JP2022024214A - Sensing system, sensing data acquisition method, and unmanned flyable aircraft - Google Patents

Abstract

To provide a sensing system capable of further efficiently acquiring non-contact sensing data which are obtained by non-contact sensing, and data obtained from soil, etc., a sensing data acquisition method, and an unmanned flyable aircraft.MEANS FOR SOLVING THE PROBLEM: A sensing system includes: a contact sensor which is embedded in the ground within a sensing area and performs contact sensing of at least any one sensing target of soil and plant in the sensing area; and an unmanned flyable aircraft. The aircraft performs non-contact sensing on the surface of the ground within the sensing area in the air and receives, from the contact sensor, contact sensing data obtained by the contact sensing of the contact sensor.SELECTED DRAWING: Figure 9

Description

It relates to a technical field such as a system for acquiring non-contact sensing data by performing non-contact sensing from the air using an unmanned aerial vehicle.

It is being considered to measure and analyze soil using unmanned aerial vehicles. For example, in Patent Document 1, the entire field is photographed from a video camera mounted on an industrial unmanned helicopter to acquire images and data showing the reflectance of natural light, while sampling a part of the soil in the field. A technique for measuring hot water extractable nitrogen and generating a distribution map of hot water extractable nitrogen in the entire field is disclosed. Further, in Patent Document 2, the reflectance calculated based on the reflected light measured from the air using an unmanned flight device and the analytical information obtained by analyzing the soil sample are used for each position of the entire field. A technique for generating field characteristic information showing characteristics is disclosed.

Japanese Unexamined Patent Publication No. 2011-254711 Japanese Unexamined Patent Publication No. 2019-002811

However, with the above techniques, apart from the non-contact sensing from the air by the unmanned aerial vehicle, it is necessary to collect the soil sample manually, for example, and it is obtained from the non-contact sensing data and the soil sample obtained by the non-contact sensing. In the case of repeatedly acquiring data, for example, the burden on the data acquisition side is large and it is inefficient.

Therefore, we provide a sensing system that can acquire non-contact sensing data obtained by non-contact sensing and data obtained from soil, etc. more efficiently, a sensing data acquisition method, and an aircraft that can fly unmanned. ..

In order to solve the above problems, the invention according to claim 1 is a contact sensor that is preliminarily embedded in the ground of a predetermined area and performs contact sensing of at least one of soil and plants in the area. A sensing system including an aircraft capable of flying unmanned, wherein the aircraft has a non-contact sensor unit for performing non-contact sensing of the ground surface in the area from the air, and contact sensing obtained by the contact sensing. It is characterized by including a receiving unit for receiving data from the contact sensor. As a result, the non-contact sensing data obtained by non-contact sensing from the air of an aircraft that can fly unmanned and the contact sensing data obtained by contact sensing of a contact sensor embedded in the ground can be acquired more efficiently. can do.

The invention according to claim 2 is a positional correspondence between the contact sensing data obtained by the contact sensing and the non-contact sensing data obtained by the non-contact sensing in the sensing system according to claim 1. It is characterized by further including a processing unit for performing the above. As a result, the non-contact sensing data obtained by non-contact sensing from the air of an aircraft that can fly unmanned and the contact sensing data obtained by contact sensing of a contact sensor embedded in the ground can be acquired more efficiently. It is possible to make a positional correspondence.

The invention according to claim 3 estimates the state of the sensing target based on the contact sensing data and the non-contact sensing data associated with the processing unit in the sensing system according to claim 1 or 2. It is characterized by further including an estimation unit. This makes it possible to obtain highly accurate estimation results for the state of the sensing target.

According to a fourth aspect of the present invention, in the sensing system according to the third aspect, the processing unit performs positional correspondence between the contact sensing data and the non-contact sensing data a plurality of times in chronological order. The estimation unit identifies the transition of the state based on the contact sensing data and the non-contact sensing data for a plurality of times associated with the processing unit, and is based on the transition of the state specified by the estimation unit. Further includes a diagnostic unit for diagnosing the health of the plant. This makes it possible to appropriately manage changes in the state of plant health over time.

The invention according to claim 5 generates treatment information indicating a treatment to be performed on the sensing target based on the state estimated by the estimation unit in the sensing system according to claim 3 or 4. It is characterized by further including a generation unit. As a result, more appropriate measures can be taken for the sensing target.

The invention according to claim 6 is the state estimated based on the contact sensing data in the sensing system according to any one of claims 3 to 5, and the said state associated with the contact sensing data. It is characterized by further including a learning control unit that performs machine learning of an estimation model that estimates the state by using the data set with the non-contact sensing data as an input. As a result, it is possible to estimate the state of the sensing target by inputting the non-contact sensing data without requiring the contact sensing data in the end.

The invention according to claim 7 is the contact sensing data from the contact sensor based on the position information indicating the position where the contact sensor is buried in the sensing system according to any one of claims 1 to 6. It is characterized by further including a flight control unit that controls the flight of the aircraft so that the aircraft can receive. As a result, the aircraft can efficiently acquire contact sensing data.

The invention according to claim 8 is the contact sensing data from the contact sensor based on the position information indicating the position where the contact sensor is buried in the sensing system according to any one of claims 1 to 7. It is characterized by further including a flight route creating unit that creates a flight route of the aircraft so as to be able to receive. As a result, the aircraft can efficiently acquire contact sensing data.

The invention according to claim 9 is characterized in that, in the sensing system according to any one of claims 1 to 8, the state of the plant includes the length of the plant. As a result, the condition of the plant can be managed more appropriately.

The invention according to claim 10 is characterized in that, in the sensing system according to any one of claims 1 to 9, the state of the plant includes the direction in which the plant grows from the ground. As a result, the condition of the plant can be managed more appropriately.

The invention according to claim 11 is a contact sensor that is preliminarily embedded in the ground of a predetermined area and performs contact sensing of at least one of soil and plants in the area, and can fly unmanned in the air. It is a sensing data acquisition method performed by a sensing system including an aircraft that performs non-contact sensing of the ground surface in the area, and acquires contact sensing data obtained by the contact sensing and received by the aircraft from the aircraft. It is characterized by including a step and a step of acquiring non-contact sensing data obtained by the non-contact sensing from the aircraft.

The invention according to claim 12 is preliminarily embedded in the ground of a predetermined area, can communicate with a contact sensor that performs contact sensing of at least one of soil and plants in the area, and is unmanned. A flightable aircraft, a non-contact sensor unit for performing non-contact sensing of the ground surface in the area from the air, and a receiving unit for receiving contact sensing data obtained by the contact sensing from the contact sensor. It is characterized by being prepared.

According to the present invention, the non-contact sensing data obtained by non-contact sensing from the air of an aircraft capable of flying unmanned and the contact sensing data obtained by contact sensing of a contact sensor embedded in the ground are further combined. It can be acquired efficiently.

It is a figure which shows the outline structure example of the sensing system S. It is a figure which shows the outline structure example of the soil sensor unit 1. It is a conceptual diagram which shows the state that the radio wave is transmitted from the soil sensor units 1a to 1c embedded in the ground of the green G in a golf course. It is a figure which shows the outline structure example of UAV2. It is a figure which shows the example of the functional block in the control part 26. It is a conceptual diagram which shows an example of the flight route Ro of UAV2 in the sensing area Ar. It is a figure which shows the outline configuration example of management server 3. It is a figure which shows the example of the functional block in the control unit 33. It is a flowchart which shows an example of the processing in the control unit 26 when UAV2 performs non-contact sensing of the earth surface in a sensing area from the air. It is a flowchart which shows an example of the processing in the control unit 33 when the management server 3 estimates the state of a sensing target.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[ 1. Configuration of sensing system S ]
First, the configuration of the sensing system S according to the embodiment of the present invention will be described with reference to FIG. FIG. 1 is a diagram showing a schematic configuration example of the sensing system S. As shown in FIG. 1, the sensing system S includes a soil sensor unit 1, an unmanned aerial vehicle (hereinafter referred to as "UAV (Unmanned Aerial Vehicle)") 2, and a management server 3. The UAV 2 and the management server 3 can communicate with each other via the communication network NW. The communication network NW is composed of, for example, the Internet, a mobile communication network, a radio base station thereof, and the like.

UAV2 is an example of an aircraft that can fly unmanned, and is also called a drone or a multicopter. The UAV2 can fly or autonomously fly from the ground according to remote control by an operator. UAV2 is managed by GCS (Ground Control Station). The GCS may be mounted on a control terminal operated by an operator as an application, or may be configured by a server such as a management server 3.

[ 1-1. Configuration and function of soil sensor unit 1 ]
Next, the configuration and function of the soil sensor unit 1 will be described with reference to FIG. FIG. 2 is a diagram showing a schematic configuration example of the soil sensor unit 1. As shown in FIG. 2, the soil sensor unit 1 includes a contact sensor 11, a data transmission unit 12, and the like, and is embedded in the ground in a predetermined area (hereinafter referred to as “sensing area”) in advance (that is, one or more). , The sensing system S operates with the soil sensor unit 1 placed in advance). For example, a plurality of soil sensor units 1 may be embedded in the ground of the sensing area in advance at regular intervals or irregular intervals. The sensing area is an area that requires maintenance such as a golf course or a ball game field. An area where the care of plants growing from the ground is particularly important (for example, a green on a golf course) may be a sensing area.

The contact sensor 11 performs contact sensing of at least one of the soil and the plant (for example, the root of a plant) in the sensing area. Here, the contact sensing means measuring at least one of the water content, temperature, salinity, electrical conductivity, acidity, etc. of the sensing target with which the contact sensor 11 is in contact. Such contact sensing may be performed continuously in a time series, and the time interval of the contact sensing may be a constant interval or an indefinite interval.

It should be noted that the water content, temperature, salinity, electrical conductivity, and acidity in the proximity range of the sensing target with which the contact sensor 11 is in contact (for example, in the range of several cm to several tens of cm where the contact sensor 11 is not in contact). At least one of the degrees and the like may be measured. The range in which contact sensing is possible is the point where the soil sensor unit 1 is embedded and the range close to the point. Therefore, the range in which contact sensing is possible by one soil sensor unit 1 is basically narrower than the sensing area.

The data transmission unit 12 modulates the contact sensing data obtained by the contact sensing of the contact sensor 11 and transmits a radio wave carrying the contact sensing data (for example, using a 920 MHz band). Here, the contact sensing data is measurement data indicating at least one of the water content, temperature, salinity, electrical conductivity, acidity, etc. measured as described above. The radio waves including the contact sensing data may be transmitted continuously in a time series, and the time interval of the transmission may be a fixed interval or an indefinite interval. FIG. 3 is a conceptual diagram showing how radio waves are transmitted from soil sensor units 1a to 1c embedded in the ground of Green G in a golf course.

In addition, the data transmission unit 12 may transmit a radio wave that modulates the position information indicating the position (latitude and longitude) of the soil sensor unit 1 together with the contact sensing data and conveys the contact sensing data and the position information. In this case, the position of the soil sensor unit 1 is measured in advance when it is buried in the ground, for example, and is stored in the soil sensor unit 1 in advance. Alternatively, the data transmission unit 12 modulates the identification information that can specify the position of the soil sensor unit 1 (for example, the identification information of the soil sensor unit 1) together with the contact sensing data, and carries the contact sensing data and the identification information. May be sent. In this case, the identification information that can specify the position of the soil sensor unit 1 is stored in advance in the soil sensor unit 1, and the identification information is managed in association with the position information indicating the position of the soil sensor unit 1. It is stored in advance in the server 3.

[ 1-2. Configuration and function of UAV2 ]
Next, the configuration and function of UAV2 will be described with reference to FIGS. 4 and 5. FIG. 4 is a diagram showing a schematic configuration example of UAV2. As shown in FIG. 4, the UAV 2 includes a drive unit 21, a wireless communication unit 22, a data reception unit 23, a sensor unit 24, a positioning unit 25, a control unit 26, and the like. Although not shown, the UAV2 includes a rotor (propeller) which is a horizontal rotor and a battery that supplies electric power to each part of the UAV2. The drive unit 21 includes a motor, a rotating shaft, and the like. The drive unit 21 rotates a plurality of rotors by a motor driven according to a control signal output from the control unit 26, a rotating shaft, and the like.

The wireless communication unit 22 is responsible for controlling the communication performed with the management server 3 via the communication network NW. The data receiving unit 23 (an example of the receiving unit) receives the contact sensing data obtained by the contact sensing by demodulating the radio wave transmitted from the soil sensor unit 1. The reception of the contact sensing data may be performed continuously in a time series, and the time interval of the reception may be a fixed interval or an indefinite interval. In addition to the contact sensing data, position information indicating the position of the soil sensor unit 1 or identification information capable of specifying the position of the soil sensor unit 1 may be received. The contact sensing data and the like received by the data receiving unit 23 are output to the control unit 26.

The sensor unit 24 includes various sensors necessary for flight control of the UAV2. Various sensors include an optical sensor, a 3-axis angular velocity sensor, a 3-axis acceleration sensor, a geomagnetic sensor and the like. The detection data detected by the sensor unit 24 is output to the control unit 26. Further, the optical sensor is composed of, for example, a camera (RGB camera or infrared camera), functions as one of the non-contact sensor units, and is also used for performing non-contact sensing of the ground surface in the sensing area from the air. Here, the non-contact sensing means observing the state (situation) of the ground surface by imaging the ground surface within the range where the non-contact sensing is possible (for example, the range within the angle of view of the camera). Such non-contact sensing is performed one or more times, for example, when arriving at the sensing area of UAV2. In order to improve the accuracy of the non-contact sensing, it is preferable that the non-contact sensing is continuously performed in a time series, and the time interval of the non-contact sensing may be a constant interval or an indefinite interval. The non-contact sensing range is preferably a wide range including the sensing area, but may be a part of the sensing area.

The positioning unit 25 includes a radio wave receiver, an altitude sensor, and the like. For example, the positioning unit 25 receives a radio wave transmitted from a satellite of a GNSS (Global Navigation Satellite System) by a radio wave receiver, and detects the current position (latitude and longitude) in the horizontal direction of the UAV2 based on the radio wave. The current position of UAV2 is the flight position of UAV2 in flight. The current position of the UAV 2 in the horizontal direction may be corrected based on the image captured by the optical sensor or the radio wave transmitted from the radio base station. The position information indicating the current position detected by the positioning unit 25 is output to the control unit 26. Further, the positioning unit 25 may detect the current position (altitude) of the UAV 2 in the vertical direction by an altitude sensor such as a barometric pressure sensor. In this case, the position information includes altitude information indicating the altitude of UAV2.

The control unit 26 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), a non-volatile memory, and the like, which are processors. FIG. 5 is a diagram showing an example of a functional block in the control unit 26. The control unit 26 functions as a sensing data processing unit 26a and a flight control unit 26b, as shown in FIG. 5, according to a program (program code group) stored in, for example, a ROM or a non-volatile memory.

The sensing data processing unit 26a associates the contact sensing data from the data receiving unit 23 with the position information indicating the position of the soil sensor unit 1 that transmitted the contact sensing data, and transmits the contact sensing data to the management server 3 by the wireless communication unit 22. (For example, it is transmitted every time it is acquired from the data receiving unit 23). Alternatively, the sensing data processing unit 26a is managed by the wireless communication unit 22 in association with the contact sensing data from the data receiving unit 23 and the identification information capable of identifying the position of the soil sensor unit 1 that has transmitted the contact sensing data. It may be transmitted to the server 3. Here, the position information (or identification information) transmitted to the management server 3 may be the position information (or identification information) transmitted from the soil sensor unit 1 together with the contact sensing data and received by the data receiving unit 23. ..

As another example, especially when the receivable distance (communication distance) of the radio wave transmitted from the soil sensor unit 1 is as short as a threshold value (for example, several meters) or less (that is, the UAV 2 does not approach the soil sensor unit 1). When the radio wave cannot be received), the position information transmitted to the management server 3 is the horizontal current position of the UAV 2 (that is, the current position at the time when the contact sensing data transmitted from the soil sensor unit 1 is received by the data receiving unit 23 (that is,). It may be the position information indicating the position directly under the UAV2). When radio waves transmitted from each of the plurality of soil sensor units 1 are received, the pair of contact sensing data and position information (or identification information) is distinguished for each soil sensor unit 1 and sent to the management server 3. Will be sent.

Further, the sensing data processing unit 26a functions as one of the non-contact sensor units, and the non-contact sensing data obtained by the non-contact sensing is transmitted to the management server 3 by the wireless communication unit 22 (for example, from the sensor unit 24). Send it every time it is acquired). Here, the non-contact sensing data may be raw detection data output from the sensor unit 24, or may be data analyzed and processed based on the output raw detection data. Such non-contact sensing data is data constituting at least one map image such as an RGB image of the ground surface in a sensing area, a NDVI (Normalized Difference Vegetation Index) distribution image, and a thermal image (temperature partial image).

NDVI is a value indicating the relationship between the health condition of plants grown from the ground and the reflectance of each wavelength in the visible to near-infrared region. For example, plants have the property of absorbing radio waves in the visible region while strongly reflecting radio waves in the near-infrared region, so the higher the NDVI value, the healthier the condition. Position information is associated with each pixel value (for example, RGB value, NDVI value, or temperature, which is a measured value) in the map image. Such position information (that is, position information in non-contact sensing data) is, for example, position information indicating the current position in the horizontal direction of UAV2 and SLAM (Simultaneous Localization and Mapping) processing (processing for performing map generation and self-position estimation at the same time). ).

Further, when the UAV 2 enters the communicable range of an access point to the Internet (for example, a Wi-Fi (registered trademark) router), the sensing data processing unit 26a is used by the wireless communication unit 22 via the access point. The contact sensing data, the non-contact sensing data, and the like may be collectively transmitted (uploaded) to the management server 3. Further, the contact sensing data and the non-contact sensing data may be transmitted to the management server 3 at predetermined time intervals (for example, 24 hours). In this case, the UAV2 will fly over the sensing area every predetermined time (for example, 24 hours). Further, the sensing data processing unit 26a wirelessly transmits the contact sensing data, the non-contact sensing data, and the like to which the contact sensing data and the non-contact sensing data are positionally associated with each other. The communication unit 22 may transmit the data to the management server 3.

The flight control unit 26b executes flight control of the UAV2. In flight control, the detection data from the sensor unit 24, the position information from the positioning unit 25, and the like are used to control the rotation speed of the rotor and control the position, attitude, and traveling direction of the UAV2. During the flight of UAV2, the flight control unit 26b sequentially transmits the position information from the positioning unit 25 to the management server 3 (or the management server 3 via GCS). Further, the flight control unit 26b may perform flight control of the UAV 2 according to the control information from the management server 3. This control information may include the flight route of UAV2. The flight route of the UAV2 is, for example, a route created so that the UAV2 can receive contact sensing data (radio waves) from the soil sensor unit 1.

FIG. 6 is a conceptual diagram showing an example of the flight route Ro of UAV2 in the sensing area Ar. In the example of FIG. 6, the flight route created so that the radio waves transmitted from the soil sensor units 1a to 1j embedded in the ground of each of the greens G1 to G5 in the sensing area Ar (golf course) can be efficiently received. Shows Ro. As a result, UAV2 can efficiently acquire contact sensing data.

If the flight control unit 26b can acquire the position information indicating the position of the soil sensor unit 1 from the management server 3 (or GCS) before receiving the contact sensing data, the soil sensor 26b is based on the acquired position information. It is advisable to control the flight of the UAV 2 so that the contact sensing data (radio wave) can be received from the unit 1. As a result, the UAV2 can efficiently acquire contact sensing data. For example, when the radio wave transmitted from the soil sensor unit 1 is present at a position where it cannot be received (communication is impossible), the flight control unit 26b can receive the radio wave transmitted from the soil sensor unit 1 (communication is possible). , The flight control of UAV2 is performed toward the position of the soil sensor unit 1 so as to approach the soil sensor unit 1. In such flight control, when the distance between the current position of UAV2 and the soil sensor unit 1 approaches a threshold value or less, it is preferable to perform control such as lowering the altitude of UAV2.

As an example of the case where the position information indicating the position of the soil sensor unit 1 can be acquired from the management server 3 or the like before receiving the contact sensing data, the position information measured when the soil sensor unit 1 is buried in the ground is used. It is registered in the management server 3 or the like, and may be acquired from the management server 3 or the like. Alternatively, as another example, the position information (received together with the sensing data) specified when the sensing data is received from the soil sensor unit 1 during the previous flight of the UAV2 (that is, when the UAV2 has already flown in the past). When the stored position information or the horizontal position information of UAV2 at the time of reception) is stored in the storage unit (not shown) of the management server 3 or UAV2, and it is acquired from the management server 3 or the storage unit. There is.

[ 1-3. Configuration and functions of management server 3 ]
Next, the configuration and function of the management server 3 will be described with reference to FIGS. 7 and 8. FIG. 7 is a diagram showing a schematic configuration example of the management server 3. As shown in FIG. 7, the management server 3 includes a communication unit 31, a storage unit 32, a control unit 33, and the like. The communication unit 31 is responsible for controlling the communication performed with the UAV 2 via the communication network NW. The contact sensing data, the non-contact sensing data, and the like transmitted from the UAV 2 are received by the communication unit 31. The storage unit 32 includes, for example, a hard disk drive or the like. Map data of the sensing area is stored in the storage unit 32. The storage unit 32 may store the position information indicating the position of the soil sensor unit 1 in association with the identification information that can specify the position of the soil sensor unit 1.

The control unit 33 includes a CPU, ROM, RAM, a non-volatile memory, and the like, which are processors. FIG. 8 is a diagram showing an example of a functional block in the control unit 33. As shown in FIG. 8, the control unit 33 has a sensing data processing unit 33a, a state estimation unit 33b (an example of an estimation unit), and a learning control unit according to a program (program code group) stored in, for example, a ROM or a non-volatile memory. It functions as 33c, a state diagnosis unit 33d (an example of a diagnosis unit), a treatment information generation unit 33e (an example of a generation unit), a flight route creation unit 33f, and the like.

The sensing data processing unit 33a refers to the position information associated with the contact sensing data from the UAV2 (that is, the position information indicating the position of the soil sensor unit 1) and the position information in the non-contact sensing data from the UAV2. Therefore, the contact sensing data and the non-contact sensing data are positionally associated with each other. As a result, the contact sensing data and the non-contact sensing data can be more efficiently and positionally associated with each other. When the identification information is received from the UAV2 in association with the contact sensing data, the position information indicating the position of the soil sensor unit 1 is specified based on the identification information.

Here, the positional correspondence between the contact sensing data and the non-contact sensing data is, for example, to the map image constituting the non-contact sensing data (that is, to the position of the soil sensor unit 1 in the map image), and the contact sensing. It is done by mapping the data. Alternatively, the positional association between the contact sensing data and the non-contact sensing data is at a position (or the closest position) that matches the position information (that is, the position of the soil sensor unit 1) associated with the contact sensing data. It may be performed by associating non-contact sensing data. Such association may be performed by UAV2 as described above. The sensing data processing unit 33a may perform positional correspondence between the contact sensing data and the non-contact sensing data a plurality of times in a time series.

The state estimation unit 33b estimates the state of the sensing target in the sensing area based on the contact sensing data and the non-contact sensing data positionally associated with each other by the sensing data processing unit 33a. This makes it possible to obtain highly accurate estimation results for the state of the sensing target. For example, regarding a plant (for example, turf) at a point where the soil sensor unit 1 is buried, the water content, temperature, and salt concentration of the root of the plant (or the soil in contact with the root of the plant) identified from the contact sensing data. At least one of the above and at least one of the RGB value, NDVI value, and temperature of the plant (stem or leaf) identified from the non-contact sensing data are comprehensively combined according to a predetermined algorithm. By analysis, the quality of the plant (which can be said to be the quality of the soil) is estimated as the current state of the plant. The quality of a plant may include the presence or absence of a disease, the degree of health condition, and the like. Further, as the state of the plant, at least one of the length of the plant (stem or leaf of the plant) and the direction in which the plant grows from the ground may be estimated. This makes it possible to more appropriately manage the condition of the plant.

The estimation of the state of the sensing target may be a rule-based estimation or a machine learning-based estimation. In the case of estimation based on machine learning, the state estimation unit 33b inputs the contact sensing data and the non-contact sensing data associated with the sensing data processing unit 33a into the trained estimation model that estimates the state of the sensing target. , The state of the sensing target can be estimated. The trained estimation model is trained in advance so as to estimate the state of the sensing target by inputting contact sensing data and non-contact sensing data as described later. The state information indicating the state estimated in this way is the position information indicating the position of the soil sensor unit 1 that transmitted the contact sensing data used for estimating the state (or the identification that can specify the position of the soil sensor unit 1). Information) and stored in the storage unit 32 in association with the estimated date and time of the state. Thereby, for example, the state information indicating the estimated state can be accumulated for each position of the soil sensor unit 1 and for each predetermined time (for example, 24 hours).

The state estimation unit 33b may specify the transition (change) of the state of the sensing target based on the contact sensing data and the non-contact sensing data for a plurality of times associated with the sensing data processing unit 33a. In this case, the state estimation unit 33b acquires, for example, state information indicating the state estimated at predetermined time (for example, 24 hours) from the storage unit 32, and the state to be sensed from the state indicated by each acquired state information. Identify the transition. Thereby, for example, it is possible to observe the transition of the health condition of the plant within the past predetermined period (for example, 1 week to 1 month).

The learning control unit 33c controls machine learning of an estimation model that estimates the state of the sensing target. For example, the learning control unit 33c uses the data set of the correct answer data regarding the state of the sensing target and the contact sensing data and the non-contact sensing data associated with the sensing data processing unit 33a as the teacher data, and the machine of the estimation model described above. A trained estimation model is obtained by training. However, the learning control unit 33c may not be provided in the management server 3. In this case, the state estimation unit 33b is a trained estimation model provided in advance to the management server 3 from a device other than the management server 3 (for example, a trained estimation model generated by machine learning performed by another device). May be used to estimate the state of the sensing target. Alternatively, in this case, the state estimation unit 33b may estimate the state of the sensing target without using the trained estimation model. Further, the set of contact sensing data and non-contact sensing data in the teacher data used for machine learning of the trained estimation model may be obtained by this sensing system S, or may be obtained by another system or method. May be. The algorithm for supervised machine learning for obtaining a trained estimation model is not particularly limited.

The state diagnosis unit 33d makes a diagnosis regarding the health of the plant based on the transition of the state estimated by the state estimation unit 33b. As a result, it is possible to obtain a diagnostic result regarding the health condition of the plant, and it is possible to appropriately manage the change in the state of the health of the plant with the passage of time. As a result of such a diagnosis, whether or not the plant needs to be cared for, whether or not the plant grows too much, and the like can be obtained. For example, if the health of a plant gradually deteriorates over time, it is diagnosed as requiring care of the plant. In addition, if the length of the plant (stem or leaf of the plant) grows beyond the reference value over time, it is diagnosed that the plant is overgrown.

The treatment information generation unit 33e generates treatment information indicating the treatment to be performed on the sensing target based on the state estimated by the state estimation unit 33b. As a result, more appropriate measures can be taken for the sensing target. Here, the treatment to be performed on the sensing target is, for example, spraying of a chemical, spraying of fertilizer, spraying of water, spraying of sand, mowing of lawn, or the like. The treatment information indicating such processing may include location information of a place where the treatment should be performed, and the like, in addition to the content of the treatment. The treatment information may be generated based on the state at a certain time point estimated by the state estimation unit 33b, or may be generated based on the time-series state transition estimated by the state estimation unit 33b. Alternatively, it may be generated based on the diagnosis result by the state diagnosis unit 33d. In addition, treatment information should be notified to soil and plant managers.

The flight route creating unit 33f creates a flight route of UAV2 so that contact sensing data (radio waves) can be efficiently received from a plurality of soil sensor units 1 based on the position information indicating the position of the soil sensor unit 1. For example, based on the information of the receivable distance (communication distance) of the radio wave transmitted from each soil sensor unit 1, the flight route so that the UAV2 passes through the area where the contact sensing data can be received from each soil sensor unit 1. To create. The control information including the flight route created in this way is transmitted to UAV2.

[ 2. Operation of sensing system S ]
Next, the operation of the sensing system S will be described.

[ 2-1. Operation of UAV2 ]
First, with reference to FIG. 9, the operation when UAV2 performs non-contact sensing of the ground surface in the sensing area from the air will be described. FIG. 9 is a flowchart showing an example of processing in the control unit 26 when the UAV 2 performs non-contact sensing of the ground surface in the sensing area from the air. As a premise of this processing, it is assumed that the UAV 2 has acquired the control information including the flight route (for example, the flight route Ro shown in FIG. 6) from the management server 3. The flight route includes, for example, position information indicating the position of each soil sensor unit 1 that performs contact sensing of the sensing target.

In the process shown in FIG. 9, first, the control unit 26 of the UAV 2 acquires the position information indicating the current position detected by the positioning unit 25 and the detection data detected by the sensor unit 24, and also acquires from the management server 3. Flight control of UAV2 is started according to the flight route included in the control information (step S1). As a result, the UAV2 flies from the starting point toward the sensing area.

Next, the control unit 26 determines whether or not the user has arrived at the sensing area based on the current position of the UAV2 (step S2). When the control unit 26 determines that it has arrived at the sensing area (step S2: YES), the control unit 26 proceeds to step S3. The case of arriving at the sensing area may be the case where the UAV2 comes near the sensing area or the case where the UAV2 enters the sensing area. On the other hand, when it is determined that the control unit 26 has not arrived at the sensing area (step S2: YES), the control unit 26 returns to step S2.

In step S3, the control unit 26 starts non-contact sensing of the ground surface in the sensing area and temporarily stores the non-contact sensing data obtained by the non-contact sensing. It should be noted that such non-contact sensing may be continuously performed in chronological order. In this case, the non-contact sensing data is temporarily stored in time series.

Next, the control unit 26 determines whether or not the radio wave from the soil sensor unit 1 that performs contact sensing of the sensing target in the sensing area is received by the data receiving unit 23 (step S4). When the control unit 26 determines that the data receiving unit 23 has received the radio wave from the soil sensor unit 1 (for example, the radio wave having a predetermined decibel value or more) (step S4: YES), the contact sensing acquired from the data receiving unit 23. Temporarily store the data and the position information indicating the position of the soil sensor unit 1 that transmitted the contact sensing data (or the identification information that can identify the position of the soil sensor unit 1 that transmitted the contact sensing data) in association with each other. (Step S5). On the other hand, when the control unit 26 determines that the radio wave from the soil sensor unit 1 has not been received by the data reception unit 23 (step S4: NO), the control unit 26 proceeds to step S6.

The control unit 26 may transmit a contact sensing request to the soil sensor unit 1 when approaching the soil sensor unit 1. In this case, when the soil sensor unit 1 receives the contact sensing request, it performs contact sensing of the sensing target and emits a radio wave including non-contact sensing data.

In step S6, the control unit 26 determines whether or not the reception of the contact sensing data from each soil sensor unit 1 is completed. When the control unit 26 determines that the reception of the contact sensing data from each soil sensor unit 1 has not been completed (step S6: NO), the control unit 26 returns to step S4. On the other hand, when the control unit 26 determines that the reception of the contact sensing data from each soil sensor unit 1 is completed (step S6: YES), the control unit 26 proceeds to step S7.

In step S7, the control unit 26 determines whether or not the UAV2 is within the communicable range of the access point to the Internet. When the control unit 26 determines that the UAV2 is not within the communicable range of the access point to the Internet (step S7: NO), the control unit 26 returns to step S7. On the other hand, when the control unit 26 determines that the UAV2 has entered the communicable range of the access point to the Internet (step S7: YES), the non-contact sensing data temporarily stored in step S3 and the non-contact sensing data temporarily stored in step S5 are temporarily stored. The contact sensing data and location information (or identification information) are transmitted to the management server 3 (step S8). After that, when the UAV2 arrives at the return point (for example, the departure point), the process shown in FIG. 9 ends.

[ 2-2. Operation of management server 3 ]
Next, with reference to FIG. 10, the operation when the management server 3 estimates the state of the sensing target will be described. FIG. 10 is a flowchart showing an example of processing in the control unit 33 when the management server 3 estimates the state of the sensing target. As a premise of this processing, it is assumed that the management server 3 has obtained a trained estimation model.

The process shown in FIG. 10 is started when, for example, non-contact sensing data, contact sensing data, and position information (or identification information) are received from UAV2. When the process shown in FIG. 10 is started, the control unit 33 of the management server 3 refers to the position information associated with the contact sensing data and the position information in the non-contact sensing data, and obtains the contact sensing data. Positionally associate with the non-contact sensing data (step S11). As a result, one or more pairs of contact sensing data and non-contact sensing data that are positionally associated with each other are created.

Next, the control unit 33 selects one pair to which the positional association is made in step S11 (step S12). Next, the control unit 33 estimates the state of the sensing target (that is, the soil or the plant at the position of the soil sensor unit 1) based on the contact sensing data and the non-contact sensing data related to the pair selected in step S12. (Step S13). For example, the control unit 33 inputs the locally associated contact sensing data and the non-contact sensing data as input variables to the trained estimation model, and obtains the objective variable output from the trained estimation model. Estimate the state of the sensing target.

Next, the control unit 33 transmits the state information indicating the state estimated in step S13 to the soil sensor unit 1 (for example, the soil sensor unit 1a shown in FIG. 6) that transmits the contact sensing data used for estimating the state. The position information indicating the position of the soil sensor unit 1 (or the identification information capable of specifying the position of the soil sensor unit 1) and the estimated date and time of the state are stored in the storage unit 32 (step S14).

Next, the control unit 33 receives a predetermined number (for example, three times) of state information indicating the state of the sensing target for the pair selected in step S12 in a predetermined period (for example, one week to one month) retroactively from the present time. It is determined whether or not the above is stored (step S15). When the control unit 33 determines that the predetermined number or more of the state information is stored in the predetermined period (step S15: YES), the control unit 33 proceeds to step S16. On the other hand, when the control unit 33 determines that the state information is not stored for a predetermined number of minutes in the predetermined period (step S15: NO), the control unit 33 proceeds to step S19.

In step S16, the control unit 33 acquires the state information (that is, the state information for a plurality of times) stored in the predetermined period retroactively from the current time from the storage unit 32. Next, the control unit 33 identifies the transition of the state of the sensing target from the state indicated by each state information acquired in step S16 (step S17). Next, the control unit 33 makes a diagnosis regarding the health of the plant based on the transition of the state of the sensing target identified in step S17 (step S18). The process of step S18 may not be executed.

In step S19, the control unit 33 relates to the pair selected in step S12 based on the state estimated in step S13, the transition specified in step S17, or the result diagnosed in step S18 (diagnosis result). Generates treatment information indicating the treatment to be performed on the sensing target. The treatment information thus generated (or the transition information indicating the transition of the above state or the above diagnosis result together with the treatment information) indicates the position of the soil sensor unit 1 that transmitted the contact sensing data used for estimating the above state. It is stored in the storage unit 32 in association with the indicated position information (or identification information that can specify the position of the soil sensor unit 1).

Next, the control unit 33 notifies the manager to be sensed of the treatment information generated in step S19 (step S20). For example, the management server 3 outputs (displays) on the display of the terminal by transmitting the treatment information when the terminal used by the administrator accesses the management server 3. Alternatively, the management server 3 sends an e-mail indicating the treatment information to the e-mail address of the administrator.

Next, the control unit 33 determines whether or not there is a pair that has not yet been selected among the pairs that have been positionally associated in step S11 (step S21). When the control unit 33 determines that there is a pair that has not been selected yet among the pairs that have been positionally associated in step S11 (step S21: YES), the control unit 33 returns to step S12 and performs the same processing as above. .. On the other hand, when it is determined in step S11 that there is no pair that has not been selected yet among the pairs that have been positionally associated with each other (step S21: NO), the control unit 33 ends the process shown in FIG.

As described above, according to the above embodiment, the UAV 2 performs non-contact sensing of the ground surface in the sensing area from the air, and is obtained by the contact sensing of the soil sensor unit 1 embedded in the ground of the sensing area. Since the contact sensing data is configured to be received from the soil sensor unit 1, the non-contact sensing data obtained by the non-contact sensing and the contact sensing can be obtained while reducing the burden for the non-contact sensing and the contact sensing. It is possible to acquire the obtained contact sensing data more efficiently.

In other words, in the prior art, the contact sensing data obtained by the contact sensing of the soil sensor was collected in the center via the relay device such as the repeater and the base station installed on the ground, but according to this embodiment, Since these relay devices are not required, the system can be greatly simplified and the cost can be reduced.

(Application example)
According to the above embodiment, since the burden for non-contact sensing and contact sensing can be reduced, a set of positionally associated contact sensing data and non-contact sensing data can be easily acquired. Then, according to the machine learning of the above embodiment, it is possible to obtain a trained estimation model capable of estimating the state of the sensing target by inputting the contact sensing data and the non-contact sensing data acquired in this way. In the application example, by applying the above embodiment, it is possible to collect teacher data for machine learning, which is an extension of the machine learning of the above embodiment. For example, the technique of estimating the state of the sensing target using only the non-contact sensing data acquired from the air is compared with the technique of estimating the state of the sensing target using only the contact sensing data acquired by the soil sensor unit. , It is difficult to obtain highly accurate results because sufficient research has not been conducted.

Therefore, the learning control unit 33c in the application example can estimate the state of the sensing target with high accuracy only by the non-contact sensing data by using the contact sensing data and the non-contact sensing data associated with each other. Generate an estimation model. For example, the learning control unit 33c uses a data set of a state estimated based on the contact sensing data and the non-contact sensing data positionally associated with the contact sensing data as teacher data, and the non-contact sensing. A trained estimation model is obtained by performing machine learning of an estimation model that estimates the state of the sensing target using data as input. That is, according to the machine learning of the application example, it is possible to obtain a trained estimation model that does not require contact sensing data and can estimate the state of the sensing target by inputting at least non-contact sensing data. As a result, it is possible to estimate the state of the sensing target by inputting the non-contact sensing data without requiring the contact sensing data in the end. Here, data other than the contact sensing data may be input to the estimation model of the application example together with the non-contact sensing data. By using a technique for estimating the state of the sensing target based on the contact sensing data, it is possible to obtain the above-mentioned data set in which the non-contact sensing data and the state of the sensing target are associated with each other.

The above embodiment is an embodiment of the present invention, and the present invention is not limited to the above embodiment, and various configurations and the like may be changed from the above embodiment without departing from the gist of the present invention. , Also included in the technical scope of the present invention. In the above embodiment, an unmanned aerial vehicle has been described as an example of an unmanned aerial vehicle, but an unmanned aerial vehicle can fly even if there is no pilot in the aircraft. It is also applicable to.

1 Soil sensor unit 2 UAV
3 Management server 11 Contact sensor 12 Data transmission unit 21 Drive unit 22 Wireless communication unit 23 Data reception unit 24 Sensor unit 25 Positioning unit 26 Control unit 26a Sensing data processing unit 26b Flight control unit 31 Communication unit 32 Storage unit 33 Control unit 33a Sensing Data processing unit 33b State estimation unit 33c Learning control unit 33d State diagnosis unit 33e Treatment information generation unit 33f Flight route creation unit S Sensing system

In order to solve the above problems, the invention according to claim 1 is a contact sensor that is preliminarily embedded in the ground of a predetermined area and performs contact sensing of at least one of soil and plants in the area. A sensing system including an aircraft capable of flying unmanned, wherein the aircraft has a non-contact sensor unit for performing non-contact sensing of the ground surface in the area from the air, and contact sensing obtained by the contact sensing. A receiving unit that receives data from the contact sensor , and a processing unit that positionally associates the contact sensing data obtained by the contact sensing with the non-contact sensing data obtained by the non-contact sensing. It is characterized by being prepared. As a result, the non-contact sensing data obtained by non-contact sensing from the air of an aircraft that can fly unmanned and the contact sensing data obtained by contact sensing of a contact sensor embedded in the ground can be acquired more efficiently. It is possible to make a positional correspondence .

The invention according to claim 2 is an estimation that estimates the state of the sensing target based on the contact sensing data and the non-contact sensing data associated with the processing unit in the sensing system according to claim 1. It is characterized by further including a part. This makes it possible to obtain highly accurate estimation results for the state of the sensing target.

According to the third aspect of the present invention, in the sensing system according to the second aspect, the processing unit performs positional correspondence between the contact sensing data and the non-contact sensing data a plurality of times in chronological order. The estimation unit identifies the transition of the state based on the contact sensing data and the non-contact sensing data for a plurality of times associated with the processing unit, and is based on the transition of the state specified by the estimation unit. Further includes a diagnostic unit for diagnosing the health of the plant. This makes it possible to appropriately manage changes in the state of plant health over time.

The invention according to claim 4 generates treatment information indicating a treatment to be performed on the sensing target based on the state estimated by the estimation unit in the sensing system according to claim 2 or 3 . It is characterized by further including a generation unit. As a result, more appropriate measures can be taken for the sensing target.

The invention according to claim 5 is the state estimated based on the contact sensing data in the sensing system according to any one of claims 2 to 4 , and the said state associated with the contact sensing data. It is characterized by further including a learning control unit that performs machine learning of an estimation model that estimates the state by using the data set with the non-contact sensing data as an input. As a result, it is possible to estimate the state of the sensing target by inputting the non-contact sensing data without requiring the contact sensing data in the end.

The invention according to claim 6 is the contact sensing data from the contact sensor based on the position information indicating the position where the contact sensor is buried in the sensing system according to any one of claims 1 to 5 . It is characterized by further including a flight control unit that controls the flight of the aircraft so as to be able to receive. As a result, the aircraft can efficiently acquire contact sensing data.

The invention according to claim 7 is the contact sensing data from the contact sensor based on the position information indicating the position where the contact sensor is buried in the sensing system according to any one of claims 1 to 6 . It is characterized by further including a flight route creating unit that creates a flight route of the aircraft so as to be able to receive. As a result, the aircraft can efficiently acquire contact sensing data.

The invention according to claim 8 is characterized in that, in the sensing system according to any one of claims 2 to 5 , the state of the sensing target includes the length of the plant. As a result, the condition of the plant can be managed more appropriately.

The invention according to claim 9 is characterized in that, in the sensing system according to any one of claims 2 to 5 , the state of the sensing target includes the direction in which the plant grows from the ground. As a result, the condition of the plant can be managed more appropriately.

The invention according to claim 10 is a contact sensor that is preliminarily embedded in the ground of a predetermined area and performs contact sensing of at least one of soil and plants in the area, and can fly unmanned in the air. It is a sensing data acquisition method performed by a sensing system including an aircraft that performs non-contact sensing of the ground surface in the area, and acquires contact sensing data obtained by the contact sensing and received by the aircraft from the aircraft. The step, the step of acquiring the non-contact sensing data obtained by the non-contact sensing from the aircraft , the contact sensing data obtained by the contact sensing, and the non-contact sensing data obtained by the non-contact sensing. It is characterized by including a step for performing positional mapping .

The invention according to claim 11 is preliminarily embedded in the ground of a predetermined area, can communicate with a contact sensor that performs contact sensing of at least one of the soil and the plant in the area, and is unmanned. A flightable aircraft, a non-contact sensor unit for performing non-contact sensing of the ground surface in the area from the air, a receiving unit for receiving contact sensing data obtained by the contact sensing from the contact sensor, and the above . It is characterized by including a processing unit for positionally associating the contact sensing data obtained by the contact sensing with the non-contact sensing data obtained by the non-contact sensing .

Claims (12)

A sensing system that includes a contact sensor that is pre-embedded in the ground of a predetermined area and performs contact sensing of at least one of the soil and plants in the area, and an aircraft that can fly unmanned.
The aircraft
A non-contact sensor unit for performing non-contact sensing of the ground surface in the area from the air,
A receiving unit that receives the contact sensing data obtained by the contact sensing from the contact sensor, and
A sensing system characterized by being equipped with. The sensing according to claim 1, further comprising a processing unit that positionally associates the contact sensing data obtained by the contact sensing with the non-contact sensing data obtained by the non-contact sensing. system. The sensing system according to claim 1 or 2, further comprising an estimation unit that estimates the state of the sensing target based on the contact sensing data and the non-contact sensing data associated with the processing unit. .. The processing unit performs positional correspondence between the contact sensing data and the non-contact sensing data a plurality of times in chronological order.
The estimation unit identifies the transition of the state based on the contact sensing data and the non-contact sensing data for a plurality of times associated with the processing unit.
The sensing system according to claim 3, further comprising a diagnostic unit for diagnosing the health of the plant based on the transition of the state specified by the estimation unit. The sensing system according to claim 3 or 4, further comprising a generation unit that generates treatment information indicating a treatment to be performed on the sensing target based on the state estimated by the estimation unit. Estimating the state by using the non-contact sensing data as an input using the data set of the state estimated based on the contact sensing data and the non-contact sensing data associated with the contact sensing data. The sensing system according to any one of claims 3 to 5, further comprising a learning control unit that performs machine learning of a model. Claim 1 is further comprising a flight control unit that controls the flight of the aircraft so that the contact sensing data can be received from the contact sensor based on the position information indicating the position where the contact sensor is buried. The sensing system according to any one of 6 to 6. A claim comprising further a flight route creation unit that creates a flight route for the aircraft so that the contact sensing data can be received from the contact sensor based on the position information indicating the position where the contact sensor is buried. Item 6. The sensing system according to any one of Items 1 to 7. The sensing system according to any one of claims 1 to 8, wherein the state of the plant includes the length of the plant. The sensing system according to any one of claims 1 to 9, wherein the state of the plant includes a direction in which the plant grows from the ground. A contact sensor that is pre-embedded in the ground of a predetermined area and performs contact sensing of at least one of the soil and plants in the area, and non-contact sensing of the ground surface in the area from the air that can fly unmanned. It is a sensing data acquisition method performed by a sensing system including an aircraft that performs
A step of acquiring contact sensing data obtained by the contact sensing and received by the aircraft from the aircraft, a step of acquiring non-contact sensing data obtained by the non-contact sensing from the aircraft, and a step of acquiring the non-contact sensing data obtained by the non-contact sensing.
A sensing data acquisition method characterized by including. An aircraft that is pre-embedded in the ground of a predetermined area, can communicate with a contact sensor that performs contact sensing of at least one of the soil and plants in the area, and can fly unmanned.
A non-contact sensor unit for performing non-contact sensing of the ground surface in the area from the air,
A receiving unit that receives the contact sensing data obtained by the contact sensing from the contact sensor, and
An aircraft characterized by being equipped with.

Images