Low-power-consumption animal individual identification, positioning and state sensing system

Abstract

The invention relates to the technical field of radio frequency identification, and discloses a low-power-consumption animal individual identification, positioning and state sensing system, which comprises an information acquisition module, an information transmission module, an information processing module and an intelligent management module, the information acquisition module is used for acquiring animal individual information and environment temperature; the information transmission module is used for efficiently transmitting the collected information and storing the information to a database; the information processing module is used for processing the information acquired by the information acquisition module; and the intelligent management module is used for visually displaying the processed data information. The problem that an ear tag on the market is single in function is solved, and meanwhile, a BLE low-power-consumption Bluetooth and intermittent data acquisition mode is adopted, so that the system power consumption is further reduced; temperature compensation, positioning and exercise amount data are combined with a neural network model to realize sensing and prediction of animal states; the visual platform displays real-time data and animal abnormal state conditions, and effective data support is provided for raising efficiency.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency identification technology, and in particular to a low-power animal individual identification, positioning and state perception system. Background Art

[0002] Radio Frequency Identification (RFID) is one of the key technologies for realizing smart agriculture. RFID technology is applied in animal husbandry, mainly using radio waves for identification and positioning. This technology uses the radio waves emitted by the ear tag to identify and locate the ear tag in real time. At present, my country's animal husbandry is still in a relatively rough breeding mode, with low automation and high labor consumption, making it difficult to achieve large-scale breeding. Most of the existing systems have problems such as single function and low intelligence.

[0003] The present invention designs an animal individual identification, positioning and state perception system, which covers functions such as information collection, information transmission, information processing and storage, and data visualization display. First of all, in view of the problems of short service life, poor fit, and single function of ear tags on the market, a new type of intelligent ear tag is proposed. The information collection device uses BLE low-power Bluetooth and intermittent working mode to further reduce power consumption and increase service life; the ear tag integrates RFID tags, acceleration sensors and thermistors, integrating individual identification, real-time positioning and state perception. Considering the complex environment in the breeding farm and the easy interference of signal transmission, the present invention adopts multi-channel frequency hopping transmission to reduce the impact of complex environment on information transmission. Build a cloud service platform covering information storage, information processing and visualization, combine temperature compensation, positioning, exercise data with neural network models, and predict the individual status of animals; the calculation results are displayed to users by the visualization platform to improve the intelligence of animal breeding. Summary of the invention

[0004] In order to solve the technical problems raised in the background technology, the present invention provides a low-power animal individual identification, positioning and state perception system, which can improve the intelligent management level and efficiency of farms.

[0005] The present invention is implemented by the following technical solutions: an animal individual identification, positioning and state perception system, the system includes an information collection module, an information transmission module, an information processing module, and an intelligent management module;

[0006] The information collection module is used to collect individual information and ambient temperature;

[0007] The information transmission module is used to receive the information collected by the information collection module to the Bluetooth gateway, and then send it to the cloud service platform. The cloud service platform forwards the data to the cloud server and stores it in the MySQL database;

[0008] The information processing module is used to process the information sent by the information transmission module, and use the temperature compensation algorithm, the triangulation positioning algorithm, and the peak detection algorithm to respectively calculate the real-time body temperature, the real-time positioning, and the amount of exercise, and use the neural network model to integrate and calculate the above results to achieve the recognition and prediction of the state of the animal;

[0009] The intelligent management module is used to visualize the processed data information and display possible abnormal situations calculated by the neural network model so that users can manage them in time.

[0010] Optionally, the information collection module includes: a smart ear tag and an ambient temperature sensor;

[0011] Among them, the smart ear tag includes a male tag and a female tag. The male tag contains a passive RFID tag with a unique ID to identify the animal's identity information. The female tag contains an accelerometer to collect the acceleration of the animal during movement, a thermistor to collect the animal's real-time body temperature, and BLE low-power Bluetooth to send the collected information. The ambient temperature sensor is used to collect the ambient temperature in the breeding farm. It also has BLE low-power Bluetooth inside, which uses the same Bluetooth protocol as the smart ear tag to send data to the Bluetooth gateway.

[0012] Optionally, the smart ear tag is made of polyurethane material, so that the ear tag has a high fit and a reduced foreign body sensation, reducing errors caused by loose fit or foreign matter between the ear tag and the animal's skin, and will not cause damage to the animal through chewing, biting, or throwing behaviors; it is made using 3D printing technology to improve the sealing and service life of the ear tag; and the BLE low-power Bluetooth in the mother tag and the ambient temperature sensor only sends data but does not receive data, further reducing the size and power consumption of the system.

[0013] Optionally, the smart ear tag and the temperature sensor in the information collection module use intermittent data collection, that is, the equipment in the information collection module uses a pulsed working current.

[0014] Optionally, the information transmission module includes: a Bluetooth gateway and a cloud service platform, wherein the communication range of the Bluetooth gateway needs to be able to fully cover the scope of the farm, the communication protocols supported by the Bluetooth gateway include Bluetooth protocol and WiFi protocol, and the communication method with the cloud server supports UDP, TCP and MQTT. In order to improve the anti-interference ability of the system, the collection device in the information collection module communicates with the Bluetooth gateway through the Bluetooth protocol and adopts multi-channel frequency hopping transmission.

[0015] Optionally, the Bluetooth gateway is connected to the cloud service platform via WIFI and uses the MQTT protocol for transmission, which enables efficient and reliable IoT device communication even when bandwidth is limited and network is unstable.

[0016] Optionally, a complete system from storage to computing to visualization is built based on the cloud service platform. In the information transmission module, after the cloud service platform receives the data sent by the Bluetooth gateway, it sends it to the cloud server and stores it in the MySQL database.

[0017] Optionally, the information processing module includes: a cloud server, wherein the cloud server includes a temperature compensation algorithm, a triangulation positioning algorithm, a peak detection algorithm and a neural network model.

[0018] Optionally, based on a neural network model, the processed data, i.e., real-time body temperature, real-time location and exercise volume, are integrated and calculated to identify and predict the state of the animal.

[0019] Optionally, the intelligent management module includes a visualization platform, which classifies and visualizes the results of the three algorithm calculations in the cloud server and the results of the neural model recognition and prediction. First, the original data information of the animal, such as real-time positioning, real-time body temperature and exercise volume, is visualized, and then the results of the neural network model recognition and prediction of possible abnormalities are visualized below the original data, wherein the predicted abnormal results include estrus, return of estrus, disease, and injury. The visualization platform provides a memo function, which retains the abnormal results predicted by the neural network model recognition until the user manually deletes them.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. The smart ear tag of the present invention integrates multiple functions of individual identification, positioning and state perception, solving the problem of single function of ear tags on the market.

[0022] 2. The smart ear tag of the present invention adopts BLE low-power Bluetooth and intermittent data collection mode, which further reduces system power consumption and prolongs service life. The transmission between BLE low-power Bluetooth and Bluetooth gateway adopts multi-channel frequency hopping transmission, which reduces the impact of the environment on the quality of information transmission.

[0023] 3. A system covering information storage, information processing and visual display was built, and it was connected with information collection equipment and Bluetooth gateway to form a complete system covering all aspects from collection to transmission to processing and display.

[0024] 4. The present invention improves the original algorithm when calculating the real-time body temperature and positioning of animals. The ambient temperature is introduced when calculating the real-time body temperature to achieve ambient temperature compensation, further reducing the error caused by external temperature changes; the triangulation positioning algorithm is used for positioning, which improves the accuracy compared with the common positioning method.

[0025] 5. The present invention visualizes the results predicted by the neural network model and adds a manual deletion function for abnormal conditions on the basis of the original platform, thereby avoiding the omission of users in handling abnormal conditions of animals. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a schematic diagram of the structure of a low-power animal individual identification, positioning and state perception system proposed by the present invention;

[0027] Figure 2 A low-power working principle flow chart of an information collection device provided by the present invention;

[0028] Figure 3 This is an overall architecture diagram of the MQTT protocol provided by the present invention. DETAILED DESCRIPTION

[0029] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0030] Reference Figure 1-Figure 3 , an animal individual identification, positioning and state perception system proposed in an embodiment of the present invention, the system may include an information collection module U1, an information transmission module U2, an information processing module U3, and an intelligent management module U4;

[0031] The information collection module U1 is used to collect individual information and ambient temperature; the information transmission module U2 is used to receive the information collected by the smart ear tag U11 and the ambient temperature sensor U12 to the Bluetooth gateway U21, and then the Bluetooth gateway U21 sends it to the cloud service platform U22. The cloud service platform U22 forwards the data to the cloud server U31 and stores it in the MySQL database U221; the information processing module U3 is used to process the information sent by the information transmission module U2, and uses the temperature compensation algorithm, triangulation positioning algorithm, and peak detection algorithm to calculate the real-time body temperature, real-time positioning and exercise volume respectively, and uses the neural network model to integrate and calculate the above results to achieve the recognition and prediction of the animal's status; the intelligent management module U4 is used to visualize the processed data information, and display the possible abnormal conditions calculated by the neural network model U314, so that users can manage it in time.

[0032] Specifically, the information collection module U1 includes: a smart ear tag U11 and an ambient temperature sensor U12. Among them, the smart ear tag U11 includes a male tag U111 and a female tag U112. The male tag U111 contains a passive RFID tag, which contains a unique ID mark to identify the identity information of the animal. The female tag U112 contains an acceleration sensor to collect the acceleration of the animal during movement, a thermistor to collect the real-time body temperature of the animal, and a BLE low-power Bluetooth to send the collected information. The ambient temperature sensor U12 is used to collect the ambient temperature in the breeding farm. It also has a BLE low-power Bluetooth inside, and uses the same Bluetooth protocol as the smart ear tag U11 to send data to the Bluetooth gateway U21.

[0033] Specifically, the smart ear tag U11 is made of polyurethane material, which makes the ear tag fit tightly and reduces the foreign body sensation, reducing the error caused by loose fit or foreign matter between the ear tag and the animal's skin, and will not cause damage caused by the animal's chewing, biting, and throwing. It is made using 3D printing technology to improve the sealing and service life of the smart ear tag. In addition, the BLE low-power Bluetooth in the mother tag U112 and the ambient temperature sensor U12 only sends data but does not receive data, further reducing the size and power consumption of the system.

[0034] The smart ear tag U11 and temperature sensor U12 in the information collection module use intermittent data collection, and set the working cycle to 10 seconds, that is, the device collects real-time data every 10 seconds; intermittent data transmission is adopted, and the transmission time slot is 60 seconds, that is, the data is sent once every 6 data collections, that is, the device in the information collection module U1 adopts pulsed working current. The low-power working principle of the smart ear tag U11 and the ambient temperature sensor U12 is as follows Figure 2 As shown, including:

[0035] At the beginning, test whether the smart ear tag U11 and the ambient temperature sensor U12 are working normally; if they are working normally (Y), determine whether the collection time has arrived; if they are not working normally (N), return to the starting step;

[0036] If the acquisition time (Y) is reached, the acquisition working current is used for acquisition until the data acquisition is completed; if the acquisition time (N) is not reached, the dormant working current is used and the data acquisition is directly assumed to be completed;

[0037] Determine whether 6 data are collected. If 6 data are collected (Y), use a larger current to transmit information; if 6 data are not collected (N), use a smaller current to keep the device working until the end. Among them, the collection working current is 0.76mA, and when entering the sleep state, the average current is 7.7μA.

[0038] The information transmission module U2 includes: a Bluetooth gateway U21 and a cloud service platform U22, wherein three Bluetooth gateways U21 need to be configured in actual use to achieve full coverage of the farm and triangulation positioning. The communication protocols supported by the Bluetooth gateway U21 include the Bluetooth protocol and the WiFi protocol, and the communication method with the cloud service platform U22 supports UDP, TCP and MQTT. In order to improve the anti-interference ability of the system, the acquisition devices U11 and U12 in the information acquisition module U1 communicate with the Bluetooth gateway U21 through the Bluetooth protocol and adopt multi-channel frequency hopping transmission. There are 40 channels in the working frequency of the BLE low-power Bluetooth in U11 and U12, and three frequency discontinuous channels are used as broadcast channels. When transmitting data, the channel with the best channel quality is selected for transmission. This method uses multiple frequency channels and can transmit more ear tag data at the same time, which not only reduces the transmission time, but also avoids mutual interference of signals when three discontinuous channels are used at the same time, thereby improving the communication quality.

[0039] The Bluetooth gateway U21 is connected to the cloud service platform U22 via WIFI and uses the MQTT protocol for transmission. MQTT (Message Queuing Telemetry Transport) is a lightweight IoT message transmission protocol based on the publish / subscribe model. Its advantage is that it can still achieve efficient and reliable IoT device communication under limited bandwidth and unstable network conditions. Its architecture diagram is as follows: Figure 3 As shown, specifically, the publisher is the client responsible for sending messages, which is the Bluetooth gateway U21 in the present invention, and publishes the collected individual information data to a specific topic. The subscriber is the client that receives the message, which is the cloud service platform U22 in this invention. The subscriber can subscribe to different topics according to their needs, and the topic is the classification label of the message, which can realize the filtering of the message and facilitate the organization and management of a large number of IoT device messages.

[0040] The cloud service platform U22 is a comprehensive cloud computing service system that provides many types of services, including database services, computing resource services, artificial intelligence development and data visualization services. The present invention builds a complete system from storage to computing to visualization, including MySQL database U221, cloud server U31 and visualization platform U41, to achieve data storage, processing and visualization. In the information transmission module U2, after the cloud service platform U22 receives the data sent by the Bluetooth gateway U21, it sends it to the cloud server U31 and stores it in the MySQL database U221.

[0041] The information processing module U3 includes: a cloud server U31, wherein the cloud server U31 includes three processing algorithms and a neural network model U314.

[0042] Specifically, the neural network model U314 is used to process the data, and different algorithms are used to calculate different data. Specifically, the temperature compensation algorithm U311 is used to calculate the real-time body temperature, the triangulation positioning algorithm U312 is used for triangulation positioning, and the peak detection algorithm U313 is used to calculate the amount of exercise. The temperature compensation algorithm U311 calculates the real-time temperature using the following formula:

[0043]

[0044] Where R0 is the resistance value at the reference temperature T0, B is the material constant of the thermistor, and R is the resistance value at the measured temperature T (also in Kelvin).

[0045] The above is a conventional temperature calculation method. Since the material constant used by the thermistor will change due to temperature changes, the present invention will go a step further on the basis of this method and take the change of ambient temperature into account. Among them, the ambient temperature is measured by the ambient temperature sensor U12 to further reduce the error obtained by the measurement. For example, when the ambient temperature changes, the material constant of the thermistor becomes B+ΔB. According to the resistance error calculation formula, the compensated temperature is calculated as:

[0046]

[0047] Then the temperature error ΔT=T new -T, the error calculated by this formula can be used to infer the range of the true temperature value, T ture =T-ΔT / 2 to T+ΔT / 2. The value in different ranges is taken according to the different ambient temperature ranges measured by the ambient temperature sensor U12. For example, if the ambient temperature is higher than 26°C, the median value of T+ΔT / 2 is taken. If the temperature is lower than 26°C, the median value of T-ΔT / 2 is taken. This further reduces the error caused by the change of ambient temperature and improves the measurement accuracy of the system.

[0048] Specifically, the triangulation positioning algorithm U312 method for positioning individual animals based on the signal strength (Received Signal Strength Indication, RSSI) received by the Bluetooth gateway U21 is as follows:

[0049] Three signal access points (APs) with known positions are arranged in the farm area, namely the Bluetooth gateway U21 in the present invention, which are marked as AP1, AP2, AP3, etc., and the access coordinates of these points are set to (x1, y1), (x2, y2), and (x3, y3) respectively.

[0050] The target device to be located, i.e., the smart ear tag U11, starts the signal sending function. The signal access point receives the RSSI values ​​of these signals with the located target. Let the signal received by AP1 be RSSI1, the signal received by AP2 be RSSI2, and the signal received by AP3 be RSSI3. According to the wireless signal propagation model, the logarithmic distance path loss model is generally used:

[0051] RSSI = P t -10nlog 10 (d)+X σ

[0052] Among them, P t is the transmission power, n is the path loss index (related to the environment, n=2 in free space, and n is between 2-5 in this invention), d is the signal propagation distance, X б is a Gaussian random variable with mean 0.

[0053] According to the above formula, for each access point, the estimated value of the distance between the target device and the gateway can be calculated. Taking AP1 as an example, substituting RSSI1 into the formula, the calculation formula for the estimated distance d1 is:

[0054]

[0055] Similarly, the distance d2 of AP2 and the distance d3 of AP3 can be calculated.

[0056] Taking AP1 and AP2 as an example, let the target device location coordinates be (x, y). According to the distance formula between two points (i represents the access point number). For AP1 and AP2, the following two equations can be obtained:

[0057] Equation 1:

[0058] Equation 2:

[0059] Subtract equation 2 from equation 1 to eliminate x 2 After adding the terms y and y2, we can get a linear equation about x and y, which is set as Equation 3. Then, AP1 and AP3 are cancelled in the same way, and another equation with x and y is set as Equation 4. By combining Equation 3 and Equation 4, we can get the position coordinates (x, y) of the smart ear tag U11.

[0060] Specifically, the principle of the peak detection algorithm U313 for motion detection based on the three-axis acceleration sensor is as follows:

[0061] The peak detection algorithm U313 identifies peak points from the acceleration signal. Peak points are often related to the movement actions of animals, and are then used to calculate exercise amount indicators such as the number of steps. First, record the time series data collected by the acceleration sensor in the master label U112, which are a x (t), a y (t) and a z (t), where t represents time. Usually, the acceleration data in the vertical direction (i.e., the z direction) is the key, and the x and y directions provide auxiliary information. For the time series a(t), if a time point t0 satisfies the following conditions, it is considered a local peak:

[0062] a(t0 - 1) < a(t0) and a(t0 + 1) < a(t0)

[0063] where t0 is the current time point, and a(t0 - 1) and a(t0 + 1) are the acceleration values at the time points before and after t0 respectively. For continuous peaks, calculate the interval Δt between them. If Δt falls within the time interval range corresponding to the normal movement step frequency of the animal (common animal step frequency intervals: (1) Pig: normally 1 to 1.5 seconds, 0.4 to 0.6 seconds when frightened; (2) Cow: normally 1 to 1.5 seconds, 0.27 seconds when frightened; (3) Sheep: normally 0.75 to 1 second, 0.33 to 0.5 seconds when frightened), it is considered a valid step frequency.

[0064] Furthermore, the present invention perceives and predicts the animal state based on the neural network model U314, integrates and calculates the processed data, namely real-time body temperature, real-time positioning and exercise amount, and identifies and predicts the animal state. For example: when the real-time body temperature of the animal is too high and the exercise amount is low, the neural network model predicts it as a disease; when the real-time body temperature of the animal is slightly high and the exercise amount increases simultaneously, the neural network model predicts it as estrus.

[0065] The intelligent management module U4 includes a visualization platform U41. The visualization platform U41 classifies and visually displays the results calculated by the three algorithms in the cloud server U31 and the results identified and predicted by the neural model U314. First, visually display the original data information of the animal, such as real-time positioning, real-time body temperature and exercise amount, and then visually display the results that may be abnormal identified and predicted by the neural network model U314 below the original data, including estrus, return to estrus, disease, injury. The visualization platform U41 provides a memo function, which retains the abnormal results identified and predicted by the neural network model U314 until the user deletes them manually, avoiding the user's omission of animal abnormal situations.

[0066] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A low-power animal individual identification, positioning and state perception system, characterized in that: The system includes an information collection module, an information transmission module, an information processing module, and an intelligent management module; The information collection module is used to collect individual information and ambient temperature; The information transmission module is used to receive the information collected by the information collection module to the Bluetooth gateway, and then send it to the cloud service platform. The cloud service platform forwards the data to the cloud server and stores it in the MySQL database; The information processing module is used to process the information sent by the information transmission module, and use the temperature compensation algorithm, the triangulation positioning algorithm, and the peak detection algorithm to respectively calculate the real-time body temperature, the real-time positioning, and the amount of exercise, and use the neural network model to integrate and calculate the above results to achieve the recognition and prediction of the state of the animal; The intelligent management module is used to visualize the processed data information and display possible abnormal situations calculated by the neural network model so that users can manage them in time.

2. The system according to claim 1, characterized in that The information collection module includes: an intelligent ear tag and an ambient temperature sensor; Among them, the smart ear tag includes a male tag and a female tag. The male tag contains a passive RFID tag with a unique ID to identify the animal's identity information. The female tag contains an accelerometer to collect the acceleration of the animal during movement, a thermistor to collect the animal's real-time body temperature, and BLE low-power Bluetooth to send the collected information. The ambient temperature sensor is used to collect the ambient temperature in the breeding farm. It also has BLE low-power Bluetooth inside, which uses the same Bluetooth protocol as the smart ear tag to send data to the Bluetooth gateway.

3. The system according to claim 2, characterized in that The smart ear tag is made of polyurethane material, so that the ear tag has a high fit and a reduced foreign body sensation, reducing errors caused by loose fit or foreign matter between the ear tag and the animal's skin, and will not cause damage to the animal through chewing, biting, or throwing behaviors; it is made using 3D printing technology to improve the sealing and service life of the ear tag; and the BLE low-power Bluetooth in the mother tag and the ambient temperature sensor only sends data but does not receive data, further reducing the size and power consumption of the system.

4. The system according to claim 3, characterized in that The intelligent ear tag and the temperature sensor in the information collection module use intermittent data collection, that is, the equipment in the information collection module uses a pulsed working current.

5. The system according to claim 1, wherein: The information transmission module includes: a Bluetooth gateway and a cloud service platform, wherein the communication range of the Bluetooth gateway needs to be able to fully cover the scope of the farm, the communication protocols supported by the Bluetooth gateway include the Bluetooth protocol and the WiFi protocol, and the communication method with the cloud server supports UDP, TCP and MQTT. In order to improve the anti-interference ability of the system, the collection device in the information collection module communicates with the Bluetooth gateway through the Bluetooth protocol and adopts multi-channel frequency hopping transmission.

6. The system according to claim 5, characterized in that The Bluetooth gateway is connected to the cloud service platform via WIFI and uses the MQTT protocol for transmission, which enables efficient and reliable communication between IoT devices even when bandwidth is limited and network is unstable.

7. The system according to claim 5, characterized in that A complete system from storage to computing to visualization is built based on the cloud service platform. In the information transmission module, after the cloud service platform receives the data sent by the Bluetooth gateway, it sends it to the cloud server and stores it in the MySQL database.

8. The system of claim 1, wherein: The information processing module includes: a cloud server, wherein the cloud server includes a temperature compensation algorithm, a triangulation positioning algorithm, a peak detection algorithm and a neural network model.

9. The system according to claim 8, characterized in that Based on the neural network model, the processed data, namely real-time body temperature, real-time positioning and exercise volume, are integrated and calculated to identify and predict the state of the animal.

10. The system according to claim 1, wherein: The intelligent management module includes a visualization platform, which classifies and visualizes the results of the three algorithm calculations in the cloud server and the results of the neural model recognition prediction. First, the original data information of the animal, such as real-time positioning, real-time body temperature and exercise volume, is visualized, and then the results of the neural network model recognition prediction that may have abnormalities are visualized below the original data. The predicted abnormal results include estrus, return of estrus, disease, and injury. The visualization platform provides a memo function, which retains the abnormal results predicted by the neural network model recognition until the user manually deletes them.

Images