JP6925638B2 - Cattle health management system and management method and health management program - Google Patents

Description

The present invention relates to a cattle health management system and management method as well as a health management program.

In the livestock industry, it is important to understand the health status of livestock animals. However, in recent years, the number of people engaged in the livestock industry has decreased, and while the scale of dairy farming has increased, the problem of labor shortage has arisen. Due to the labor shortage, the physical condition and health status of the cattle raised may be neglected, which may affect the breeding of cattle and the management of diseases, which in turn may lead to a decrease in the productivity of milk and beef.
Therefore, in order to improve such a situation, a system for managing livestock animals such as cattle using various sensors and computers has been proposed (for example, Patent Documents 1 to 4).

Patent Document 1 includes a system in which a tag equipped with a temperature sensor and a telemeter is used for a cow's ear tag to enable individual identification of a cow, and a system for collectively managing the detected body temperature of the cow using a computer. It is disclosed.

Patent Document 2 describes a sensor that is attached to a controlled animal to measure a predetermined physical quantity such as acceleration and wirelessly transmits the measurement data, and a predetermined frequency based on the measurement data by the sensor. An animal health management system including a determination computer that makes a predetermined determination and transmits the determination result to a predetermined reporting destination by a predetermined method is disclosed.

Patent Document 3 describes an acceleration sensor and / or a ruminant that is attached to a ruminant collar and measures the acceleration in the anteroposterior direction when viewed from the ruminant as a method and system for managing the health condition of the ruminant such as a cow. A technology that has temperature sensors that measure the temperature of the lower jaw and body, and grasps the health condition, estrus, and delivery timing of the ruminant 11 based on the acceleration data and / or temperature data measured by these sensors. It is disclosed.

In Patent Document 4, a pressure sensor and an acceleration sensor are attached to a collar attached to the neck of a ruminant in order to more accurately grasp whether the ruminant is resting in a lying position or a standing position. A health management system including a data analysis processing device including a ruminant state determination unit that is attached and determines the health state of a ruminant animal based on the respective measurement data acquired by a pressure sensor and an acceleration sensor is disclosed.

However, such a conventional system has a problem in the health management of cattle that it is not possible to detect, treat and manage diseases having different progression rates in cattle at an early stage.

Japanese Unexamined Patent Publication No. 2003-333950 JP-A-2007-306804 Japanese Unexamined Patent Publication No. 2008-228573 JP-A-2017-60407

The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to discriminate various activity states of cattle being bred and calculate from the amount of activity for each discriminated activity. The purpose is to provide a cattle health management system that can detect diseases with different progression rates based on changes in indicators.

The cow health condition management system according to one aspect of the present invention, which has been made to achieve the above object, connects an activity status sensor module mounted on a cow in a controlled area to the activity status sensor module via a communication network. A cow health condition management system including the above-mentioned management device, wherein the activity state sensor module includes a 3-axis acceleration sensor and a pressure sensor, and the cow is identified by the measured values measured by the respective sensors. Control that transmits the activity data of the cow associated with the data to the management device, and the management device controls the activity state determination unit, the health condition determination unit, the communication unit, the storage unit, the display unit, and the entire system. The activity state determination unit specifies the activity state of the cow by using a preset behavior identification model based on the activity data of the cow received from the activity state sensor module, and the identification unit includes the unit. The cow's behavioral index, stationary index, and reflexive index corresponding to the activity state of the cow are calculated, and the health condition determination unit calculates the calculated behavioral index, stationary index, and And the temporal fluctuation amount of the anticorrosion system index is compared with a preset threshold value, and when the fluctuation amount of any of the behavioral system index, the stationary system index, and the anticorrosion system index exceeds the threshold value, it is determined. The determination model determines the presence or absence of abnormalities in the health condition of the cow, and the control unit acquires the activity data of the cow from the activity state sensor module at a predetermined cycle and operates the activity state determination unit. , The index data including the behavioral index, the stationary index, and the reflexive index of the cow calculated based on the acquired activity data of the cow is stored in the storage unit in the order of the acquired time, and every predetermined time. The health condition determination unit is operated, and index data for a predetermined period including immediately before is read from the index data stored in the storage unit in chronological order and transmitted to the health condition determination unit, and the health condition determination unit operates. Control to generate a command to display a warning when it is determined that there is an abnormality in the health condition of the cow based on the amount of fluctuation of the index data immediately before the index data for the predetermined period. It is a feature.

The method for managing the health condition of cattle according to one aspect of the present invention, which has been made to achieve the above object, connects the activity status sensor module mounted on the cattle in the controlled area and the activity status sensor module via a communication network. A method of managing the health condition of a cow in a cow health condition management system including the above-mentioned management device, wherein the management device acquires cow activity data from the activity state sensor module at a predetermined cycle. Based on the step and the acquired activity data of the cow, the step of calculating the behavioral index, the stationary index, and the reflexive index of the cow, and the calculated behavioral index and the stationary index of the cow. , And the step of associating the index data including the anticorrosion system index with the identification data of the cow and storing it in the storage unit of the management device in the order of acquisition time, and the step of storing the index data stored in the storage unit at predetermined time intervals. A step of reading the index data of a predetermined period including the immediately preceding index data from the index data and comparing the fluctuation amount of the index data immediately before with respect to the index data of the predetermined period with a preset threshold value, and the compared index data. When the fluctuation amount of any of the behavioral index, the stationary index, and the reflexive index included in the above exceeds the threshold value, a step of determining the presence or absence of an abnormality in the health condition of the cow by a predetermined determination model, and It is characterized by having a step of generating a command for displaying a warning when it is determined that the health condition of the cow is abnormal.

The cow health condition management program according to one aspect of the present invention made to achieve the above object connects an activity state sensor module mounted on a cow in a controlled area and the activity state sensor module via a communication network. In a cow health condition management system including the above-mentioned management device, the management device including an activity state determination unit, a health condition determination unit, a communication unit, a storage unit, a display unit, and a control unit that controls the entire system. A program that causes a computer to execute a process for managing the health condition of a cow, which is a step of causing the computer to acquire cow activity data from the activity state sensor module at a predetermined cycle, and a step of acquiring the acquired cow activity data. Index data including the step of calculating the behavioral index, the stationary index, and the anticorrosion index of the cow based on the activity data, and the calculated behavioral index, the stationary index, and the anticorrosion index of the cow. Is associated with the cow identification data and stored in the storage unit in the order of acquired time, and an index for a predetermined period including immediately before from the index data stored in the storage unit at predetermined time intervals. A step of reading data and comparing the amount of fluctuation of the index data immediately before with respect to the index data of the predetermined period with a preset threshold, a behavioral index, a stationary index, and a stationary index included in the compared index data. When the fluctuation amount of any of the anticorrosion system indexes exceeds the threshold value, it is determined that there is an abnormality in the health condition of the cow and a step of determining the presence or absence of an abnormality in the health condition of the cow by a predetermined determination model. It is characterized by executing a step of generating a command for displaying a warning.

According to the present invention, diseases having different progression rates in cattle, that is, acute diseases and chronic diseases can be detected at an early stage. In addition, it prevents oversight of changes in the physical condition of cattle due to labor saving in monitoring the active state, and has the effect of facilitating and improving the efficiency of health management.

It is a schematic diagram which shows the structure of the health condition management system of cattle by one Embodiment of this invention. It is a block diagram which shows the structure of the activity state sensor module by one Embodiment of this invention. It is a block diagram of the management apparatus by one Embodiment of this invention. It is a flowchart which shows the activity state determination process of the management apparatus by one Embodiment of this invention. It is a flowchart which shows the method of determining the health condition of a cow by the health condition determination part of the management apparatus by one Embodiment of this invention. It is a figure which shows an example of the time transition of the behavioral index calculated by the management apparatus by this embodiment. It is a flowchart which shows the other method of determining the health condition of a cow by the health condition determination part of the management apparatus by one Embodiment of this invention. It is a figure which shows another example of the time transition of the behavioral index calculated by the management apparatus by this embodiment.

Hereinafter, specific examples of embodiments for carrying out the present invention will be described in detail with reference to the drawings. In the embodiment of the present invention, a cow will be described as an example, but the present invention is not limited thereto.

<Overall configuration>
First, the overall configuration of the cattle health condition management system according to the present invention will be described.

FIG. 1 is a schematic view showing a configuration of a cattle health condition management system according to an embodiment of the present invention. As shown in FIG. 1, the health condition management system 1 is derived from an activity status sensor module 20 mounted on one or more cows A to E in a controlled area 10 such as a barn or a ranch, and an activity status sensor module 20. It includes a management device 30 that processes information and manages the activity state and health state of cattle. The details of the activity state sensor module 20 will be described later.

The management device 30 is a general-purpose computer or server computer equipped with a CPU, a main storage device, an auxiliary storage device, a network module, and the like, and has each functional unit realized in cooperation with installed programs and software. .. Details of each functional unit of the management device 30 will be described later (see FIG. 3).

Each functional unit included in the management device 30 may be realized by a control unit (see FIG. 2) including a CPU in the activity state sensor module 20. That is, the management device 30 acquires data on the activity state of the cow determined by the activity state sensor module 20, and is on the cloud via the storage device in the management device 30 and / or the network 40 (LAN, Internet, etc.). It may be stored in the database 60.

The health state management system 1 may include a repeater 17 that wirelessly communicates with the activity state sensor module 20 and communicates with the management device 30 by wire or wirelessly. In this case, one or a plurality of repeaters 17 are installed in (or outside) the controlled area 10 to cover the entire controlled area 10.

Wireless communication between the repeater 17 and the activity state sensor module 20 is based on standards such as wireless LAN such as Wi-Fi (registered trademark), PHS (registered trademark), Bluetooth (registered trademark), or ZigBee (registered trademark). It may be based on. Further, between the repeater 17 and the management device 30, wireless communication based on these standards or wired communication based on standards such as serial communication and Ethernet (registered trademark) is used. The management device 30 and the activity state sensor module 20 may directly communicate with each other wirelessly without providing the repeater 17.

The health condition management system 1 measures various environmental conditions (atmospheric pressure, temperature, humidity, wind direction and wind power, power consumption, water consumption, etc.) in the control area 10 and transmits the information to the management device 30. , A plurality of sensors 11-16 may be included. As a sensor for measuring various environmental conditions in the controlled area 10, one or more of a reference pressure sensor 11, a temperature sensor 12, a humidity sensor 13, a wind speed sensor 14, a power sensor 15, and a water amount sensor 16 may be used. can. These sensors 11 to 16 communicate with the management device 30 directly or via the repeater 17 wirelessly or by wire. Further, the health condition management system 1 may include a position information sensor (not shown) for detecting the position of the cow in the controlled area 10, that is, the location. As a method for detecting the location of a cow using a position information sensor, a conventionally known position detection method (for example, Japanese Patent No. 6409028) can be used.

Here, the reference atmospheric pressure sensor 11 measures the reference atmospheric pressure in the controlled area 10, and is installed at a predetermined position (for example, on the ground in the barn) of the predetermined area 10A in the controlled area 10. Will be done. The reference atmospheric pressure sensor 11 transmits the atmospheric pressure value measured at predetermined time intervals (for example, several tens of milliseconds to several seconds) to the management device 30 as reference atmospheric pressure data. The reference atmospheric pressure data transmitted to the management device 30 is stored (stored) in the database 313 in the management device 30 and / or the database 60 on the cloud.

The number and position of each of the reference pressure sensor 11, the temperature sensor 12, the humidity sensor 13, the wind speed sensor 14, the power sensor 15, and the water amount sensor 16 depend on the size of the control area 10, the number of areas in the control area, and the like. Is adjusted.

Further, in the health condition management system 1, a user terminal (50, 51) such as a personal computer PC, a smartphone, or a tablet terminal can communicate with a management device 30 via a network 40 such as the Internet, LAN, or WAN. Will be done. As a result, the management device 30 transmits an alarm, that is, a warning, according to the health condition of the cow, which will be described later, to the user terminal (50, 51), and the user terminal (50, 51) accesses the management device 30. Information on the activity and health status of cattle and various environmental conditions can be obtained. At this time, the user terminals (50, 51) access the management device 30 through a web browser or a predetermined application, and acquire information from the management device 30 and / or the database 60 on the cloud.

<Activity sensor module>
Next, the activity state sensor module according to the present invention will be described.

FIG. 2 is a block diagram showing a configuration of an activity state sensor module according to an embodiment of the present invention. The activity state sensor module 20 is a sensor module including a sensor for measuring at least the acceleration of a cow and the atmospheric pressure around it. The activity state sensor module 20 is an active sensor module equipped with a battery and an antenna (not shown) and performs wireless communication by itself, and does not require, for example, reading work of measurement data by a reading device.

As shown in FIG. 2, the activity state sensor module 20 includes a 3-axis acceleration sensor 21, a barometric pressure sensor 22, a control unit 24, a storage unit 25, a communication unit 26, and a timer 27. Further, the activity state sensor module 20 may further include an angular velocity sensor (not shown) such as a temperature sensor 23, a gyroscope or an optical fiber gyro.

The activity state sensor module 20 is mounted on the side or lower side of the cow's neck using a collar or the like so that the built-in 3-axis acceleration sensor 21 and the barometric pressure sensor 22 can measure the vertical movement of the cow and the acceleration of the entire neck. NS. In order to accurately measure changes in acceleration and changes in the height position of the cow's head (neck) due to the activity state of the cow, it is preferable to attach the activity state sensor module 20 near the cow's throat. good. The method of attaching the activity state sensor module 20 is not limited to the collar, and it can be attached to the head using a bridle or embedded in a cow.

The activity state sensor module 20 manages cow activity data (including various data described later) detected by the 3-axis acceleration sensor 21 and the barometric pressure sensor 22 every predetermined period or in response to a request from the management device 30. Send to 30. At this time, the activity state sensor module 20 uses the identifier of the activity state sensor module 20 itself and / or the identifier of the cow to which the activity state sensor module 20 is attached as the data for identifying the cow, and the data of the measurement date and time obtained from the timer 27. It is added to the measured value data of these sensors and transmitted.

The measurement date and time data may be added by the repeater 17, or when the management device 30 receives the cattle activity data transmitted from the activity state sensor module 20, the management device 30 side determines the acquisition date and time. It may be added as data. In this case, the activity state sensor module 20 may be configured not to include the timer 27.

By configuring the activity data of the cattle as described above, the management device 30 can identify and manage the cattle, the activity state of the cattle, and the measurement date and time (or acquisition date and time) of the activity state. In addition, on the management device 30 side, a correspondence table in which the identifier of the activity state sensor module 20 and the identifier of the cow to which it is attached is associated in advance is created and stored in the storage unit 25, and the activity state sensor module 20 is stored. When the management device 30 transmits its own identifier to the management device 30, the management device 30 examines the cow's identifier based on the correspondence table and associates it with the cow's identifier to manage various data including the measured value data. You may.

The 3-axis acceleration sensor 21 measures the acceleration in the front-back (X-axis), left-right (Y-axis), and up-down (Z-axis) directions of the cow at predetermined time intervals (for example, several tens of milliseconds to several seconds). The measured value is output to the control unit 24.

The barometric pressure sensor 22 is a high-precision and high-resolution barometric pressure sensor capable of detecting a barometric pressure change in units of several centimeters at predetermined time intervals (for example, several tens of milliseconds to several seconds), and is the height of the activity state sensor module 20. The atmospheric pressure at the position is measured, and the measured value of the atmospheric pressure is output to the control unit 24. Since there is a difference of about several tens of cm in the height of the activity state sensor module 20 between the lying posture and the standing posture of the cow, there is a difference in the measured value of the atmospheric pressure. The lying posture and the standing posture can be discriminated from the difference in the measured values of the atmospheric pressure.

Since there is a change in atmospheric pressure due to changes in the weather on a daily basis, it is desirable to determine the posture of the cow by the relative pressure that offsets the change in atmospheric pressure due to the change in the weather in the measured value of the atmospheric pressure corresponding to the posture of the cow. The "relative atmospheric pressure" in the present embodiment means an atmospheric pressure converted so that the posture of the cow can be determined without being influenced by the atmospheric pressure change due to climate change or the like. The relative atmospheric pressure will be described later.

The temperature sensor 23 measures the temperature in the vicinity of the temperature sensor 23 at predetermined time intervals (for example, several seconds), and outputs the measured value to the control unit 24.

The control unit 24 includes a CPU, an I / O, an A / D conversion circuit, and the like, and converts each signal detected by the 3-axis acceleration sensor 21 and the pressure sensor 22, that is, the measured value into digital data to measure the measured value data. A set of data in which the activity state sensor module 20 identifier and / or the cow identifier is associated with the measured value data as data for identifying the cow, and the data of the measurement date and time (or acquisition date and time) is associated with the data. A process of generating cow activity data consisting of a tuple) and outputting it to a communication unit 26 is performed.

The storage unit 25 includes a non-volatile memory and the like, and is required for information on the identifier of the activity state sensor module 20 and / or the identifier of the cow to which the storage unit 25 is mounted, a correspondence table thereof, and each process executed by the control unit 24. Each program and software is saved. In addition, the measured value data of various sensors acquired by the activity state sensor module 20 is temporarily stored.

The communication unit 26 has a data transmission / reception function, and receives cow activity data generated by the control unit 24 and each built-in in the activity state sensor module 20 at predetermined periods or in response to a request from the management device 30. The measurement data from the sensors 21 to 23 are transmitted to the management device 30 via the repeater 17 by wireless communication.

The activity state sensor module 20 may further include a biosensor (not shown) that measures a cow's body temperature, heartbeat, myoelectric signal, or the like. Information (measured value data) from these biosensors is also transmitted to the management device 30, and the management device 30 can use this information to support the determination of the activity state of the cow and the physical condition management of the cow.

<Management device>
Hereinafter, in the cattle health state management system according to the present invention, a management device which is a device for determining the activity state and health state of cattle and managing the information will be described.

FIG. 3 is a block diagram of a management device according to an embodiment of the present invention. As shown in FIG. 3, the management device 30 includes a control unit 300 composed of a CPU, a main storage device, an auxiliary storage device, and the like to control the entire system, and a program or software installed in the auxiliary storage device of the control unit 300. As a functional unit realized by executing the above on the CPU, the posture determination unit 301, the feeding determination unit 302, the drinking water determination unit 303, the walking determination unit 304, the running determination unit 305, the stationary determination unit 306, the rebellion determination unit 308, It includes at least one or all of the environmental condition determination unit 309 and the health condition determination unit 310. The estrus determination unit may be further included as the functional unit.

Here, the foraging determination unit 302, the drinking water determination unit 303, the walking determination unit 304, the running determination unit 305, the stationary determination unit 306, and the rumination determination unit 308 are collectively referred to as an activity state determination unit. The functions of the functional units (301 to 10) described below include another device (not shown) that can access the database 313 of the management device 30 and / or the database 60 on the cloud, and a user terminal (50, It may be realized by 51).

The management device 30 includes a communication unit 311 having a function of communicating with a plurality of activity state sensor modules 20 and various sensors 11 to 16 via a network 40, a determination result by each function unit (301 to 10), and each activity state. It further includes a storage unit 312 that stores information and the like acquired from the sensor module 20 and various sensors 11 to 16, and a display unit 314 having a function of displaying various information on a screen such as a liquid crystal display according to a predetermined format.

In the storage unit 312, a database 313 that stores the determination results (types of active states) by each functional unit (301 to 10) and the like is constructed. The database 313 is managed as a table having a field of the identifier of the cow and / or the identifier of the activity state sensor module 20, a field of the measurement date and time, a field of the activity state, and a field of the environmental state. However, the configuration of the database 313 is not limited to this.

The posture determination unit 301 changes the specific acceleration (for example, when the cow is standing or lying down) measured by the 3-axis acceleration sensor 21 of the activity state sensor module 20 at a predetermined cycle (for example, several tens of milliseconds to several seconds). Based on the information of the height position based on the pressure measured by the pressure sensor 22 of the activity state sensor module 20) and the information of the height position based on the pressure, it is determined whether or not the cow is in the standing state or the lying state. .. In addition, the method of determining whether the cow is in the standing state or the lying state from the measured value of the atmospheric pressure is based on the "relative atmospheric pressure" which is the difference between the respective atmospheric pressures measured by the reference atmospheric pressure sensor 11 and the atmospheric pressure sensor 22. (For example, Patent Document 4) may be used.

When the posture determination unit 301 determines that the cow is in a lying state, it is unlikely that the cow is in a feeding activity, a drinking activity, a walking activity, a running activity, and a stationary state. The determination unit 302, the drinking water determination unit 303, the walking determination unit 304, the running determination unit 305, and the rest determination unit 306 do not determine these activities, but the rumination determination unit 308 determines whether or not the cow is performing rumination activity while lying down. Is determined. On the other hand, when the posture determination unit 301 determines that the cow is in an upright state, which of the activities of feeding activity, drinking activity, walking activity, running activity, resting state, and rumination activity is performed. Is determined by the corresponding functional units (302 to 308). By determining the posture (standing or lying down) of the cow by the posture determination unit 301 in this way, it is possible to omit a part of the subsequent determination of the activity state of the cow, thereby reducing the amount of data and the length of the battery. It leads to a long life.

Following the posture determination of the cow by the posture determination unit 301, a predetermined period ( Which active state the cow is in (including the determination of the lying state by the posture determination unit 301) every time period during which an amount of data that can determine which active state the cow is in is obtained, for example, several seconds to several minutes). Is determined. Then, the activity amount (the amount of time that the cow lays down, eats, drinks water, walks, runs, stands still, or recoils), the identifier of the cow and / or the identifier of the activity state sensor module 20, and the information of the measurement date and time. Is associated and stored in the database 313 (or the database 60 on the cloud) in the storage unit 312 of the management device 30. For these determinations, a behavior-specific model in which the relationship between the measured value of the sensor and the behavior of the cow is defined is used. The details of the behavior specific model will be described later.

Here, the control unit 300 classifies each activity amount when the cow is determined to be in the lying state by the posture determination unit 301 into the activity amount when the cow is lying down, and when it is determined that the cow is in the standing state. Each activity amount of is classified into the activity amount at the time of standing up. In addition, the amount of activity (unit: time) calculated by each of the feeding determination unit 302, the drinking water determination unit 303, the walking determination unit 304, the running determination unit 305, the stationary determination unit 306, and the rumination determination unit 308 is taken. It is called the amount of food activity, the amount of drinking water activity, the amount of walking activity, the amount of running activity, the amount of stationary activity, and the amount of rumination activity.

For example, a cow (A, B, C, ...) Is in which active state in one minute based on each measured value data generated by the active state sensor module 20 by the management device 30 every minute. It is determined whether or not the cow is warm, and as a result of the determination, the state of the cow is stored as "state judgment data" in the database 313 and / or the database 60 on the cloud in association with the identification information and the time information of the cow. As an example, the data on the activity status of each cow shown in Table 1 below for each minute is stored in the database 313 in chronological order.

Next, the operation of each functional unit will be described in detail.

The foraging determination unit 302 determines whether or not the cow is in the foraging activity mainly based on the measured value (that is, the measured value data) from the 3-axis acceleration sensor 21. In general, the feeding behavior of cattle in the pasture is to wrap the grass in the mouth with the tongue, sandwich it between the incisors of the lower jaw and the denticle plate of the upper jaw, and shake the head back and forth and left and right to tear off the grass. After performing the movement several times, chew and swallow. Even when mixed feed is fed in the barn, the behavior is almost the same. From this, the foraging determination unit 302 shows a fluctuation pattern (hereinafter referred to as a measurement pattern) in which the measured value data of the 3-axis acceleration sensor 21 corresponds to a small movement peculiar to the feeding of cattle, and the pressure sensor. When the height of the head (neck) of the cow obtained from 22 is continuously low for a predetermined time, it is determined that the cow is engaged in foraging activity. This is because cows generally lower their heads to a lower position for feeding, so if the cow's head is continuously low for a predetermined period of time in a controlled area 10 that is a rangeland or barn, the cow Is highly likely to be engaged in foraging activities. Therefore, the foraging determination unit 302 uses the measured value data of the height of the head (neck) of the cow obtained from the barometric pressure sensor 22 for determining the foraging activity. Further, since the movement of the cow's neck differs between the feeding activity and the rumination behavior described later, it can be discriminated by the measurement pattern of the acceleration sensor.

In addition, in order to improve the determination accuracy of the foraging behavior, the fact that the animal is approaching the breeding tank is acquired by using the position information sensor installed in the controlled area 10, and this information is given to the foraging behavior determination unit. You may. In order to determine the activity status of cattle based on the measured value data of each sensor, a behavior identification model including a measurement pattern for determination and the like is previously stored in the database 313 of the management device 30 (or the database 60 on the cloud). It will be saved.

The drinking water determination unit 303 determines whether or not the cow is drinking water, mainly based on the measured value data from the 3-axis acceleration sensor 21. In general, cows lower their heads when drinking water, move their mouths (especially the lower jaw) up, down, left and right, and then drink water in a stationary state. From this, in the drinking water determination unit 303, the measurement value data of the 3-axis acceleration sensor 21 shows the measurement pattern of the movement peculiar to the cow drinking water, and the height of the cow's head (neck) obtained from the pressure sensor 22. When the cow is maintained at the height position of the drinking fountain continuously for a predetermined time, it is determined that the cow is engaged in drinking activity. The drinking water determination unit 303 acquires the measurement pattern of the acceleration sensor used for determining the drinking activity and the information on the height position of the cow's head (neck) from the behavior specific model. However, since the change in acceleration during drinking is minute, it may be judged including the fact that the cow is approaching the aquarium based on the position information of the cow acquired by using the position information sensor in order to improve the judgment accuracy. ..

The walking determination unit 304 determines whether or not the cow is in walking activity, mainly based on the measured value data from the 3-axis acceleration sensor 21. Generally, when walking, a cow sends out one forefoot toward the ground in the moving direction, and when the foot lands, it sends out the other forefoot toward the ground in the moving direction, and this is repeated. Walk and move. From this, the walking determination unit 304 determines that the cow is performing walking activity when the measured value data of the 3-axis acceleration sensor 21 shows a measurement pattern of a movement peculiar to the cow when walking.

The travel determination unit 305 determines whether or not the cow is in a running state (running activity) mainly based on the measured value data from the 3-axis acceleration sensor 21. Generally, the measured value data of the 3-axis acceleration sensor 21 based on the running activity of a cow shows a measurement pattern similar to that of walking activity, but its intensity is relatively large. Therefore, the traveling determination unit 305 determines that the cow is performing the traveling activity when the measured value data of the three-axis acceleration sensor 21 shows a measurement pattern of a movement peculiar to the cow when traveling.

The rest determination unit 306 determines whether or not the cow is in a rest state, mainly based on the measured value data from the 3-axis acceleration sensor 21. For example, it is generally considered that a cow is in a resting state if it does not move while standing still or lying down. In this way, the rest determination unit 306 determines that the cow is in the rest state based on the measured value data from the three-axis acceleration sensor 21 and the barometric pressure sensor. The stationary state is determined by distinguishing between a standing and stationary state and a lying and stationary state.

The rumination determination unit 308 determines whether or not the cow is in rumination activity, mainly based on the measured value data from the 3-axis acceleration sensor 21. In general, rumination of cows is seen in the standing or lying down state of cows, and once eaten grass is bolus from the rumen and returned to the mouth, the lower jaw is moved left and right, and the bolus is chewed with the molar teeth. Heal, re-chew about 40 to 50 times per bolus, and then swallow. Therefore, the measured value data from the 3-axis acceleration sensor 21 shows that after the bolus is returned from the rumen to the mouth, the back of the head moves in small steps about 40 to 50 times, then the movement is temporarily stopped and the bolus is swallowed. It shows a measurement pattern that changes regularly according to the movement of returning the bolus from the rumen to the mouth, moving about 40 to 50 times, then stopping and swallowing. From this, the rumination determination unit 308 determines that the cow is performing rumination activity when the measured value data from the 3-axis acceleration sensor 21 shows a measurement pattern of movement peculiar to the rumination of the cow. The rumination activity is determined by distinguishing between a state of standing up and ruminating and a state of lying down and ruminating.

The behavior-specific model used for determining the activity state described above is the relationship between the maximum value and standard deviation of the L2 norm of the measured value of the acceleration sensor during a predetermined time (for example, 1 to 10 minutes) and the behavior of the cow. This is a model in which a region representing the above is defined (for example, Japanese Patent Application Laid-Open No. 2018-170969). Such a behavior-specific model is created in advance by a machine learning method such as SVM (Support Vector Machine). For learning the behavior specific model, for example, various machine learning methods such as a neural network are used.

The environmental condition determination unit 309 determines the environmental condition (atmospheric pressure, temperature, humidity, wind direction and speed, power consumption, etc.) of the controlled area 10 based on the measured value data from various sensors 11 to 16 provided in the controlled area 10. Water usage, etc.) is determined.

The health condition determination unit 310 provides information on whether the cow determined by the posture determination unit 301 is in a lying state or an upright state (that is, state determination data), the amount of foraging activity calculated by the foraging determination unit 302, and the drinking water determination. The amount of drinking water activity calculated by the unit 303, the amount of walking activity calculated by the walking determination unit 304, the amount of traveling activity calculated by the traveling determination unit 305, the amount of stationary activity calculated by the stationary determination unit 306, and the rumination determination unit. Based on the ruminant activity information calculated by 308, the health status of cattle is determined. The details of the health condition determination by the health condition determination unit 310 will be described later.

The management device 30 receives measurement value data from the activity state sensor module 20 attached to each cow and various sensors 11 to 16 provided in the control area 10 at predetermined time intervals (or when the management device 30 requests). Is temporarily stored in the storage unit 312, and each determination unit (301 to 10) determines the activity state of the cow, the environmental state of the controlled area 10, etc. based on the measured value data, and stores the result. Add to database 313 and / or database 60 on the cloud in part 312.

By constructing a database 313 and / or a database 60 on the cloud regarding the activity status of multiple types of cattle (big data conversion), the activity status of cattle, which is a precursor to cattle diseases, is analyzed, and cattle diseases, It enables highly accurate prediction of the effect on milking amount.

FIG. 4 is a flowchart showing an activity state determination process of the management device according to the embodiment of the present invention. The flow from start to end shown in FIG. 4 is repeated at predetermined time intervals and accumulated in the database 313 and / or the database 60 on the cloud.

As shown in FIG. 4, first, in step S401, the management device 30 changes the activity state of the cow from the activity state sensor module 20 attached to each cow at predetermined time intervals (or when the management device 30 requests). The information is acquired and stored in the storage unit 312.

In step S402, the posture determination unit 301 of the management device 30 determines whether the cow is standing or lying down based on the information on the activity state of the cow acquired in step S401. When the posture determination unit 301 determines that the cow is in a lying state, the rumination determination unit 308 subsequently determines whether or not the cow is performing rumination activity, and if it is not performing rumination activity, the cow's behavior is It is determined that the patient is lying down. On the other hand, when the posture determination unit 301 determines that the cow is standing, the feeding determination unit 302, the drinking water determination unit 303, the walking determination unit 304, the running determination unit 305, the stationary determination unit 306, and the rumination determination unit 308. Determines which type of activity the cow's behavior corresponds to: foraging, drinking water, walking, running, stationary, and rumination.

In step S403, the management device 30 associates the information regarding the type of the activity state of the cow determined in step S402 with the identifier of the cow and / or the identifier of the activity state sensor module 20 and the information on the date and time when the activity state was measured. , As "activity data" of the cow, it is stored (accumulated) in the database 313 of the storage unit 312 and / or the database 60 on the cloud.

In step S404, the management device 30 acquires each measured value data from various sensors 11 to 16 provided in the control area 10 at predetermined time intervals (or when the management device 30 requests), and stores the measurement value data in the storage unit 312. save.

In step S405, the management device 30 processes each measured value data acquired in step S404 by the environmental state determination unit 309, and the environmental state (atmospheric pressure, air temperature, humidity, wind direction and speed, power consumption, power consumption, etc. of the controlled area 10). Water usage, etc.) is determined.

In step S406, the management device 30 stores the information regarding the environmental state determined in step S405 in the database 313 of the storage unit 312 and / or the database 60 on the cloud in association with the date and time information.

In step S407, the management device 30 sends predetermined information from the database 313 and / or the database 60 on the cloud to the user terminal (50, 51) at predetermined intervals or in response to a request from the user terminal (50, 51). Is transmitted, and the user terminals (50, 51) display various information in a predetermined format on the screen. Similarly, the management device 30 can display various types of information on the display unit 314 in a predetermined format according to a predetermined interval or a user's request.

The order of steps S401 to S407 is not limited to this order, and may be changed as appropriate.

<How to judge the health condition of cattle>
Hereinafter, a method of determining the health condition of cattle by the management device according to the embodiment of the present invention will be described with reference to FIGS. 3 and 5.

The health condition determination unit 310 of the management device 30 reads (acquires) the activity amount data in each activity state for each cow from the database 313 (or the database 60 on the cloud), and the behavioral index and the stationary system for each predetermined time t. Aggregate as an index and a ruminant index. Here, the behavioral index is a numerical value obtained by totaling the foraging activity amount, the drinking activity amount, the walking activity amount, and the running activity amount, and the stationary index is a numerical value obtained by totaling the stationary activity amount. The amount of static activity is calculated separately for the amount of standing stationary activity and the amount of lying stationary activity according to the state of standing or lying down. The ruminant index is a numerical value obtained by totaling the amount of ruminant activity, and is divided into the amount of standing ruminant activity and the amount of lying ruminant activity according to the state of standing or lying down.

Further, the health state determination unit 310 calculates a behavioral index, a stationary index, and a rebellion index from the activity data of each active state included in the time period of the predetermined time T (an integral multiple of 2 or more of t). do. Then, the health condition determination unit 310 determines the health condition of the cow based on the calculated index.

FIG. 5 is a flowchart showing a method of determining the health condition of cattle by the health condition determination unit of the management device according to the embodiment of the present invention.

In step S501, the health condition determination unit 310 accesses the database 313 (or the database 60 on the cloud) every predetermined time t (tens of minutes to several hours, for example, 3 hours), and each activity amount for each cow. Read the data.

In step S502, the health condition determination unit 310 aggregates each activity amount data included in the period from the present time to the predetermined time t in the past, and calculates the behavioral index, the stationary index, and the ruminant index, respectively.

For example, if the predetermined time t is 3 hours (the minimum unit is 1 minute), the total amount of activity included in the predetermined time t is 180 minutes. In the example shown in Table 2 below, as the aggregated values of the activity amount for each activity item at the present time, the amount of foraging activity is 30 minutes, the amount of drinking water activity is 10 minutes, the amount of walking activity is 70 minutes, and the amount of lying still activity is 30. If the amount of standing and stationary activity is 10 minutes and the amount of standing and ruminant activity is 30 minutes, the behavioral index is 110 = (30 + 10 + 70), the stationary index is 40 = (30 + 10), and the ruminant index is 30. Become. Further, when the same calculation is performed for the next time (3 hours after the present time), the behavioral index is 40, the stationary index is 120, and the ruminant index is 20. That is, each index is defined as a numerical value corresponding to the sum of each activity amount included in the index during a predetermined time t.

In this way, the health condition determination unit 310 calculates the current behavioral index, the stationary index, and the ruminant index for each cow, and uses these as the cow's identifier (or the identifier of the activity state sensor module 20) and the current time. It is stored in the database 313 (or the database 60 on the cloud) in association with the time. In addition, each aggregated activity data (hereinafter referred to as activity aggregate data) is also stored in the database 313 (or database 60 on the cloud) in association with the cow identifier and the current time. In the present embodiment, each index is calculated as the sum of the corresponding activity amounts, but any index whose purpose is to detect changes in each activity amount at different times or times. An index may be used. For example, the ratio (ratio) of each activity amount to the predetermined time t may be used.

In step S503, the health condition determination unit 310 sets a period from the present time to a predetermined time T1 (an integral multiple of 2 or more of t, for example, 3 hours to 24 hours) as a unit evaluation period, and within this unit evaluation period. In, the average value of the behavioral index, the stationary index, and the ruminant index for each predetermined time t is calculated. As a result, it is possible to obtain an evaluation index in which the influence of the noise component included in each index at predetermined time t is suppressed.

In this way, the health condition determination unit 310 calculates the average value of the behavioral index, the stationary index, and the reflexive index included in the unit evaluation period for each cow, and the cow identifier (or the activity state sensor module 20). It is stored in database 313 (or database 60 on the cloud) in chronological order in association with the identifier) and the current time. In the following description, the average value of each index calculated for each unit evaluation period is referred to as unit evaluation data of each index. Therefore, the unit evaluation data of each index is accumulated in the database in chronological order.

In step S504, the health condition determination unit 310 uses the unit evaluation data of each index in the unit evaluation period from the present time to the predetermined time T1 calculated in step S503, and the predetermined time T2 (2 or more of t) before T1. It is compared with the unit evaluation data of each index calculated in the same manner in the period up to (integer multiple. It may be the same as T1). The comparison method may simply calculate the difference between the current unit evaluation data and the immediately preceding unit evaluation data, but the difference from the fluctuation amount predicted from a plurality of past unit evaluation data by time series analysis. May be calculated and compared.

In step S504, as a method of comparing the unit evaluation data of the current behavioral index, stationary index, and ruminant index, it is conceivable to use the moving average value of the average value of each index from the past to the present. However, in comparison with the long-term moving average value, the difference between the average value of each index at the present time, which has changed significantly due to the change in the health condition of cattle, and the respective moving average value becomes small. In addition, the user may not be able to know the change (deterioration or improvement) of the health condition of the cow in a timely manner (immediately after the change occurs). Furthermore, the user's operation cycle is often daily (24 hours). Considering these, the predetermined time T2 is preferably 6 to 48 hours, preferably 6 to 24 hours, and more preferably 24 hours.

In this embodiment, the average value or the moving average value of each index is used, but any function can be used as long as it is a function whose purpose is to obtain the fluctuation of each activity amount at different times or times. good. For example, a moving average, an autoregressive moving average (AR model), an autoregressive moving average (ARMA model), an autoregressive integrated moving average (ARIMA model), an oblivion learning algorithm (SDAR model), a K-nearest neighbor method, or the like may be used. ..

In step S505, the health condition determination unit 310 determines whether or not an alert has occurred based on the difference (or fluctuation amount) of each index calculated in S504. In the judgment, the difference (or fluctuation amount) between each calculated index and each activity amount aggregated data included in the index sets a preset threshold value for each index and each activity amount aggregated data. If it is exceeded, an alert will be flagged. These threshold values are derived from past achievements and empirical values, but it is also possible to update them by machine learning or the like using each of the past aggregated data.

FIG. 6 is a diagram showing an example of the time transition of the behavioral index calculated by the management device according to the present embodiment. In the example shown in FIG. 6, the unit evaluation data of the behavioral index of a certain cow is plotted in chronological order together with the foraging activity aggregated data and the dynamic activity aggregated data constituting the behavioral index, and is plotted with the time period T1 (current). The difference ΔACT of the behavioral index, the difference Δfed of the foraging activity aggregated data, and the dynamic activity aggregated data Δmove all exceed the respective thresholds (Tha, Thf, Thm) with the time period T2 before that. An example is shown. In this case, the cattle behavioral index, feeding activity and dynamic activity are flagged for alert generation. The dynamic activity amount is the total activity amount of the drinking activity amount, the walking activity amount, and the running activity amount.

The health condition determination unit 310 associates the index and activity amount items flagged with the alert with the cow identifier (or the identifier of the activity status sensor module 20) and the current time as alert information, and associates them with the database 313 (or the current time). Save in the database 60) on the cloud.

In step S506, the health condition determination unit 310 determines whether or not there is an abnormality in the health condition of the cow based on the determination model from the alert information determined in step S505. That is, it is determined whether or not there is an abnormality in the health condition of the cow based on the determination model from the combination pattern of the index in which the alarm is set and the item of the activity amount. In the example shown in FIG. 6, an alert flag is set for the cattle behavioral index and the foraging activity and dynamic activity included in this index, and the cattle foraging activity and dynamic activity decrease sharply. Is recognized. The health condition determination unit 310 classifies patterns in which alerts are issued for these items at the same time according to the determination model, and determines that the health condition of the cow is abnormal. In addition, the content of the abnormality (that is, illness or poor physical condition) may be determined from the combination pattern of the index in which the alert is set and the item of the amount of activity. Note that FIG. 6 records cases in which the behavioral index and the amount of dynamic activity exceeded the respective thresholds during the past T1'and T2', but at this time, the amount of foraging activity was the threshold. Since it was as follows, it was not determined that the health condition of the cattle was abnormal.

The judgment model is preset based on past cases and recorded data that each index and a specific combination in the amount of activity contained in the index are associated with a specific abnormality (disease or poor physical condition) of cattle. It is a model composed of a plurality of criteria, and a rule-based model and / or a model classified and extracted by machine learning can be used.

When the health condition determination unit 310 determines that there is an abnormality in the health condition of the cow, a command for displaying a warning that the health condition is deteriorating together with the information of the cow corresponding to the user terminal (50, 51). (Command) is generated and transmitted (S507). The warning may include information in a predetermined form (for example, character information, an icon, etc.) from which the content of the abnormality can be confirmed. This allows the user to detect acute illness in cattle at an early stage.

Next, another method for determining the health condition of cattle by the management device according to the embodiment of the present invention will be described.

FIG. 7 is a flowchart showing another method of determining the health condition of cattle by the health condition determination unit of the management device according to the embodiment of the present invention.

In this embodiment, a method for determining a change in the health condition of cattle over a longer period from the present time to the past is shown. This provides a way to lead to the detection of chronic diseases.

The stage in which the health condition determination unit 310 aggregates the activity amount of each activity state included in the period from the present time to the past predetermined time t, calculates the unit evaluation data of each index, and accumulates them in the database in chronological order. Since it is the same as the steps from step S501 to step S503 in FIG. 5, the description thereof will be omitted.

In step S704, the health condition determination unit 310 uses the unit evaluation data of each index in the unit evaluation period from the present time to the predetermined time T1 calculated in the steps up to S503 of FIG. 5, and the predetermined time T3 (predetermined time T3) before T1. The unit evaluation data of each index in the period up to (an integer multiple of 2 or more of t. Longer than T2) is compared with each other. As a comparison method, the difference (or fluctuation amount) between the current unit evaluation data and the unit evaluation data before this may be simply compared, but the time-series analysis shows that the time period is T3. The fluctuation pattern of the unit evaluation data may be extracted and determined from the degree of agreement with the determination model created from the fluctuation pattern of the unit evaluation data corresponding to a plurality of past chronic diseases stored in the database 313 in advance.

In step S704, as a method of comparing the unit evaluation data of the current behavioral index, stationary index, and ruminant index, in order to detect chronic disease in cattle, the moving average value of the average value of each index from the past to the present. When using the above, the predetermined time T3 is preferably 3 days to 60 days, preferably 7 days to 42 days, from the viewpoint of accurately grasping the change in the health condition.

In step S705, the health condition determination unit 310 determines whether or not an alert has occurred based on the difference (or fluctuation amount) of each index calculated in S704. The determination raises an alert when the preset threshold value exceeds the difference (or fluctuation amount) between each calculated index and the activity amount aggregated data included in the index.

The threshold value for determining the health condition of the cattle used in the present embodiment is derived from past achievements and empirical values, but is different from the condition used in step S505, that is, the condition corresponding to long-term fluctuation. It is desirable to be composed of.

FIG. 8 is a diagram showing another example of the time transition of the behavioral index calculated by the management device according to the present embodiment. In FIG. 8, the unit evaluation data of the behavioral index of a certain cow is arranged in chronological order together with the foraging activity aggregated data and the dynamic activity aggregated data constituting the behavioral index, and T1 (current) and T3 before it. Example in which the difference ΔACT of the behavioral index, the difference Δfed of the foraging activity aggregated data, and the dynamic activity aggregated data Δmove all exceed the respective thresholds (Tha', Thf', Thm') in the time period of Is shown.

The health condition determination unit 310 associates the index and activity amount items flagged with the alert with the cow identifier (or the identifier of the activity status sensor module 20) and the current time as alert information, and associates them with the database 313 (or the current time). Save in the database 60) on the cloud.

In step S706, the health condition determination unit 310 determines whether or not there is an abnormality in the health condition of the cow based on the determination model from the alert information determined in step S705. That is, it is determined whether or not there is an abnormality in the health condition of the cow based on the determination model from the combination pattern of the index in which the alarm is set and the item of the activity amount. In the example shown in FIG. 8, the behavioral index of cattle and the foraging activity and dynamic activity included in this index are flagged as alerts, and the foraging activity and dynamic activity of cattle are set for a long period of time. It can be seen that it gradually decreases. The health condition determination unit 310 classifies patterns in which alerts are issued for these items at the same time according to the determination model, and determines that the health condition of the cow is abnormal. In addition, the content of the abnormality (that is, chronic illness or poor physical condition) may be determined by the combination pattern of the index and the activity amount item for which the alert is set. The determination model for determining the health condition of cattle used here is composed of a model different from the one used in step S506, that is, a model corresponding to an abnormality highly related to long-term fluctuation. Is desirable.

When the health condition determination unit 310 determines that there is an abnormality in the health condition of the cow, a command for displaying a warning indicating that the user has a chronic disease together with the information of the cow corresponding to the user terminal (50, 51). (Command) is generated and transmitted (step S707).

Further, as another embodiment of the present invention, the posture determination unit 301, the foraging determination unit 302, the drinking water determination unit 303, and the walking determination unit 304 are based on the measured values of the three-axis acceleration sensor and the pressure sensor mounted on the cow. The present invention relates to a computer program for operating a computer as a functional unit including a running determination unit 305, a stationary determination unit 306, and a rebellion determination unit 308, and a computer program for operating the computer as a health condition determination unit 310. The computer program may be installed in the activity state sensor module 20, the management device 30, or the user terminals (50, 51). Further, a server that realizes each functional unit of the management device 30 may be provided on the cloud, and this server may determine various activity states and health states of cattle.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the present invention is not limited to the above-described embodiments and is variously modified within a range that does not deviate from the technical scope of the present invention. It is possible to carry out.

1 Health condition management system 10 Control area 10A Predetermined area in the control area 11 Reference pressure sensor 12 Temperature sensor 13 Humidity sensor 14 Wind speed sensor 15 Power sensor 16 Water volume sensor 17 Repeater 20 Activity state sensor module 21 3-axis acceleration sensor 22 Pressure Sensor 23 Temperature sensor 24 Control unit 25 Storage unit 26 Communication unit 27 Timer 30 Management device 40 Network 50, 51 User terminal 60 Database on the cloud 300 Control unit 301 Attitude judgment unit 302 Foraging judgment unit 303 Drinking water judgment unit 304 Walking judgment unit 305 Traveling judgment unit 306 Static judgment unit 308 Rebellion judgment unit 309 Environmental condition judgment unit 310 Health condition judgment unit 311 Communication unit 312 Storage unit 313 Database 314 Display unit

Claims (10)

The activity sensor module attached to the cows in the controlled area,
A cattle health condition management system including a management device connected to the activity state sensor module via a communication network.
The activity state sensor module includes a 3-axis acceleration sensor and a pressure sensor, and the measured value of the 3-axis acceleration sensor showing a different measurement pattern depending on the activity state of the cow and the height position where the activity state sensor module is mounted. The activity data of the cow in which the data for identifying the cow is associated with the measured value of the pressure sensor indicating the above is transmitted to the management device.
The management device includes a posture determination unit, an activity state determination unit, a health condition determination unit, a communication unit, a storage unit, a display unit, and a control unit that controls the entire system.
The activity state determination unit includes a feeding determination unit, a drinking water determination unit, a walking determination unit, a running determination unit, a stationary determination unit, and a rumination determination unit.
Based on the measured value of the 3-axis acceleration sensor and the measured value of the pressure sensor included in the activity data of the cow received from the activity state sensor module, the activity state of the cow is set to a preset behavior specific model. Using the posture determination unit and the activity state determination unit, the cow is in any active state of foraging, drinking water, walking, running, standing still, lying still, standing ruminant, and lying ruminant. Identify if there is
The health condition determination unit
For each of the specified activity states, the amount of time that the cow was in the specified activity state during the period from the present time to the past predetermined time t is totaled as the activity amount of the activity state, and the total is calculated. The activity amount is stored in the storage unit as the activity amount total data for each specified activity state, and the activity amount is stored in the storage unit.
Wherein one of the activity of each aggregated active, eating, drinking, walking, and summed action based indicator activity amount of active travel, stationary obtained by summing the activity of active standing still and lying still system metrics, calculates a ruminant system indices obtained by summing the activity of the standing ruminant and lying ruminants in parallel beauty stored in the storage unit,
Behavior-based index of the cattle in the period from the present time to the past predetermined time T1 (2 or more integer times the time t), the stationary system indicator, anti芻系index, and for each of activity aggregated data for each active Each of the calculated average values and the behavioral index, stationary index, anticorrosion index, and activity state of the cow in the period up to the predetermined time T2 (the time obtained by multiplying t by an integer of 2 or more) before the predetermined time T1. the difference between respective average values calculated for each of activity aggregated data each, compared to the respective threshold value set in advance, the comparison and action-based indicators, stationary system indicator, anti芻系indicators, and active among the items of activity aggregated data each, a pattern of a combination of items exceeds the respective threshold, to determine I by the predetermined determination model to determine the presence or absence of an abnormality in health status of the cow ,
The control unit
The activity data of the cow is acquired from the activity state sensor module at a predetermined cycle, the posture determination unit and the activity state determination unit are operated, and the behavior of the cow calculated based on the acquired activity data of the cow. system indicators, still based indicators, anti芻系index, and for each active amount of activity aggregate data, stored in the storage unit on the obtained time order,
Wherein by operating the health status determination section for each of the plant constant-time t, behavior-based index of the cattle in the period from the present time to the past predetermined time (T1 + T2), the stationary system indicators, rumination based index, and for each active by the activity amount aggregate data out read from the storage unit and sent to the health status determination section,
If it is determined that there is an abnormality in the health state before Kiushi to control the health status determination section to generate a command to display a warning,
The behavior-specific model is a model that defines the measurement pattern of the measured value of the three-axis acceleration sensor and the relationship between the measured value of the barometric pressure sensor and the behavior of the cow.
The determination model is a model in which a pattern of a specific combination in the behavioral index, a stationary index, a reflexive index, and the momentum aggregated data for each activity state of the cattle is associated with a specific abnormality of the cattle. A cattle health management system characterized by this. The action-based indicators, foraging activity amount is Hakasan by the feeding determining unit, drinking amount of activity was Hakasan by the drinking water determination unit, locomotor activity amount that is Hakasan by the walking determining part, and by the running determination unit include at least one of Hakasan to the travel amount of activity,
The rest system index includes at least one of the standing still activity amount and the lying still activity amount calculated by the posture determination unit and the rest determination unit.
The ruminant-based index is seen at least Tsuo含of said posture determination unit and standing ruminant activity amount is Hakasan by the ruminant determination unit and lying ruminant activity amount,
The amount of feeding activity, the amount of drinking activity, the amount of walking activity, the amount of running activity, the amount of standing still activity, the amount of lying still activity, the amount of standing rumination activity, and the amount of the lying ruminant activity are collected by cows, respectively. The cow according to claim 1, wherein the amount of time is the amount of time during which the cow is in a state of eating, drinking water, walking, running, standing still, lying still, standing up and ruminating, and lying down and ruminating. Health management system. Behavior-based index of the cow to be read from the storage unit, the stationary system indicators, rumination based index, and the predetermined time between T1 and T2 of the activity aggregated data for each active, the control unit is the health status determination section equal to or longer than the said plant scheduled between t to be operated,
The plant scheduled between t depending upon the health management system of cattle according to claim 1 or 2, characterized in that from 0.5 to 3 hours. The predetermined period T2 of the behavioral index, the stationary index, the ruminant index, and the activity amount aggregated data for each activity state read from the storage unit is from either the first hour period or the second hour period. Selected ,
The first hour period is 6 hours to 48 hours.
The cattle health condition management system according to claim 3, wherein the second hour period is 3 to 60 days. The health status determination section, and characterized in that action-based indicators of past the cattle stored in the storage unit, the stationary system indicator, and by utilizing the ruminant system indicators on the training data to update the respective threshold The cattle health management system according to claim 1. The cow activity state management system further includes a user terminal connected to the management device via a communication network.
The management device transmits a command for displaying a warning corresponding to the abnormality of the health condition of the cow to the user terminal.
The cow health condition management system according to claim 1, wherein the user terminal displays the transmitted warning on a display means included in the user terminal. The activity state sensor module includes an activity state sensor module mounted on a cow in a controlled area and a management device connected to the activity state sensor module via a communication network, and the activity state sensor module includes a 3-axis acceleration sensor and a pressure sensor. Including , And a method of managing the health condition of cattle in a cattle health condition management system including a control unit that controls the entire system.
The management device
From the activity state sensor module at a predetermined cycle, the measured value of the 3-axis acceleration sensor showing a different measurement pattern according to the activity state of the cow and the pressure sensor indicating the height position where the activity state sensor module is mounted. And the step of acquiring the activity data of the cow including the measured value of
Based on the measured value of the 3-axis acceleration sensor and the measured value of the pressure sensor included in the acquired activity data of the cow, the activity state of the cow is determined by using a preset behavior specific model. Judgment by the judgment unit, foraging judgment unit, drinking water judgment unit, walking judgment unit, running judgment unit, stationary judgment unit, rumination judgment unit, the cow eats, drinks, walks, runs, stands still, lies still, A step to identify which of the standing rumination and the lying rumination is active, and
For each of the specified activity states, the amount of time that the cow was in the specified activity state during the period from the present time to the past predetermined time t is totaled as the activity amount of the activity state, and the total is calculated. A step of saving the activity amount in the storage unit as the activity amount aggregated data for each specified activity state, and
Of the total activity amount for each activity state, the behavioral index which is the sum of the activity amounts of the foraging, drinking, walking, and running activity states, and the stillness which is the sum of the activity amounts of the standing stillness and the lying stillness. and storing system metrics, calculates a ruminant system indices obtained by summing the activity of the standing ruminant and lying ruminants in parallel beauty, the acquired time sequentially previous Kiki憶部,
For each of the plant constant-time t, behavior-based index of the cow period from the storage unit or al current to past predetermined period (T1 + T2), the stationary system indicators, rumination based index, and the activity amount aggregated data for each active The activity amount aggregated data for each behavioral index, stationary index, reflexive index, and activity state of the cow within the period from the present time to the predetermined period T1 (time obtained by multiplying t by an integral number of 2 or more). Each of the average values calculated for each of the above, and the behavioral index, the stationary index, and the rebellion system of the cow within the period before the predetermined time T2 (the time obtained by multiplying t by an integral number of 2 or more) than the predetermined time T1. comparing an index, and the difference between the respective average values calculated for each of activity aggregated data for each active, with the respective threshold value set in advance,
Behavior-based index that said comparison, static system indicator, anti芻系index, and among the items of activity aggregated data for each active, the pattern of the combination of items exceeds the respective thresholds, the predetermined determination The step of determining the presence or absence of abnormalities in the health condition of the cow by discriminating with a model,
If it is determined that there is an abnormality in the health of the cow, and generating a command to display a warning, it was closed,
The behavior-specific model is a model that defines the measurement pattern of the measured value of the three-axis acceleration sensor and the relationship between the measured value of the barometric pressure sensor and the behavior of the cow.
The decision model, action-based index of the cow, the stationary system indicators, rumination based index, and the particular combination in momentum aggregated data for each active, Oh model that associates the cow particular abnormal Rukoto A method of managing the health condition of cattle. The method for managing the health condition of a cow according to claim 7, further comprising a step of transmitting a command for displaying the generated warning to a user terminal connected to the management device via a communication network. .. How to manage the health of cattle
A claim characterized by further including a step of updating each of the above thresholds by utilizing the past behavioral index, resting activity amount, and reflexive index stored in the storage unit as learning data. Item 7. The method for managing the health condition of cattle according to Item 7. The activity state sensor module includes an activity state sensor module mounted on a cow in a controlled area and a management device connected to the activity state sensor module via a communication network, and the activity state sensor module includes a 3-axis acceleration sensor and a pressure sensor. Including , And in a cow health management system that includes a control unit that controls the overall system.
The posture determining unit, the feeding determining unit, the drinking determination unit, the walking determining part, the travel determination unit, the stillness determination unit, the ruminant determination unit, the health status determination section, the communication unit, said storage unit, the display unit, and a program for processing the execution for managing the health of cattle in the computer of the management apparatus including the control unit for controlling the system in general,
On the computer
From the activity state sensor module at a predetermined cycle, the measured value of the 3-axis acceleration sensor showing a different measurement pattern according to the activity state of the cow and the pressure sensor indicating the height position where the activity state sensor module is mounted. And the step of acquiring the activity data of the cow including the measured value of
Based on the measured value of the 3-axis acceleration sensor and the measured value of the pressure sensor included in the acquired activity data of the cow, the activity state of the cow is determined by using a preset behavior specific model. Judgment by the judgment unit, foraging judgment unit, drinking water judgment unit, walking judgment unit, running judgment unit, stationary judgment unit, rumination judgment unit, the cow eats, drinks, walks, runs, stands still, lies still, A step to identify which of the standing rumination and the lying rumination is active, and
For each of the specified activity states, the amount of time that the cow was in the specified activity state during the period from the present time to the past predetermined time t is totaled as the activity amount of the activity state, and the total is calculated. A step of saving the activity amount in the storage unit as the activity amount aggregated data for each specified activity state, and
Of the total activity amount for each activity state, the behavioral index which is the sum of the activity amounts of the foraging, drinking, walking, and running activity states, and the stillness which is the sum of the activity amounts of the standing stillness and the lying stillness. system metrics, a step of calculating a ruminant system indices obtained by summing the activity of the standing ruminant and lying ruminants in parallel beauty, is stored in the storage unit on the obtained time order,
For each of the plant constant-time t, behavior-based index of the cow period from the storage unit or al current to past predetermined period (T1 + T2), the stationary system indicators, rumination based index, and the activity amount aggregated data for each active The activity amount aggregated data for each behavioral index, stationary index, reflexive index, and activity state of the cow within the period from the present time to the predetermined period T1 (time obtained by multiplying t by an integral number of 2 or more). Each of the average values calculated for each of the above, and the behavioral index, the stationary index, and the rebellion system of the cow within the period before the predetermined time T2 (the time obtained by multiplying t by an integral number of 2 or more) than the predetermined time T1. comparing an index, and the difference between the respective average values calculated for each of activity aggregated data for each active, with the respective threshold value set in advance,
Behavior-based index that said comparison, static system indicator, anti芻系index, and among the items of activity aggregated data for each active, the pattern of the combination of items exceeds the respective thresholds, the predetermined determination The step of determining the presence or absence of abnormalities in the health condition of the cow by discriminating with a model,
When it is determined that the cow's health condition is abnormal, a step of generating a command for displaying a warning is executed .
The behavior-specific model is a model that defines the measurement pattern of the measured value of the three-axis acceleration sensor and the relationship between the measured value of the barometric pressure sensor and the behavior of the cow.
The decision model, action-based index of the cow, the stationary system indicators, rumination based index, and the particular combination in momentum aggregated data for each active, Oh model that associates the cow particular abnormal Rukoto A cattle health management program featuring.

Images