Data mining algorithms to determine frost forecasts.

Figure 2 shows the components that make up the system0), which operates as follows:

• In the first instance, a plurality of nodes (200) connected to sensors, is located in different parts of the agricultural property as well as a central node (Gateway) (400) is placed, which is permanently connected to the internet through a plan of telemetry In one embodiment of the invention, each node (200) corresponds to a Waspmote of Libelium, which has connected different sensors that measure various weather variables. Being this technology focused on low consumption, it allows nodes (200) to make measurements of up to 5 years. It is mainly programmed in C language and is relatively simple to use. An important feature that these types of nodes possess is the communication protocol, which is ZigBee. This protocol allows the formation of wireless networks of the mesh network type. Mesh type networks have the great quality of being very fault tolerant, since if one of its components fails, it can easily be replaced. In addition to the basic features, different plates can be incorporated into the waspmote. In these plates the sensors that measure the required variables are inserted. In the particular case of this invention, the Agriculture Board was used. This Board allows to integrate a wide variety of sensors, among which the following stand out:

At least one SHT75 temperature and humidity sensor.

At least one MPX41 15A atmospheric pressure sensor.

At least one SQ-1 10 solar radiation sensor

At least one humidity sensor of the LWS sheet

At least one PT-1000 ground temperature sensor

At least one Irrometer soil moisture sensor

Watermark

At least one anemometer, wind vane and rain gauge Nodes (200) send information to the central node (300) in a given period of time t, preferably every 10 minutes. In a preferred embodiment of the invention, the central node (300) is formed by a small computer, of the Raspberry Pl type, which has an antenna that receives the information provided by the nodes (200). Among the technical characteristics that it has, we can highlight this type of computing, it has an ARM1 176JZF 700 MHz processor, a VideoCore IV video card, 512 MB of RAM, as storage it uses SD memory of at least 4 GB of memory It also has two USB ports, a network port and an HDMI port. As an operating system, it uses a version of Linux They should have been developed for processors of the ARM type. By owning this type of operating system, it is possible to develop applications in various languages for this.

The computer has a small program that captures the information received and then stores it in a local database (sqllite). Every time t, that is 10 minutes, the program checks if there is an internet connection, if there is a connection, it sends the data to a central server (600). Otherwise, it stores the data locally and I have tried to send t (10 more minutes) in that period of time. In one embodiment of the invention, this connection is made through an interface (500) of the RestFUL type. The REST interfaces define a group of architectural principles for the design of services that, which are mainly focused on system resources. This means that the different system resources are manipulated and transferred through the HTTP protocol to different clients that may not always be the same. One of the key features of a RESTFul interface is the use of explicit http method for content manipulation. These methods are as follows:

or POST, which is used to create some resource.

or GET, to obtain some resource.

or PUT, to modify some resource.

or DELETE, to eliminate some resource.

When using this method, an api (Application Programming Interface) is defined, which will be used by all clients that connect to a central server (600). As a summary it can be said that the RESTFul interfaces are an alternative to traditional Web services, where RPC calls are widely used.

• The data received by the server (600) is stored in a database (700). In one embodiment of the invention, the database (700) is a MongoDB database, which has the characteristic of storing files in JSON format. Since the data is stored in a time t, every approximately 10 minutes, it is possible to have the meteorological information of each property updated.

• On the other hand there is a Web interface (500), which allows you to view the collected data. In this web interface (500) it is possible to obtain historical summaries of the meteorological data collected for one or several nodes located in a particular property. The data is shown in graph form, additionally it can be exported to a data processing form. An example of a web page image can be seen in figure 3.

• The information of the different climatic variables is shown in the form of hourly averages, while others such as maximums, minimums and evapotranspiration are shown in the form of daily values. Since the information is stored it is stored every time t (10 minutes), the server at the time of receiving a request of some variable performs an operation of aggregation of the data and then sends them to the web portal, which only displays the information . This allows to display large periods of time quickly. These aggregations are made using the model method described above. It should be mentioned that each node captures in particular at least 1 1 different climatic variables, also based on these and through the use of aggregation methods it allows obtaining other calculations of interest for the farmer such as evapotranspiration, minimum temperature prediction and calculation of degrees day. Figure 4 shows a scheme of all the variables used.

Although there are very similar monitoring systems to the one of the present invention, the great difference of the present system is the possibility of real estate and non-regional monitoring as is the vast majority. However, by using a distributed computing mode on the server side, the system allows you to combine information from multiple data sources which improves the accuracy of predictive models. In addition, this system fully integrates both software and hardware.

Another remarkable feature of the system of the present invention is the possibility of seeing the state of soil moisture in a current and accurate way. This variable becomes very important and relevant, since with it the water consumption of the property can be monitored. Additionally, the simple act of controlling soil moisture has a significant impact on many monetary areas.

Claims

A real-time interactive monitoring system for precision agriculture to obtain the real conditions of a property or land, CHARACTERIZED because it comprises a plurality of sensors located in different parts of an agricultural property, connected through a network of nodes with a node central, a data processing system to work with large volumes of data, at least one database located on at least one server and a Web interface, which allows you to view the collected data.

The system of claim 1, CHARACTERIZED in that the data processing system for working with large volumes of data is based on a MapReduce model.

The system of claim 1, CHARACTERIZED in that each sensor or node corresponds to a Waspmote sensor of Libelium, which has connected a plurality of sensors that measure various weather variables.

The system of claim 3, CHARACTERIZED in that the plurality of sensors comprises the following types: at least one temperature and humidity sensor; at least one atmospheric pressure sensor; at least one solar radiation sensor; at least one leaf moisture sensor; at least one soil temperature sensor; at least one soil moisture sensor; and at least one anemometer, wind vane and rain gauge.

5. The system of claim 1, CHARACTERIZED in that the central node is formed by a small computer, of the Raspberry Pl type, which has an antenna receiving the information provided by the nodes.

6. The system of claim 1, CHARACTERIZED in that the interface is of the RestFUL interface type.

7. The system of claim 1, CHARACTERIZED the database on the server where they are stored in a database, is one is a MongoDB database.

8. A real-time interactive monitoring procedure for precision agriculture to obtain the real conditions of a property or land, CHARACTERIZED because it comprises the following stages:

to. Arrange a plurality of sensors associated with respective nodes, located in different parts of an agricultural property for data collection;

b. Place a central node connected to the internet using a telemetry plan;

C. Send the data collected by the nodes to the central node in a period of time t;

d. Capture the information received in the central node and then store it in a local database;

and. Verify in a period of time í if there is an internet connection, if there is a connection, it sends the data to a central server, and if not, it stores the data locally and I try to send in the same period of time t; F. Store the data received by the server in a database every period of time t; Y

g. Display the data collected in a Web interface, which shows historical summaries of the meteorological data collected for one or several nodes located in a particular property.

9. The method of claim 8, CHARACTERIZED in that large volumes of data can be indexed in parallel through distributed mapping operations (MAP) and whose results are consolidated through aggregation operations (REDUCE).

10. The method of claim 8, CHARACTERIZED because the communication protocol between the nodes is the ZigBee protocol.

eleven . The method of claim 8, CHARACTERIZED in that it comprises steps of inserting plates in each node to place the respective sensors.

12. The method of claim 1, CHARACTERIZED in that it comprises the step of measuring temperature and humidity, atmospheric pressure, solar radiation, leaf moisture, soil temperature, soil moisture, wind and rain through the sensors connected to each node

13. The method of claim 8, CHARACTERIZED in that the period of time t is about 10 minutes.

14. The method of claim 8, CHARACTERIZED in that it comprises the step of storing the data in a local database on a computer located in the central node.

15. The method of claim 8, CHARACTERIZED in that the interface is an interface of the RestFUL type.

16. The method of claim 8, CHARACTERIZED in that the data is displayed in the interface in the form of a graph, and can additionally be exported to a data processing form.

17. The method of claim 8, CHARACTERIZED in that the information of the different climatic variables is shown in the form of hourly averages, while others such as maximum, minimum and evapotranspiration are shown in the form of daily values.

18. The method of claim 8, CHARACTERIZED because since the information is stored is stored every time t, the server at the time of receiving a request for some variable performs an operation of aggregation of the data and then sends them to the web portal , which only displays the information.