Animal management systems and animal management methods

Abstract

To provide an animal management system and an animal management method capable of managing the state of an animal easily.SOLUTION: An animal management system 10 includes a sensor 14 attached to an animal 12, a communication device 16 for communicating data acquired by the sensor 14, and a computer 18 for determining the state of the animal 12 from the acquired data. A program is stored in storage means of the computer 18, and the computer 18 functions by the program as mentioned below. The program causes the computer 18 to function as means 28 for converting the value of the sensor 14 to a visible graph, means 30 for converting the visible graph to a degree distribution, means 32 for obtaining the inclination of the degree distribution, and means 34 to determine the state of the animal from the inclination.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] This invention relates to an animal management system and an animal management method for managing the condition of animals. [Background technology]

[0002] Conventionally, systems for managing livestock using acceleration sensors, communication devices, and computers have been developed. For example, Patent Document 1 discloses a system for managing cattle diseases using values ​​from an acceleration sensor. Patent Document 1 discloses a cattle health management system and method that determines the time progression of behavioral indicators of cattle from acceleration sensor values, compares them with previously recorded data, and detects abnormal health conditions in cattle. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2019-122368 [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] The health management system and method described in Patent Document 1 involve collecting and comparing large amounts of historical data and performing complex calculations, resulting in complicated processing. This leads to high costs in raising cattle and high selling prices for livestock. The objective of the present invention is to realize an animal management system and method that can easily manage the condition of animals, and to provide a low-cost method for raising livestock and livestock with low selling prices. [Means for solving the problem]

[0005] The animal management system of the present invention comprises a sensor attached to an animal, a communication device that communicates data obtained by the sensor, means for converting the data obtained by the sensor into a visibility graph, means for converting the visibility graph into a degree distribution, means for determining the form of the degree distribution, and means for determining the state of the animal from the form of the degree distribution.

Effects of the Invention

[0006] According to the present invention, the form of the degree distribution is determined from the data of the sensor attached to the animal, and the state of the animal is determined from that form. Since a large amount of data is not collected nor complex calculations are performed, the state of the animal can be easily managed. This makes it possible to provide a low-cost livestock animal rearing method and livestock animals with a low selling price.

Brief Description of the Drawings

[0007] [Figure 1] It is a diagram showing the configuration of the animal management system of the present invention. [Figure 2] It is a flowchart of the animal management method of the present invention. [Figure 3] It is a graph showing an example of the relationship between acceleration intensity and time. [Figure 4] This is a graph in which two nodes are connected by an edge (straight line) when the two nodes are within line of sight in FIG. 3. [Figure 5] This is the visibility graph constructed from FIG. 4. [Figure 6] It is a graph for explaining the determination of the presence or absence of visibility between nodes. [Figure 7] It is a logarithmic graph of the degree distribution.

Mode for Carrying Out the Invention

[0008] The animal management system and animal management method of the present invention will be described with reference to the drawings.

[0009] The animal management system 10 of the present invention includes a sensor 14 attached to an animal 12, a communication device 16 that communicates data obtained by the sensor 14, and a computer 18 that determines the state of the animal 12 from the obtained data.

[0010] The animal 12 to be managed is an animal that stays within a certain area. For example, cattle raised in a cowshed or a pasture within a certain area can be cited as the animal 12.

[0011] An example of the sensor 14 is an acceleration sensor. The acceleration sensor includes a capacitive three-axis acceleration sensor. The acceleration sensor is attached to the animal 12 to detect the movement of the animal 12. If the animal 12 is a cow, the acceleration sensor is attached to a collar 20 mounted on the cow's neck to detect the movement of the cow. The detection by the sensor 14 may be performed constantly, or may be performed at regular time intervals.

[0012] The communication device 16 includes a device that wirelessly transmits and receives data obtained by the sensor 14, and transmits the data to the computer 18 via the network 22. A transmitter 24 is arranged on the animal 12, and a repeater 26 is arranged in or around the area where the animal 12 is located. For example, the transmitter 24 and the repeater 26 may be provided with a wireless communication device such as Bluetooth (registered trademark) or WiFi, a router, and the like. The transmitter 24 is attached to the animal 12 together with the sensor 14. The sensor 14 and the transmitter 24 attached to the animal 12 are configured to be drivable by a power source such as a battery. The data obtained by the sensor 14 is wirelessly transmitted from the transmitter 24 to the repeater 26. The repeater 26 transmits the data to the computer 18 that determines the state of the animal 12 via the network 22.

[0013] The computer 18 for determining the state of the animal 12 is connected to the network 22. The network 22 may be a communication network such as the Internet, or may be a local area network (LAN) closed within a certain area.

[0014] The values ​​detected by sensor 14 are transmitted to computer 18 via network 22 in chronological order, and computer 18 receives the values ​​from sensor 14 in chronological order via network 22 and processes them into time-series data. Alternatively, the values ​​detected by sensor 14 may be processed into time-series data and then transmitted to computer 18 as a single data set. The values ​​detected by sensor 14 may be transmitted randomly, and computer 18 may arrange them in chronological order to create time-series data. The program described later may include a function to arrange the data from sensor 14 in chronological order.

[0015] As shown in the flowchart of Figure 2, the animal management method of the present invention acquires time-series data from a sensor 14 attached to an animal 12 (S1), sets nodes corresponding to individual times in the time series from the time-series data (S2), calculates whether each node is visible or not (S3), calculates the number of nodes that are visible at each node as the degree of that node (S4), obtains the shape of the distribution of the number of nodes against the degree (S5), and determines the state of the animal from the shape of the node number distribution (S6). To this end, the computer 18 provided in the present invention functions as a means as described by the program. The program causes the computer 18 to function as a means 28 for converting the values ​​of the sensor 14 into a visibility graph, a means 30 for converting the visibility graph into a degree distribution, a means 32 for determining the slope of the degree distribution, and a means 34 for determining the state of the animal from the slope.

[0016] The program may be stored in the storage means of the computer 18 and executed, or the computer 18 may execute a program stored in the storage means of a server connected to the network 22 via the network 22.

[0017] The means 28 for converting to a visibility graph constructs a visibility graph from the data obtained from the sensor 14. In the data in which the values ​​of the sensor 14 are arranged in a time series, the visibility graph is constructed by connecting two nodes with an edge (straight line) when they are visible from each other.

[0018] Specifically, the values obtained by the sensor 14 are arranged in time series. Data containing only acceleration intensity is extracted from the time-series acceleration data. To obtain only the acceleration intensity, the gravitational acceleration is measured in advance, and the gravitational acceleration is subtracted from the acceleration data. The measurement of gravitational acceleration may be a value obtained when the acceleration sensor is kept stationary. The extracted acceleration intensity time-series data is, for example, as shown in Figure 3. The acceleration intensity time-series data is time-series data for a predetermined period of time such as one day. The process of S1 in Figure 2 is performed by the above processing.

[0019] From the extracted time-series data, nodes corresponding to individual times in the time series are set (S2 in Figure 2), and the presence or absence of Visibility between the nodes is determined (S3 in Figure 2). When there is visibility between two nodes of the acceleration intensity time-series data and the nodes are in line of sight, the nodes are connected by an edge (straight line) (Figure 4) to construct a visibility graph (Figure 5).

[0020] A method for determining the presence or absence of node visibility will be described using a simple example. For example, as shown in Figure 6, assume that there are two nodes i and j, and a node k is present therebetween. The time points of the respective nodes i, j, k are t i , t j , t k , and the acceleration intensities at the respective t i , t j , t k are respectively denoted as x(t i ), x(t j ), x(t k ). i, j, and k are natural numbers, and the time points satisfy t i <t k <t j . The time-series data of the respective nodes i, j, k are (t i , x(t i )) , (t j , x(t j )) , and (t k , x(t k )) .

[0021] When the node i and the node j are connected by a straight line L, the slope of the straight line L is (x(t j )-x(ti )) / (t j -t i ) becomes. k The value of the line L in x a Therefore, the following number is obtained.

[0022]

number

[0023] For nodes i and j to be visible, x(t k ) is x a It must be less than the given value, and the formula in equation 2 below must be satisfied.

[0024]

number

[0025] The acceleration intensity x(t) of node k located between two nodes i and j k For nodes i and j to be visible to each other, it is necessary that the above equation 2 is satisfied. This determination is made for all nodes between the two nodes i and j, and if the equation 2 is satisfied for any node in all of those nodes, it is determined that there is visibility between the two nodes.

[0026] For each node, the number of visible nodes is calculated as the degree of that node (S4 in Figure 2), and the number of nodes with that degree is calculated for each degree (S5 in Figure 2). The rate of change in the logarithm of the number of nodes relative to the change in the logarithm of the degree is calculated as a power exponent, and the form of the node number distribution with respect to the degree is obtained (S6 in Figure 2). The form of the node number distribution with respect to the degree is an exponent in the region where the number of nodes can be considered as an exponential function of the degree. The means 30 that converts the visibility graph to a degree distribution using the above method, for example, converts the visibility graph of Figure 5 to a degree distribution as shown in Figure 7. The degree distribution may also be represented by a logarithmic graph. The degree distribution can be represented by P(k), where k is the number of edges (lines) connected to the nodes. If the slope (γ) of the degree distribution plotted on a logarithmic graph is a straight line, then the degree distribution follows the power law P(k)∝k -γ Follow the rules.

[0027] The method 32 for determining the slope of the degree distribution determines the slope (γ) of the degree distribution plotted on a logarithmic graph. This slope (γ) is the slope of the linear regression line in the degree distribution plotted on the logarithmic graph. The method 32 for determining the slope of the degree distribution determines the linear regression line from the logarithmic graph and determines the slope of that linear regression line. For example, in Figure 7, the horizontal axis is 10 1 ~10 2 The slope of the portion of the logarithmic graph plotted as a straight line within the given range matches the slope of the linear regression line.

[0028] The means 34 for determining the state of the animal 12 uses this slope to determine the state of the animal (S7 in Figure 2). If the slope is a predetermined value or within a predetermined range, the animal is determined to be normal; if it is not within the predetermined value or range, it is determined that there is something wrong with the animal. If the animal 12 is a cow, if the slope (γ) is -3.5, the cow is determined to be healthy; if it is not -3.5, it is determined that the cow is unwell, such as having a high body temperature. The predetermined value or range of the slope (γ) may be changed depending on the animal 12.

[0029] A high acceleration data value indicates an animal is actively moving, a low acceleration data value indicates a slow movement, and zero acceleration data value indicates that the animal is at rest. There will be differences in the data obtained from the acceleration sensor between healthy and unhealthy animals, resulting in differences in the slope of the degree distribution in the above steps. Since animals of the same species will move the same or similarly, the slope (γ) will be the same or fall within a predetermined range. When there is a change in physical condition, the slope (γ) will change, and an abnormality can be detected.

[0030] The predetermined value or range of the slope (γ) is determined in advance by obtaining a degree distribution using the above method with healthy animals 12. The determined predetermined value or range of slope (γ) is stored in the storage means of the computer 18 and used by the determination means 34. Since all healthy animals 12 will have the same predetermined value or range of slope (γ), it is not necessary to obtain slope data from many animals 12 in advance.

[0031] Each piece of data, such as the data from the sensor 14, the visibility graph, the degree distribution, and the judgment result, may be stored in the storage means of the computer 18.

[0032] As described above, the animal management system 10 of the present invention simply obtains a degree distribution from the data obtained from the sensor 14 and utilizes its slope. It does not require a large amount of data on the animals 12, nor does it perform complex calculations for determination. The state of the animals 12, especially their health status, can be easily managed.

[0033] When raising calves using the livestock animal rearing method according to the present invention, calves are often raised in large numbers of several to dozens, making it difficult to constantly monitor the health of individual calves. This can sometimes lead to several calves dying in the rearing group due to infectious diseases, but this situation is improved, enabling inexpensive health management in near real-time, and significantly improving the yield of the cattle being raised. As a result, the unit price of the raised cattle is reduced, making it possible to sell the cattle at a lower price.

[0034] The present invention is not limited to the embodiments described above. For example, while the slope of the degree distribution was used, any form of the degree distribution, such as the density of the degree distribution, may be used. The form of the degree distribution, such as the slope and density, changes depending on the state of the animal, and this change is utilized.

[0035] The managed animals 12 are not limited to one; multiple animals 12 may be managed. A sensor 14 is attached to each animal 12, and data is transmitted to a computer 18 using a communication device 16. Data that identifies the animal 12 is added to the data detected by the sensor 14 and transmitted to the computer 18. The computer 18 performs the above data conversion for each animal 12 and makes a determination.

[0036] The sensor 14 is not limited to an acceleration sensor; other sensors 14 may be used, and their values ​​may be converted to a degree distribution in the same way. For example, an optical heart rate sensor or a temperature sensor for measuring body temperature may be used.

[0037] The number of sensors 14 is not limited to one; multiple sensors 14 may be used, and there may be multiple types of sensors 14. For example, an accelerometer, an optical heart rate sensor, and a temperature sensor may be used. Each sensor 14 performs the above-described data conversion to manage the state of the animal 12. Increasing the number and types of sensors 14 makes it easier to determine the state of the animal 12.

[0038] If the area in which the animals 12 are kept expands, multiple repeaters 26 may be provided in different locations. The system is designed so that any one of the multiple repeaters 26 can communicate with the animal 12's transmitter 24.

[0039] The data stored in the storage means may be accessible from other computers connected to the network 22. This allows the status of the animal 12 to be monitored remotely. Furthermore, when it is determined that something is wrong with the animal 12, this information may be transmitted to other computers connected to the network 22.

[0040] The data detected by the sensor 14 was transmitted to the computer 18 via the network 22, but the data may also be transmitted directly from the transmitter 24 to the computer 18. A communication device may be attached to the computer 18 to communicate directly with the transmitter 24. The computer 18 is installed in or near the area where the animals 12 are kept.

[0041] The animals to be managed are not limited to cattle; the present invention can be applied to any animal raised within a certain area, such as pigs, wild boars, sheep, horses, and goats. Furthermore, it is not limited to animals on farms, but may include various animals kept in zoos, or pets such as dogs or cats that live with people in ordinary households. The present invention can be applied to the health management of each of these animals.

[0042] Furthermore, the present invention can be implemented in various forms with improvements, modifications, and changes based on the knowledge of those skilled in the art, without departing from its spirit. [Explanation of Symbols]

[0043] 10: Animal Management Systems 12:Animal 14: Sensor 16: Communication equipment 18: Computer 20: Collar 22: Network 24: Transmitter 26: Repeater 28: Means of converting to a visibility graph 30: Means of converting to a degree distribution 32: Method for determining the slope of a degree distribution 34: Means for determining the condition of an animal

Claims

1. Sensors attached to animals, A means for converting data obtained from the aforementioned sensor into a visibility graph, A means for converting the visibility graph into a degree distribution, A means for determining the form of the aforementioned degree distribution, A means for determining the state of an animal from the morphology of the aforementioned degree distribution, An animal management system equipped with [unspecified features].

2. The animal management system according to claim 1, wherein the form of the degree distribution is the slope of the linear regression line on the logarithmic graph of the degree distribution.

3. The animal management system according to claim 1 or 2, wherein the sensor is an acceleration sensor.

4. The steps include acquiring data about the animal using sensors attached to the animal, The steps include converting the acquired data into a visibility graph, The steps include converting the visibility graph into a degree distribution, The steps include determining the form of the aforementioned degree distribution, The steps include determining the animal's condition from the slope of the aforementioned degree distribution, Animal management methods that include these.

5. The animal management method according to claim 4, wherein the step of determining the form of the degree distribution is the step of determining the slope of the linear regression line on the logarithmic graph of the degree distribution.

6. The aforementioned sensor is an acceleration sensor, The animal management method according to claim 4, wherein the step of acquiring data relating to the animal includes the step of subtracting gravitational acceleration from data obtained by an acceleration sensor.

7. The animal management method according to claim 6, wherein the data obtained from the acceleration sensor is acceleration intensity data.

8. The steps include obtaining time-series data from sensors attached to animals, The steps include setting each node corresponding to an individual time in the time series from the aforementioned time series data, A step of calculating whether or not there is visibility between each of the aforementioned nodes, For each of the aforementioned nodes, the step of calculating the number of visible nodes as the degree of that node, A step of calculating the number of nodes having that order for each order, The steps include obtaining the form of the node number distribution for the aforementioned order, The steps include determining the animal's condition based on the shape of the node number distribution, An animal management method characterized by comprising the following:

9. The animal management method according to claim 8, characterized in that the form of the node number distribution with respect to the degree is an exponent in the region where the number of nodes can be considered as an exponential function of the degree.

10. The step of calculating whether or not each of the aforementioned nodes is visible is characterized in that the time series data when one element of the time t series data is x(ti) is a set of (t = 1, 2, ..., n, n is a natural number), and at time tiii, the node corresponding to x(tiii) is node i, and for two nodes i < j, node i and node j, for any natural number k such that i < k < j, it is determined that there is visibility if x(tk - tk) / (tj - t) < x(tj) - (x(ti) - x(tj))(tj) holds.

11. The aforementioned sensor is an acceleration sensor, The animal management method according to claim 8, characterized in that the time series data is data calculated based on time series gravity-corrected acceleration data obtained by subtracting a known acceleration component due to gravity from time series acceleration data obtained directly from the acceleration sensor.

12. The animal management method according to claim 9, characterized in that the step of determining the condition of the animal is to determine that an abnormality exists when the index exceeds a predetermined range of values.

13. A method for raising animals, characterized by raising them while managing their health condition using the animal management method described in claim 4 or claim 8.

Images