System and method for determining feed intake of a bovine

Abstract

The present application relates to systems and methods for determining feed intake of a cow. A system for determining feed intake of a dairy cow, the system comprising: at least one sensor configured to be positioned on a collar and / or ear tag of the dairy cow, wherein the sensor is configured to collect data associated with head movement of the dairy cow; a processor in communication with a memory module having program code stored thereon, the program code executable by the processor to: receive signals from the at least one sensor, preprocess the received signals by classifying different time blocks of the received signals into behavior categories of the dairy cow, extract a plurality of features based at least in part on the classified time blocks, and calculate a feed intake of the dairy cow based on the extracted one or more features.

Description

Technical Field

[0001] This disclosure generally relates to systems and methods for determining feed intake in cattle. background

[0002] Today, determining feed intake in cattle (e.g., dairy cows) sometimes involves calculations based on animal factors such as milk energy, body weight (BW), days at lactation (DIM), and parity. However, the accuracy of such calculations is often relatively low and has a high slope bias, meaning the results are even less accurate for cows with the highest and lowest feed intake. This can be a significant problem because identifying these cows is crucial for making informed breeding and culling decisions to improve herd efficiency and profitability. Cow monitoring devices that can identify feed intake based on cow movement can be helpful; however, due to the high variability in feed rates within the herd, the timing of feeding itself is a poor predictor of feed intake. Furthermore, for some tracking devices, tracking cow / dairy cow movement over long periods can prove unsustainable, and these devices may require large amounts of storage space for detected signals and continuous communication with remote receivers. The hardware required for such tracking methods can also exhaust the tracking sensors and the batteries they need to maintain on the cow or dairy cow's body for at least several years.

[0003] Therefore, there is a need in the art for systems and methods for accurately tracking feed intake in cattle or dairy cows, and for systems and methods that can operate continuously for several years at a time.

[0004] Overview

[0005] According to some embodiments, this document provides a system for determining feed intake in dairy cows, the system comprising: at least one sensor configured to be positioned on a collar and / or ear tag of a dairy cow, wherein the sensor is configured to collect data associated with head movements of the dairy cow; a processor in communication with a memory module having program code stored thereon, the program code being executable by the processor to: receive signals from at least one sensor; preprocess the received signals by classifying different time blocks of the received signals into behavioral categories of the dairy cow; extract multiple features at least in part based on the classified time blocks; and calculate feed intake of the dairy cow based on one or more of the extracted features.

[0006] According to some embodiments, a method for measuring feed intake of dairy cows is provided, the method comprising: receiving signals from at least one sensor located on a collar and / or ear tag of the dairy cow; preprocessing the received signals by classifying different time blocks of the received signals into behavioral categories of the dairy cow; extracting multiple features based at least in part on the classified time blocks; and determining feed intake of the dairy cow by applying at least one algorithm to one or more of the extracted features.

[0007] According to some embodiments, the processor is also configured to calculate feed intake without calculating rumination in cows.

[0008] According to some embodiments, the processor is also configured to determine one or more of the cow's feeding time and feeding rate, and wherein the calculation of the cow's feed intake is based at least in part on one or more of the cow's feeding time and feeding rate.

[0009] According to some embodiments, the program code can also be executed by a processor to receive metadata associated with cows and / or herds.

[0010] According to some embodiments, the metadata includes any one or more of the following: the age of the cow, the number of days since calving, the health status of the cow, the cow's profile, the cow's weight, the type of food consumed by the cow and / or herd, the number of calves the cow has given birth to, the gestation period of the cow, and the herd number.

[0011] According to some embodiments, the health status of a cow includes any one or more of the following: the cow's pregnancy status, the amount of milk produced by the cow, the milk composition of the milk produced by the cow, the number of days since the last calving, and any reported injuries to the cow.

[0012] According to some embodiments, the system is also configured to convert the received signal into a frequency domain representation.

[0013] According to some embodiments, the sensor includes an accelerometer, and the signal is associated with the kinematics of the cow.

[0014] According to some embodiments, the received signal includes a resolution of at least one millisecond.

[0015] According to some embodiments, the length of each time block ranges from approximately 30 seconds to approximately 5 minutes.

[0016] According to some embodiments, the behavior category includes any one or more of the following: eating, rumination, walking, jumping, breathing rate, or any combination thereof.

[0017] According to some embodiments, the storage module is configured to store the received signals in one or more clusters, each cluster being associated with a behavioral category of the cow.

[0018] According to some embodiments, the storage module is configured to store the received signals in the collar and / or ear tag.

[0019] According to some embodiments, the processor is configured to send the received signals from the collar and / or ear tag to the receiver at predetermined time intervals.

[0020] According to some embodiments, the predetermined time period includes approximately 5 minutes to approximately 24 hours.

[0021] According to some embodiments, the processor is configured to transmit received signals from the collar and / or ear tag to the receiver based on a predetermined transmission mechanism.

[0022] According to some embodiments, the memory module or a portion thereof is located on the collar and / or ear tag.

[0023] According to some embodiments, the method also includes sorting the cows relative to the herd.

[0024] According to some embodiments, the method also includes sorting the dairy herd relative to other dairy herds.

[0025] According to some embodiments, the method also includes determining the feed intake of the dairy cows based at least in part on the amount of food given to the herd and / or the price of the food given to the herd.

[0026] According to some implementations, the feed intake of dairy cows includes feed intake over a period of approximately one week.

[0027] According to some embodiments, the method further includes determining the amount of food intake by applying one or more machine learning modules to the time blocks classified as feeding time blocks.

[0028] According to some embodiments, the method further includes calculating the frequency based on the received signal.

[0029] According to some embodiments, the frequency is between 1 GHz and 2 GHz. According to some embodiments, the frequency is between 2 GHz and 3 GHz. According to some embodiments, the frequency is between 1 GHz and 3 GHz.

[0030] According to some embodiments, the method also includes classifying time blocks at least in part based on the variance of signal frequencies within the time block.

[0031] According to some embodiments, the method further includes clustering multiple feeding time blocks into one or more meal time blocks, and wherein determining the cow's feed intake includes applying at least one algorithm to one or more features extracted from the data in the meal time blocks.

[0032] According to some embodiments, determining feed intake includes applying extracted features to a machine learning algorithm.

[0033] According to some embodiments, feed intake of dairy cows is determined by applying at least one algorithm to one or more extracted features and one or more additional features.

[0034] According to some embodiments, additional features include any one or more of the following: metadata of the cow and data associated with the milk produced by the cow, or any combination thereof.

[0035] According to some embodiments, the metadata of a cow includes any one or more of the following: the cow's age, the cow's health status, the cow's profile, the cow's weight, the type of food consumed by the cow and / or herd, the number of calves the cow has given birth to, the gestation period of the cow, and the herd number.

[0036] According to some embodiments, the health status of a cow includes any one or more of the following: the cow's pregnancy status and reported injuries to the cow.

[0037] According to some embodiments, data associated with milk produced by a cow includes any one or more of the following: the amount of milk produced by the cow, the milk composition of the milk produced by the cow, and the number of days since the last calving.

[0038] According to some embodiments, the data associated with milk produced by cows includes data input by the user and / or data received from milk sensors.

[0039] Some embodiments of this disclosure may include some, all, or none of the advantages described above. One or more other technical advantages will readily occur to those skilled in the art based on the accompanying drawings, description, and claims contained herein. Furthermore, while specific advantages have been set forth above, various embodiments may include all, some, or none of the listed advantages.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. In case of conflict, the patent specification (including definitions) shall prevail. As used herein, unless the context clearly indicates otherwise, the indefinite articles “a” and “an” mean “at least one” or “one or more”. Brief description of the attached diagram

[0041] Some embodiments of this disclosure are described herein with reference to the accompanying drawings. The description, taken in conjunction with the drawings, makes it apparent to those skilled in the art how some embodiments can be practiced. The drawings are for illustrative purposes and are not intended to show structural details of the embodiments in more detail than necessary for a basic understanding of this disclosure. For clarity, some objects depicted in the drawings are not drawn to scale. Furthermore, two different objects in the same drawing may be drawn to different scales. In particular, the scale of some objects may be significantly exaggerated compared to other objects in the same drawing.

[0042] In block diagrams and flowcharts, optional elements / components and optional stages can be included within dashed boxes.

[0043] In the attached diagram:

[0044] Figure 1 This is a system for determining feed intake in dairy cows according to some embodiments of the present invention;

[0045] Figure 2 This is a flowchart of functional steps in a method for transmitting data from collars and / or ear tags of a large herd of cattle, according to some embodiments of the present invention; and

[0046] Figure 3 This is a flowchart of functional steps in a method for determining feed intake according to some embodiments of the present invention. Detailed description

[0047] Referring to the accompanying descriptions and figures will provide a better understanding of the principles, uses, and implementations taught herein. By carefully reading the descriptions and figures presented herein, those skilled in the art will be able to implement the teachings without much effort or experimentation. In the figures, the same reference numerals consistently refer to the same parts.

[0048] In the following description, various aspects of the invention will be described. For purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without the specific details presented herein. Furthermore, well-known features may have been omitted or simplified so as not to obscure the invention.

[0049] refer to Figure 1 , Figure 1 A system for determining feed intake in dairy cows according to some embodiments of the present invention is shown.

[0050] According to some embodiments, the system 100 may include a processor 102. According to some embodiments, the system 100 may include at least one sensor 104 communicating with the processor 102. According to some embodiments, the at least one sensor 104 may be located on a cow's collar and / or ear tag (or a dairy cow's collar and / or ear tag). According to some embodiments, the system 100 may include a storage module 106 communicating with the processor 102 and / or the at least one sensor 104. According to some embodiments, the system may include a user interface module 114 communicating with any one or more of the processor 102, the at least one sensor 104, and the storage module 106.

[0051] According to some embodiments, at least one processor 102 may be configured to analyze signals received from at least one sensor 104. According to some embodiments, the at least one processor may include a processor embedded in a collar and / or ear tag of a cow or dairy cow. According to some embodiments, the at least one processor 102 may include a remote server communicating with the collar and / or ear tag. According to some embodiments, the at least one processor 102 may be part of a receiving system configured to receive and analyze signals collected by at least one sensor 104.

[0052] According to some embodiments, the user interface module 114 may be configured to receive data (e.g., data associated with cows) from a user and / or other sensors separate from the collar and / or ear tag. According to some embodiments, the user interface module 114 may include a keyboard, screen, mouse, buttons, microphone, etc., or any combination thereof. According to some embodiments, the user interface module 114 may include a smartphone application or software. According to some embodiments, the user interface module 114 may be configured to input data associated with cows and / or herds.

[0053] According to some embodiments, the system may include a collar configured to surround the neck of a cow or dairy cow. According to some embodiments, the collar may include at least one processor 102 embedded therein. According to some embodiments, the collar may include a storage module 106 embedded therein and / or a portion of the storage module 106 (such as, for example, the collar storage 108 portion of the storage module 106).

[0054] According to some embodiments, the system may include an ear tag configured to be positioned on the ear of a cow or dairy cow. According to some embodiments, the ear tag may include at least one processor 102 embedded therein. According to some embodiments, the ear tag may include a storage module 106 embedded therein and / or a portion of the storage module 106 (such as, for example, the ear tag storage 108 portion of the storage module 106).

[0055] According to some embodiments, at least one sensor 104 may include a sensor positioned on the cow's collar and / or ear tag. According to some embodiments, at least one sensor 104 may include an accelerometer. According to some embodiments, at least one sensor 104 may include an accelerometer and an additional sensor (e.g., a Global Positioning System (GPS)) positioned on the collar and / or ear tag. According to some embodiments, at least one sensor 104 may be configured to collect data associated with the cow's head movements. According to some embodiments, at least one sensor 104 may be configured to collect data associated with the cow's kinematics. According to some embodiments, at least one sensor 104 may be configured to continuously collect data associated with the cow's movements. According to some embodiments, at least one sensor 104 may be configured to send the collected data associated with the cow's movements to a storage module 106 and / or a portion of the storage module. According to some embodiments, at least one sensor 104 may be configured such that the collected signal can have a resolution of at least 1 millisecond, at least 5 milliseconds, at least 10 milliseconds, at least 50 milliseconds, at least 100 milliseconds, or any range therebetween. Each possibility is a separate embodiment.

[0056] According to some embodiments, a sensor 104 may be configured to transmit collected signals to a receiver or receiving system. According to some embodiments, the receiver or receiving system may communicate with any one or more of the processor 102, at least one sensor 104, and / or storage module 106. According to some embodiments, the receiver system may be configured to receive data associated with the movement of a cow or dairy cow, or in other words, to receive signals collected by at least one sensor 104. According to some embodiments, the receiver system may be configured to receive data associated with the movement of multiple cows or dairy cows, or in other words, the receiver may be configured to receive multiple collected signals from multiple collars and / or ear tags. According to some embodiments, the data from the collars and ear tags may be comparable, or in other words, data received from the collar of one cow may be compared with data received from the ear tags of different cows.

[0057] According to some embodiments, and as described in more detail elsewhere herein, the receiver can receive data collected by at least one sensor 104 using methods for transmitting data from collars (and / or ear tags) of cattle in a large herd. According to some embodiments, the processor can be configured to be based on (e.g., Figure 2The predetermined transmission mechanism described herein transmits the collected signals from the collar (and / or ear tag) to the receiver. According to some embodiments, at least one sensor 104 may be configured to transmit the collected signals to the receiver at predetermined time intervals. According to some embodiments, at least one sensor 104 may be configured to transmit the collected signals to the storage module 106.

[0058] According to some embodiments, storage module 106 may include a collar (and / or ear tag) storage portion 108 embedded in a cow's collar. According to some embodiments, collar (and / or ear tag) storage 108 may be configured to store data (or signals) received from at least one sensor 104. According to some embodiments, collar (and / or ear tag) storage 108 may be configured to store up to one week's worth of data received from at least one sensor 104. According to some embodiments, collar (and / or ear tag) storage 108 may be configured to store data collected by sensor 104 before it is sent to a receiver. According to some embodiments, and as described in more detail elsewhere herein, collar (and / or ear tag) storage 108 may be configured to store data collected by sensor 104 until it can be sent to a receiver. For example, if system 100 detects a communication line in use, meaning that the collected signal cannot be sent to the receiver at that moment, the collected signal will be stored in collar (and / or ear tag) storage 108.

[0059] refer to Figure 2 , Figure 2 A flowchart illustrating functional steps in a method for transmitting data from collars and / or ear tags of a large herd of cattle, according to some embodiments of the present invention, is shown. According to some embodiments, a storage module may store program code or one or more algorithms 112 thereon. According to some embodiments, one or more algorithms may be configured to implement, as in... Figure 2 The method described in 200.

[0060] According to some embodiments, in step 202, method 200 may include collecting signals from at least one sensor located on the cow's collar (and / or ear tag) in a collar (and / or ear tag) storage device within at least one time block. According to some embodiments, in step 204, method 200 may include checking for unobstructed communication lines for transmitting the collected signals. According to some embodiments, in step 206, method 200 may include periodically checking for unobstructed communication lines until an unobstructed communication line is detected. According to some embodiments, in step 208, method 200 may include transmitting the collected signals to a receiver (or receiving system) for detected unobstructed communication lines.

[0061] According to some embodiments, method 200 can be configured to manage the transmission of collected signals from multiple cows to one or more receivers (or receiving systems). According to some embodiments, method 200 can be configured to transmit the collected signals from multiple cows such that no data (or signal) is lost due to unavailable communication lines (or the communication lines in use). Advantageously, by having different collars (and / or ear tags) communicate with the receivers at different times, method 200 enables the transmission of signals collected from multiple collars (and / or ear tags) to a single receiver without data loss.

[0062] According to some embodiments, in step 202, method 200 may include collecting signals from at least one sensor located on the cow's collar (and / or ear tag) in a collar (and / or ear tag) storage device within at least one time block. According to some embodiments, the signals may be collected in the collar (and / or ear tag), such as, for example, in a collar (and / or ear tag) storage device. According to some embodiments, method 200 may include tagging the collected data to be associated with a specific time block. According to some embodiments, the time block may be about 30 seconds, 1 minute, 2 minutes, 5 minutes, 7 minutes, 10 minutes, 13 minutes, 15 minutes, 25 minutes, or any range therebetween. Each possibility is a separate embodiment.

[0063] According to some embodiments, the method may include transmitting the collected signals using a processor. According to some embodiments, the processor may be configured to transmit the collected signals from a collar (and / or ear tag) to a receiver. According to some embodiments, the method may include transmitting the collected signals from the collar (and / or ear tag) to the receiver at predetermined time intervals. According to some embodiments, the predetermined time interval may be approximately 3 minutes, 5 minutes, 10 minutes, 30 minutes, 1 hour, 3 hours, 24 hours, or any range therebetween. Each possibility is a separate embodiment.

[0064] According to some embodiments, sending the collected signal to the receiver may include checking whether the communication line between the processor (or collar and / or ear tag) and the receiver is idle (or available). According to some embodiments, an unavailable communication line may be caused by other collars (and / or ear tags) sending the collected signal to the same receiver. According to some embodiments, an unavailable communication line may be caused by a loss of communication between the processor (or collar and / or ear tag) and the receiver, i.e., for example, if the cow is outside the receiver's range (or has traveled too far from the receiver).

[0065] According to some embodiments, in step 204, method 200 may include checking for accessible communication lines for transmitting the collected signals. According to some embodiments, in step 208, method 200 may include periodically checking for accessible communication lines for detected unavailable communication lines until an accessible communication line is detected. According to some embodiments, the method may include checking communication lines at least after each time block. According to some embodiments, the method may include storing the collected signals in packets, one or more of which may be associated with the time block in which the signals were collected. According to some embodiments, after collecting signals within a single time block, the method may include storing the collected signals in one or more packets in a collar (and / or ear tag) storage. According to some embodiments, the method may include storing the collected signals associated with one or more time blocks as one or more packets in a collar (and / or ear tag) storage for detected unavailable communication lines until an accessible communication line is detected.

[0066] According to some embodiments, in step 206, method 200 may include sending a collected signal to a receiver (or receiving system) for a detected unobstructed communication line. According to some embodiments, sending the collected signal to the receiver may include sending one or more packets from the collar (and / or ear tag) storage to the receiver. According to some embodiments, the method may include sending, along with one or more packets, a flag indicating that more packets are to be sent to the receiver (or in other words, a binary indicator that a communication line should be maintained between the collar (and / or ear tag) storage and the receiver). According to some embodiments, for the flag indicating that more packets are to be sent, the method may include maintaining a communication line between the collar (and / or ear tag) storage and the receiver. According to some embodiments, the method may include sending, along with the last sent packet, a flag indicating that no packets are to be sent. According to some embodiments, the method may include disconnecting the communication line between the collar (and / or ear tag) storage and the receiver after receiving the flag indicating that no packets are to be sent.

[0067] According to some embodiments, the method may include sending a flag, along with a first packet of one or more packets, indicating the number of packets to be sent from the collar (and / or ear tag) storage to the receiver. According to some embodiments, the method may include disconnecting the communication line between the collar (and / or ear tag) storage and the receiver after the indicated number of packets has been sent. According to some embodiments, the method may include erasing one or more packets from the collar (and / or ear tag) storage once one or more packets have been sent to the receiver.

[0068] According to some embodiments, the system can be configured to receive the collected signals using a receiver (or receiving system). According to some embodiments, the method may include analyzing the received signals using one or more algorithms (e.g., algorithm 112).

[0069] refer to Figure 3 , Figure 3 A flowchart illustrating functional steps in a method for determining feed intake according to some embodiments of the present invention is shown. According to some embodiments, one or more algorithms 112 may be executable by a processor to implement, as described above. Figure 3 Method 300 is shown.

[0070] According to some embodiments, in step 302, method 300 may include receiving signals from at least one sensor positioned on the cow's collar (and / or ear tag). According to some embodiments, in step 304, method 300 may include preprocessing the received signals by classifying different time blocks of the received signals into behavioral categories of the cow. According to some embodiments, in step 306, method 300 may include extracting multiple features at least in part based on the classified time blocks. According to some embodiments, in step 308, method 300 may include determining the cow's feed intake by applying at least one algorithm to one or more of the extracted features.

[0071] According to some embodiments, the method may include receiving signals (e.g., the collected signals described above) from at least one sensor positioned on the cow's collar (and / or ear tag). According to some embodiments, the collected signals may be the same as the received signals, or in other words, these signals may be referred to as the received signals once they are received by a receiver (or receiving system).

[0072] According to some embodiments, the received signal may include data associated with the kinematics of the cow. According to some embodiments, the received signal may include data associated with any one or more of X-axis motion, Y-axis motion, and Z-axis motion, or any combination thereof. Each possibility is a separate embodiment. According to some embodiments, the received signal may be received by a receiver at intervals of one or more predetermined time periods, such as those described in more detail elsewhere herein. According to some embodiments, the received signal may be received by a receiver at intervals of approximately 3 minutes, 5 minutes, 10 minutes, 30 minutes, 1 hour, 3 hours, 24 hours, or any range thereof. Each possibility is a separate embodiment.

[0073] According to some embodiments, the method may include storing data associated with each cow in a database. According to some embodiments, the data may include received signals and other data associated with the cow. According to some embodiments, the database may include a timeline composed of multiple time blocks. According to some embodiments, the time blocks may be organized chronologically. According to some embodiments, each received signal is associated with a time block, and each time block may be stored in the database such that the data in the database consists of a continuous stream of data associated with the cow's movement (or kinematics).

[0074] According to some embodiments, the database may include providing an accurate timestamp associated with each time block (or in other words, timestamping the received signal, wherein each timestamp is configured to identify the start time of the time block). According to some embodiments, the timestamp may differ from the timestamp received along with the received signal due to an accurate clock in at least one sensor and / or collar (and / or ear tag). Due to energy-saving methods, the clock in at least one sensor and / or collar (and / or ear tag) may be unreliable and / or inaccurate. Therefore, timestamping the received signal with an accurate timestamp avoids the need for an accurate clock in the collar (and / or ear tag) and / or at least one sensor. Advantageously, having an accurate timestamp associated with the received signal allows for comparison of data received from the sensor with other cows during the real-time period of data reception, thereby enabling more accurate calculation of feed intake. According to some embodiments, the method may include receiving multiple packets associated with multiple time blocks at once, and retrospectively and accurately marking these time blocks by reading the relative time of the multiple time blocks (or the time associated with each packet, each packet being related to other packets received from the same collar (and / or ear tag) or sensor).

[0075] According to some embodiments, the method may include converting a received signal into a frequency domain representation. According to some embodiments, the method may include applying one or more Fourier transform algorithms to the received signal. According to some embodiments, the method may include determining one or more frequencies within the received signal. According to some embodiments, the method may include determining one or more frequencies within the received signal for each time block. According to some embodiments, the method may include determining one or more frequencies within the received signal for each time block separately. According to some embodiments, a continuous data stream associated with cow movement stored in a database may include the determined frequencies (as a function of time). According to some embodiments, the frequency range may be between 0.5 GHz and 4 GHz. According to some embodiments, the frequency range may be between 1 GHz and 3 GHz. According to some embodiments, the frequency range may be between 1 GHz and 2 GHz. According to some embodiments, the frequency range may be between 2 GHz and 3 GHz. Each possibility is a separate embodiment.

[0076] According to some embodiments, the method may include calculating the average frequency within each time block. According to some embodiments, the method may store the average frequency of each time block separately in a database. For example, for a 1.5-minute time block, the database would store the average frequency for each 1.5-minute interval. According to some embodiments, the method may include calculating the variance within each time block. According to some embodiments, the method may include storing the variance of each time block separately in a database.

[0077] According to some embodiments, the method may include receiving metadata associated with cows. According to some embodiments, the method may include receiving metadata from a user via a user interface module (such as user interface module 114 described above). According to some embodiments, the method may include storing the metadata associated with cows in a database.

[0078] According to some embodiments, metadata may include any one or more of the following: the cow's age, the cow's health status, the cow's profile, the cow's weight, the type of food consumed by the cow and / or herd, the number of calves born, the duration of gestation, and the herd number, or any combination thereof. Each possibility is a separate embodiment. According to some embodiments, metadata may include environmental parameters. According to some embodiments, the cow's health status may include any one or more of the following: the cow's pregnancy status, the amount of milk produced, the milk composition of the milk produced, the number of days since the last calving, the cow's milking status, and reported cow injuries, or any combination thereof. Each possibility is a separate embodiment.

[0079] According to some embodiments, a dairy herd can refer to the actual group in which the dairy cows are located, such as the herd of livestock in which the dairy cows belong. According to some embodiments, a dairy herd can refer to the shed in which the dairy cows are located.

[0080] According to some embodiments, a cow's milking status can refer to the cow's current position in a milking or lactation cycle. According to some embodiments, milking status can include conditions such as "dry" or "heifer".

[0081] According to some embodiments, the number of calves that a cow has given birth to can refer to the total number of calves that a cow has given birth to and / or the number of times a cow has given birth and / or the number of calves given birth to each time a cow gives birth, or any combination thereof. Each possibility is a separate embodiment.

[0082] According to some embodiments, a cow's pregnancy status may include whether the cow is pregnant, the stage of pregnancy, the sex of the fetus, the number of fetuses, the health status of one or more fetuses, or any combination thereof. Each possibility is a separate embodiment.

[0083] According to some embodiments, a cow profile may include any data that a user has about the cow that may not be associated with other metadata tags, such as, for example, the number of days the cow has been lost or gone.

[0084] According to some embodiments, environmental parameters may refer to input data and / or data received in other ways that are associated with the environment in which the cows live, such as, for example, weather, seasons of the year, temperature, living conditions, etc.

[0085] According to some embodiments, data associated with milk produced by a cow may include any one or more of the following: the amount of milk produced by the cow, the milk composition of the milk produced by the cow, and the number of days since the last calving, or any combination thereof. Each possibility is a separate embodiment. According to some embodiments, data associated with milk produced by a cow may be input by a user and / or received from a milk sensor. According to some embodiments, the milk sensor may measure yield, fat, protein, and / or any combination thereof. According to some embodiments, the milk sensor may include a spectrometer. According to some embodiments, the milk sensor may be configured to detect the amount of total fat (or total fatty acids) in milk produced by a cow, the amount of saturated fat in milk produced by a cow, the amount of unsaturated fat in milk produced by a cow, the amount of casein in milk produced by a cow, the amount of lactose in milk produced by a cow, or any combination thereof. Each possibility is a separate embodiment.

[0086] According to some embodiments, the database can be organized by: a numbering system for each cow, a numbering system for each herd, a numbering system for each pen, the type of feed the cows are receiving, the health status of the cows, the pregnancy status of the cows, etc., or any combination thereof. Each possibility is a separate embodiment.

[0087] According to some embodiments, in step 304, method 300 may include preprocessing the received signal by classifying different time blocks of the received signal into cow behavior categories. According to some embodiments, the method may include preprocessing a determined frequency for each time block. According to some embodiments, preprocessing may include classifying each time block of a cow into a cow behavior category. According to some embodiments, the method may include classifying each time block of a cow into a cow behavior category by applying the frequency of each time block to one or more classification algorithms. According to some embodiments, the method may include classifying time blocks at least in part based on the variance of the signal frequencies within the time block.

[0088] According to some embodiments, the method may include applying a model comprising one or more classifiers. According to some embodiments, the model may include one or more classifiers configured to classify time blocks into any one or more behavioral categories of cows. According to some embodiments, the model may include one or more classifiers configured to classify time blocks at least in part based on: the frequency of a signal within the time block, the variance of the frequency, one or more frequencies associated with other time blocks (unclassified), one or more previous classifications of one or more other time blocks, or any combination thereof. Each possibility is a separate embodiment. According to some embodiments, the model may include one or more Hidden Markov Models (HMMs). According to some embodiments, the model may include classifiers configured to use HMMs. According to some embodiments, the classifier may be configured to classify time blocks at least in part based on data received from at least one sensor. According to some embodiments, the classifier may be configured to classify time blocks at least in part based on extracted features.

[0089] According to some embodiments, the method may include classifying each time block of a cow into a behavioral category by applying the variance of each time block to one or more classification algorithms. According to some embodiments, for signal frequencies with variance above a predetermined threshold, a time block may be classified as a feeding time block. According to some embodiments, for signal frequencies with variance below a predetermined threshold, a time block may be classified as a rumination time block.

[0090] According to some embodiments, the classification algorithm may include one or more machine learning models. According to some embodiments, the classification algorithm may include any one or more of the following: logistic regression, Naive Bayes, k-Nearest Neighbors (k-NN), decision tree, support vector machine (SVM), hidden Markov model, boosting algorithms (such as, for example, adaptive boosting, gradient tree boosting, XGBoost, etc.), linear regression, decision tree, K-means, random forest, or any combination thereof. Each possibility is a separate embodiment.

[0091] According to some embodiments, one or more classification algorithms can be configured to classify each time block as associated with one or more behavioral categories of the cow. According to some embodiments, behavioral categories may include any one or more of the following: feeding, rumination, walking, walking speed, activity (such as, for example, jumping), respiratory rate, or any combination thereof. Each possibility is a separate embodiment.

[0092] According to some embodiments, respiratory rate classification may refer to respiratory rates that are above or below a predetermined threshold. According to some embodiments, respiratory rate may be associated with other health conditions of the cow (such as, for example, if the cow is experiencing heat stress).

[0093] According to some embodiments, the method may include clustering time blocks for each cow, wherein the clustering is associated with a classification of one or more behaviors of the cow. According to some embodiments, the method may include separating clusters associated with rumination and clusters associated with feeding.

[0094] According to some embodiments, the method may include determining that clusters in which time blocks are classified as feeding behaviors are feeding time blocks. According to some embodiments, the method may include clustering multiple feeding time blocks into one or more mealtime blocks. According to some embodiments, the method may include identifying feeding time blocks that are outside of (or separate from) mealtime blocks, or in other words, that are outliers and considered as non-feeding time blocks even if their frequency and / or variance may be similar to other feeding time blocks of the cows.

[0095] According to some embodiments, mealtime blocks may include one or more time blocks that are not classified as having feeding behavior. For example, sometimes during a meal, a cow may rest between chewing its food. In cases where rest may exist between feeding time blocks, if the meal is still in progress and the time block is not classified as feeding behavior, the method may include identifying some time blocks within the meal as feeding-rest.

[0096] According to some embodiments, in step 306, method 300 may include extracting multiple features at least partially based on time blocks of classification. According to some embodiments, the method may include implementing feature construction and / or feature selection techniques. According to some embodiments, the method may include extracting multiple features from any one or more of the following: the frequency of time blocks, the variance of time blocks, the number of time blocks in a specific cluster associated with a particular behavior (or classification), the ratio between specific clusters, metadata associated with cows, data associated with other cows in the same herd, herd, and / or barn, cow health status, data associated with milk production, and / or any combination thereof. Each possibility is a separate embodiment. According to some embodiments, features may be based on metadata associated with cows. According to some embodiments, features may be based on time blocks of classification associated with cows. According to some embodiments, features may be based on a combination of metadata associated with cows and time blocks of classification associated with cows.

[0097] According to some embodiments, one or more algorithms may be used to extract, select, and / or construct features. According to some embodiments, the method may include applying time blocks and / or metadata of cows to an algorithm configured to extract one or more features. According to some embodiments, the algorithm may include one or more machine learning models. According to some embodiments, the algorithm may not have a machine learning model. According to some embodiments, the algorithm may include one or more models that are functions generated by one or more machine learning models.

[0098] According to some embodiments, in step 308, method 300 may include determining the cow's feed intake by applying at least one algorithm to one or more extracted features. According to some embodiments, the method may include applying one or more algorithms configured to calculate feed intake to the extracted features. According to some embodiments, calculating feed intake may include calculating the amount of consumed feed already ingested by the cow. According to some embodiments, the algorithm may be configured to calculate the amount of time the cow has been eating and / or the cow's feeding rate (or chewing rate). According to some embodiments, the algorithm may be configured to calculate feed intake without calculating rumination by the cow.

[0099] According to some embodiments, determining a cow's feed intake can be achieved by applying at least one algorithm to one or more features extracted from data in feeding time blocks. According to some embodiments, the algorithm can be configured to identify the amount of feed consumed by a cow during one or more feeding time periods. According to some embodiments, the method may include applying the algorithm to one or more feeding time blocks of the cow and extracting features.

[0100] According to some embodiments, the algorithm can be configured to rank (or score) a cow's feed intake relative to the feed intake of other cows in the same herd (e.g., a herd and / or a herd with the same metadata parameters). According to some embodiments, the algorithm can be configured to rank (or score) a cow's feed intake relative to the feed intake of other cows in the same herd (e.g., a stockyard or a specific shed). According to some embodiments, the algorithm can be configured to rank (or score) a cow's feed intake relative to the feed intake of other cows with one or more of the same metadata parameters (e.g., the same pregnancy status, health status, milk yield data, etc.). For example, the feed intake of a pregnant cow can be ranked relative to the feed intake of other pregnant cows. For example, the feed intake of a pregnant cow carrying twins can be ranked relative to the feed intake of other pregnant cows carrying twins. For example, the feed intake of a lame cow can be ranked relative to the feed intake of other lame cows. For example, the feed intake of non-milk-producing cows can be ranked relative to the feed intake of other non-milk-producing cows.

[0101] According to some embodiments, the method may include receiving input data from a user associated with the amount of feed provided to a herd of dairy cows (e.g., a herd or shed). According to some embodiments, the data associated with the amount of feed provided to the herd may include the weight of the feed. According to some embodiments, the data associated with the amount of feed provided to the herd may include the type of feed. According to some embodiments, the data associated with the amount of feed provided to the herd may include the price of the feed. According to some embodiments, the data associated with the amount of feed provided to the herd may include the percentage of dry matter in the feed. According to some embodiments, the algorithm may be configured to output the amount of feed consumed by the cows based at least in part on the input data associated with the amount of feed provided to the herd.

[0102] According to some embodiments, the algorithm can be configured to output the amount of food consumed by a cow per day. According to some embodiments, the algorithm can be configured to output the amount of food consumed by a cow per week. According to some embodiments, the algorithm can be configured to output the amount of food consumed by a cow per month.

[0103] In the specification and claims of this application, the words “comprising” and “having” and their various forms are not limited to members of the list that may be associated with these words.

[0104] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. In case of conflict, the patent specification (including definitions) shall prevail. As used herein, unless the context clearly indicates otherwise, the indefinite articles “a” and “an” mean “at least one” or “one or more”.

[0105] It should be understood that certain features of this disclosure described in the context of individual embodiments for clarity may also be provided in combination in a single embodiment. Conversely, various features of this disclosure described in the context of individual embodiments for brevity may also be provided individually or in any suitable sub-combination, or deemed suitable for any other described embodiments of this disclosure. Features described in the context of an embodiment are not considered essential features of that embodiment unless expressly stated otherwise.

[0106] Although the stages of a method according to some embodiments may be described in a specific order, the method of this disclosure may include some or all of the described stages performed in a different order. The method of this disclosure may include some or all of the described stages. Unless expressly specified as such, a particular stage in the disclosed method is not considered a necessary stage of the method.

[0107] While this disclosure has been described in conjunction with specific embodiments thereof, it will be apparent to those skilled in the art that many alternatives, modifications, and variations may exist. Therefore, this disclosure covers all such alternatives, modifications, and variations falling within the scope of the appended claims. It should be understood that this disclosure is not necessarily intended to limit its application to the details and / or methods of the construction and arrangement of the components set forth herein. Other embodiments may be practiced, and embodiments may be performed in various ways.

[0108] The wording and terminology used herein are for descriptive purposes and should not be construed as restrictive. Any reference or identification of any source in this application should not be construed as an admission that such reference can be used as prior art in this disclosure. Section headings used herein are for ease of understanding and should not be construed as necessary limitations.

[0109] This invention can be a system, method, and / or computer program product. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to perform aspects of the invention.

[0110] A computer-readable storage medium can be a tangible device capable of holding and storing instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable optical disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices on which instructions are recorded, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium should not be construed as a transient signal, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires. Rather, a computer-readable storage medium is a non-transitory (i.e., non-volatile) medium.

[0111] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a corresponding computing / processing device, or downloaded via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network) to an external computer or external storage device. This network may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them for use in a computer-readable storage medium stored within the corresponding computing / processing device.

[0112] Computer-readable program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer (or cloud) may be connected to the user's computer via any type of network (including local area networks (LANs) or wide area networks (WANs)), or may be connected to an external computer (e.g., via the Internet provided by an Internet service provider) (including wired or wireless connections such as, for example, Wi-Fi, BitTorrent, mobile, etc.). In some embodiments, electronic circuits including, for example, programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) can execute computer-readable program instructions by utilizing state information of computer-readable program instructions to personalize the electronic circuits in order to perform aspects of the present invention.

[0113] Various aspects of the invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block and combination of blocks in the flowchart illustrations and / or block diagrams can be implemented by computer-readable program instructions.

[0114] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus for manufacturing machines, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / actions specified in one or more boxes of a flowchart and / or block diagram. These computer-readable program instructions may also be stored in a computer-readable storage medium that instructs a computer, programmable data processing equipment, and / or other device to function in a particular manner, such that the computer-readable storage medium storing the instructions includes an article of manufacture comprising instructions for implementing aspects of the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0115] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer-implemented process, such that the instructions that execute on the computer, other programmable apparatus or other device implement the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0116] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, paragraph, or part of an instruction, which may include one or more executable instructions for implementing a specified logical function (multiple specified logical functions). In some alternative implementations, the functions recorded in the blocks may not appear in the order shown in the drawings. For example, two blocks shown consecutively may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart illustrations, and combinations of blocks in the block diagram and / or flowchart illustrations, can be implemented by a dedicated hardware-based system that performs a specified function or action or executes a combination of dedicated hardware and computer instructions.

[0117] For illustrative purposes, various embodiments of the invention have been described, but are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein has been chosen to best explain the principles of the embodiments, practical applications of techniques found in the market, or improvements to the technology, or to enable those skilled in the art to understand the embodiments disclosed herein.

Claims

1. A system for determining an amount of food consumed by a dairy cow, the system comprising: at least one sensor configured to be positioned on a collar and / or ear tag of a dairy cow, wherein the sensor is configured to collect data associated with head movement of the dairy cow; a processor in communication with a memory module having program code stored thereon, the program code executable by the processor to: receive signals from the at least one sensor; pre-process the received signals by classifying different time blocks of the received signals into behavior categories of the dairy cow; extract a plurality of features based at least in part on the classified time blocks, wherein at least one of the features is used to rank the dairy cow relative to a herd of dairy cows to which the dairy cow belongs; receive input data associated with an amount of food provided to the herd of dairy cows; and calculate the amount of food consumed by the dairy cow based on the extracted feature or features, wherein the calculation of the amount of food is based at least on the ranking of the amount of food consumed by each dairy cow relative to the amount of food consumed by the herd of dairy cows to which the dairy cow belongs and the input data associated with the amount of food provided to the herd of dairy cows.

2. The system of claim 1, wherein, the processor is further configured to separate between time blocks classified as eating time blocks and between time blocks classified as ruminating time blocks, wherein the calculation of the amount of food is based on the eating time blocks and not on the ruminating time blocks.

3. The system of any one of claims 1-2, wherein, the processor is further configured to determine one or more of an eating time and an eating speed of the dairy cow, and wherein calculating the amount of food consumed by the dairy cow is based at least in part on one or more of the eating time and the eating speed of the dairy cow.

4. The system of any one of claims 1-3, further comprising receiving metadata associated with the dairy cow and / or the herd of dairy cows.

5. The system of claim 4, wherein, the metadata comprises any one or more of an age of the dairy cow, a number of days since the dairy cow gave birth, a health condition of the dairy cow, a profile of the dairy cow, a weight of the dairy cow, and a type of food consumed by the dairy cow and / or the herd, a number of calves the dairy cow has given birth to, a length of time the dairy cow is pregnant, and a herd number of the herd of dairy cows.

6. The system of claim 5, wherein, the health condition of the dairy cow comprises any one or more of a pregnancy condition of the dairy cow, an amount of milk produced by the dairy cow, a milk composition of the milk produced by the dairy cow, a number of days since the dairy cow gave birth, and a reported injury of the dairy cow.

7. The system of any one of claims 1-6, wherein, the system is further configured to convert the received signals into a frequency domain representation.

8. The system of any one of claims 1-7, wherein, the sensor comprises an accelerometer, and wherein the signals are associated with kinematics of the dairy cow.

9. The system of any one of claims 1-8, wherein, the received signals comprise a resolution of at least one millisecond.

10. The system of any one of claims 1-9, wherein, a length of each time block comprises 30 seconds to 5 minutes.

11. The system of any one of claims 1-10, wherein, the behavior categories comprise any one or more of eating, ruminating, walking, jumping, breathing rate.

12. The system of any one of claims 1-11, further comprising a storage module, wherein, the memory module is configured to store the received signals in one or more clusters, wherein each cluster is associated with a behavior category of the dairy cow.

13. The system of any one of claims 1-12, wherein, The storage module is configured to store the received signals in the collar and / or ear tag.

14. The system of claim 13, wherein, The processor is configured to send the received signals from the collar and / or ear tag to a receiver every predetermined time period.

15. The system of claim 14, wherein, The predetermined time period comprises between 5 minutes and 24 hours.

16. The system of any one of claims 1-15, wherein, The processor is configured to send the received signals from the collar and / or ear tag to a receiver based on a predetermined sending mechanism.

17. The system of any one of claims 1-16, wherein, The memory module or a part of the memory module is located on the collar and / or ear tag.

18. A method for measuring an amount of food consumed by a dairy cow, the method comprising: receiving signals from at least one sensor positioned on a collar and / or ear tag of a dairy cow; preprocessing the received signals by classifying different time blocks of the received signals into behavior categories of the dairy cow; extracting a plurality of features based at least in part on the classified time blocks, wherein at least one of the features is used to rank the dairy cow relative to a herd of dairy cows to which the dairy cow belongs; receiving input data associated with an amount of food provided to the herd of dairy cows; and determining an amount of food consumed by the dairy cow by applying at least one algorithm to the extracted feature or features, wherein the calculation of the amount of food is based at least on the ranking of the amount of food consumed by each dairy cow relative to the amount of food consumed by the herd of dairy cows to which the dairy cow belongs and the input data associated with the amount of food provided to the herd of dairy cows.

19. The method of claim 18, further comprising ranking the dairy cow relative to the herd of dairy cows.

20. The method of any one of claims 18-19, further comprising ranking the herd of dairy cows relative to other herds of dairy cows.

21. The method of any one of claims 18-20, further comprising determining the amount of food consumed by the dairy cow based at least in part on an amount of food given to the herd of dairy cows and / or a price of food given to the herd of dairy cows.

22. The method of any one of claims 18-21, wherein, The amount of food consumed by the dairy cow comprises an amount of food consumed over a period of one week.

23. The method of any one of claims 18-22, wherein, For time blocks classified as eating, the amount of food consumed is determined by applying one or more machine learning modules to the time blocks classified as eating time blocks.

24. The method of any one of claims 18-23, further comprising calculating a frequency from the received signals.

25. The method of claim 24, wherein, The frequency is between 1 GHz and 2 GHz.

26. The method of any one of claims 24-25, further comprising classifying the time blocks based at least in part on a variance in the frequency of the signals within the time blocks.

27. The method of any one of claims 18-26, further comprising clustering a plurality of eating time bins into one or more meal time bins, and wherein, Determining the amount of food consumed by the dairy cow comprises applying at least one algorithm to the extracted feature or features of data in the meal time blocks.

28. The method of any one of claims 18-27, wherein, Determining the amount of food consumed comprises applying the extracted features to a machine learning algorithm.

29. The method of any one of claims 18-28, wherein, Determining the amount of food consumed by the dairy cow comprises applying at least one algorithm to the extracted feature or features and one or more additional features.

30. The method of claim 29, wherein, The additional features comprise any one or more of: metadata of the dairy cow and data associated with milk produced by the dairy cow.

31. The method of claim 30, wherein, The metadata of the cow includes any one or more of: an age of the cow, a health of the cow, a profile of the cow, a weight of the cow, and a type of food consumed by the cow and / or herd, a number of calves the cow has given birth to, how long it takes for the cow to become pregnant, and a herd number of a herd of the cow.

32. The method of claim 31, wherein, The health of the cow includes any one or more of: a pregnancy status of the cow and a reported injury of the cow.

33. The method of any one of claims 30-32, wherein, The data associated with the milk produced by the cow includes any one or more of: an amount of milk produced by the cow, a milk composition of the milk produced by the cow, and a number of days since the last calf birth.

34. The method of any one of claims 30-33, wherein, The data associated with the milk produced by the cow includes data input by a user and / or data received from a milk sensor.

Images