Method and control device for investigating the feeding behavior of animals

Abstract

The present disclosure relates generally to keeping animals (10) and in particular to a method for investigating eating behavior of animals (10) in a livestock area (30). The present disclosure also relates to a control device (100) and to a computer program for performing the method. According to a first aspect, the present disclosure relates to a method for investigating eating behavior of animals (10) in a livestock area (30). The method comprises monitoring animal data for a period of time after feed has been dispensed at a feeding station (31) in the livestock area (30), the animal data being collected using tags carried by the animals (10) and being indicative of activity of the animals (10) and / or position of the animals (10) relative to the feeding station (31). The method further comprises performing an action when the monitored animal data fails to meet one or more eating criteria, the one or more eating criteria defining normal eating behavior of the animals (10) in terms of activity of the animals (10) and / or position of the animals (10) relative to the feeding station (31). The present disclosure also relates to a control device (100) and to a computer program for performing the method.

Description

Technical Field

[0001] This disclosure generally relates to the raising of animals, and more specifically to a method for investigating the feeding behavior of animals in livestock enclosures. This disclosure also relates to a control device and a computer program for performing the method. Background Technology

[0002] Feeding dairy animals or other livestock is an important daily task for farmers. Animals receive rations such as roughage or complete or partial feed at feeding platforms or pens. Regular feed intake is essential for animals to ensure high feed intake, thereby maintaining a healthy body capable of supporting high milk production and pregnancy. Therefore, ensuring that each animal can eat what it needs is crucial for maintaining high milk production.

[0003] However, various factors (such as overcrowding, aggression at the feeding station, poor health, and other factors) can affect the frequency of animal visits to the feeding area and feeding station, and thus affect feed intake.

[0004] WO2018111180A1 proposes a method for providing adequate feed to animals (such as dairy cows) by grouping them based on information about their Body Condition Score (BCS) in order to provide each animal with the optimal amount of energy.

[0005] However, sometimes it is desirable to take measures to ensure that animals can eat what they need before the BCS is affected. Therefore, there is a need to improve methods to help farmers ensure that animals can eat what they need. Summary of the Invention

[0006] The objective of this disclosure is to mitigate at least some of the disadvantages of the prior art. Therefore, the aim is to provide a method for investigating the feeding behavior of animals in livestock enclosures. In particular, the aim is to implement a method for investigating animal feeding behavior that allows measures to be taken before deviant feeding behaviors affect the animal's physical condition and milk production.

[0007] According to a first aspect, this disclosure relates to a method for investigating the feeding behavior of animals in a livestock area. The method includes monitoring animal data for a period of time after feed is dispensed at a feeding station in the livestock area. This animal data is collected using tags carried by the animals and indicates the animals' activity and / or their position relative to the feeding station. The method also includes taking action when the monitored animal data fails to meet one or more feeding criteria, which define normal feeding behavior for the animals regarding their activity and / or their position relative to the feeding station. The proposed method allows for the detection of abnormal feeding behavior before consequences such as deviations from the BCS (Breakthrough Condition System) and reduced milk production occur. Therefore, measures can be taken when necessary to mitigate abnormal feeding behavior and prevent adverse effects on animal health and milk production. This method facilitates the most efficient use of available resources while ensuring good animal health and enabling immediate intervention when feeding behavior changes.

[0008] According to some implementation schemes, one or more feeding criteria can be used to assess feeding behavior based on one or more of the following: animal location, animal movement, animal posture, chewing activity, and animal speed, as indicated by animal data. By assessing multiple animal parameters, a better assessment of feeding behavior can be made.

[0009] According to some implementation schemes, one or more feeding standards include individual feeding standards for individual animals, feeding standards effective for a subset of animals, and / or common feeding standards effective for all animals. In some cases, it is possible to use the same feeding standards for all animals in a herd, which is beneficial for monitoring. In other cases, by customizing feeding standards for individual animals or groups of animals, more accurate assessments can be made.

[0010] According to some implementation schemes, the method includes: determining feeding behavior of one or more animals based on activity and / or location indicated by monitored animal data; and evaluating the determined feeding behavior using one or more feeding criteria that define normal feeding behavior. By first determining feeding behavior based on animal data, it is possible to compare the actual behavior with normal feeding behavior defined by the criteria.

[0011] According to some implementation schemes, the identified feeding behaviors include one or more of the following: time spent at the feeding station, time pattern of time spent at the feeding station, actual feeding time, time pattern of actual feeding time, feeding or chewing rate, pressure level during feeding, movement during feeding, posture during feeding, and position at the feeding station. Therefore, multiple behaviors associated with feeding can be assessed to determine whether feeding behaviors are normal.

[0012] According to some implementation schemes, this method includes analyzing the location and animal activity indicated by the monitored animal data to determine potential factors contributing to the failure to meet the feeding criteria when the monitored animal data fails to meet one or more feeding criteria. Therefore, the causes of abnormal feeding behavior can also be identified, which can help mitigate such behavior.

[0013] According to some implementation schemes, the analysis of potential factors includes: detecting the location and / or activity of one or more of the following: overcrowding, aggressive behavior of animals, and animal displacement. Therefore, potential problems can be addressed through regrouping or similar methods.

[0014] According to some implementation plans, the actions taken include providing users with information about failure to meet one or more feeding criteria. This informs farmers of abnormal feeding behavior and allows them to take timely action to avoid consequences.

[0015] According to some implementation plans, the information provided indicates that one or more potential factors identified in one or more feeding criteria have not been met. Therefore, farmers will also be informed of the root cause of the abnormal feeding behavior.

[0016] According to some implementation schemes, the information provided includes instructions for performing activities. Therefore, farmers are informed of activities that enable them to resolve the situation. In some implementation schemes, automating these activities includes adjusting the feeding robot's work schedule, trajectory, speed, and / or operating mode.

[0017] According to some implementation schemes, the actions performed include automatically executing activities associated with animals in the livestock area. Therefore, adverse situations can sometimes be resolved without human interaction, such as that of farmers.

[0018] According to some implementation schemes, this action includes one or more of the following: regrouping animals into other groups, separating a pair or group of animals that affect each other's feeding behavior, instructing for health checks, and implementing automated livestock management. By performing these actions, reduced BCS and milk production can be avoided.

[0019] According to some implementation schemes, this method involves adjusting one or more feeding criteria based on location and activity indicated by monitored animal data. Therefore, feeding criteria can be optimized for individual herds, which reduces the risk of false alarms, etc.

[0020] According to some implementation plans, adjustments are made when, during the period of animal data monitoring, the deviation of milk production or physical condition scores from predefined reference values ​​remains within a predefined tolerance level. Therefore, feeding standards are adjusted when a behavior is confirmed to be normal and does not affect the physical condition or milk production of a particular herd or group of animals.

[0021] In some implementations, monitoring is performed using a Real-Time Location System (RTLS). Therefore, if RTLS is already deployed, this method can be implemented without additional hardware.

[0022] According to the second aspect, this disclosure relates to a control device for monitoring the feeding behavior of animals in a livestock area, wherein the control device is configured to perform the method according to the first aspect.

[0023] According to a third aspect, this disclosure relates to a computer program that includes instructions that, when executed by a computer, cause the computer to perform the method according to the first aspect.

[0024] According to a fourth aspect, this disclosure relates to a computer-readable medium comprising instructions that, when executed by a computer, cause the computer to perform the method described in accordance with the first aspect. Attached Figure Description

[0025] Figure 1 This is a top view of an exemplary livestock area including a feeding platform.

[0026] Figure 2 This is a flowchart of a proposed method for investigating animal feeding behavior.

[0027] Figure 3 This is a conceptual diagram of a real-time positioning system.

[0028] Figure 4 The control device according to the second aspect is shown. Detailed Implementation

[0029] Regular and sustained visits to the feeding station are essential for animals to ensure high feed intake, thus maintaining a healthy body capable of supporting high milk production and pregnancy. Animals that are unwilling or unable to visit the feeding station at will or for as long as they wish may suffer impaired health, lower milk production, and a higher risk of disease. Individual animals may have limited access to the feeding station due to problems within the herd, such as overcrowding or relocation. Furthermore, poor feed palatability may reduce an animal's feeding motivation. In other cases, reduced motivation to visit the feeding station may be caused by illness or disease.

[0030] Feeding stations are typically surrounded by an area where animals that are feeding or intend to feed are expected to reside. The proposed technology is based on the understanding that by using animal data collected using tags carried by the animals, it is possible to monitor, for example, when animals enter the feeding area and their behavior within it. The maximum and minimum number of visits to the feeding area and feeding station per day can be defined, depending on the lactation stage, milk production level, and milking permission in the case of an automated milking system (AMS) in the barn. Furthermore, the time animals need to remain in a position where they can reach the feed (e.g., with their heads above the feeding station) can be defined to ensure a specific feed intake based on the animal's production level. If an animal or group of animals deviates from this predefined feeding behavior, action should be taken. For example, an alarm can be triggered, allowing the farmer to intervene and ensure the approach of the animal or group of animals.

[0031] By analyzing historical animal data, it is also possible to determine the optimal number of animals that should reside in a specific portion of the feeding area to ensure that all animals in that area receive adequate feed intake. This number can vary within the herd and over time, depending on feed quality and management (feeding times, animal flow settings, feed delivery).

[0032] Now refer to Figures 1 to 4 The proposed technology will be described in more detail. Figure 1 A feeding platform 31 is shown arranged in the livestock area 30. The livestock area 30 includes a living area 2 in which animals 10 can roam freely, small compartments 4, and a feeding robot 1. The proposed technology is suitable for dairy animals, such as, for example, dairy cows, buffalo, sheep, or goats.

[0033] Feeding platform 31, as used herein, refers to any surface on which feed is disposed and does not need to be separately arranged, or it may be part of the floor. In other words, feeding platform 31 is the area where feed is distributed for animals to eat. Feeding platform 31 typically comprises separate feeding platform sections arranged along one or more aisles of the livestock area 30, such as a barn. Feeding platform 31 may include several (separate or connected) feeding platform sections, which may be considered as separate feeding platforms 31. However, for simplicity, only one feeding platform 31 is shown. In this disclosure, these separate feeding platforms 31 are referred to as a single feeding platform because feeding platform 31 is typically just part of the floor on which feed 32 is intended to be placed. Feeding platform 31 is typically divided into feeding positions. In some embodiments, a feeding position is a head lock.

[0034] A feeding fence 34 is arranged along the aisle between the feeding station 31 and the animal 10. The feeding fence 34 is arranged to separate the feeding station 31 from the living area 2. More specifically, the feeding fence 34 is a barrier to prevent the animal 10 from entering the feeding station 31, but through this barrier, the animal can reach the feed 32 located near the feeding fence 34 by passing its head through the feeding fence 34. Headlocks arranged in the feeding fence 34 are typically used to fix the position of an animal 10 to a feeding position.

[0035] When an animal wants to eat, it moves toward the feeding station 31. It can be assumed that most animals residing in the feeding area 33 near the feeding station 31 (e.g., in the aisle adjacent to the feeding station 31) are there because they are eating, have already eaten, or intend to eat. In other words, the feeding area 33 is defined as a predefined area of ​​the livestock area 30 near the feeding fence 34. In other words, the predefined area in which animals that are eating or intend to eat are expected to reside is referred to herein as the feeding area 33. The feeding area 33 may be defined differently depending on the livestock area 30 and the feeding station 31. Therefore, the feeding area 33 is typically predefined based on the design of the barn and observations of feeding behavior. For example, the feeding area 33 is the area within a predetermined distance (e.g., 3 meters, 4 meters, or 5 meters) from the feeding fence 34 on the side of the feeding fence 34 that allows the animal 10 to access the feed 32.

[0036] The feeding robot 1 operates along the feeding platform 31. The feeding robot 1 includes a feed dispensing mechanism 13 configured to deliver and / or redistribute feed within the livestock area 30. In the example shown, the feed dispensing mechanism 13 is a feed redistribution mechanism configured to redistribute or move feed 32. The illustrated feed redistribution mechanism includes a rotary propulsion feeder. This rotary propulsion feeder lifts, mixes, and aerates the feed 32 while repositioning it closer to the feeding fence. However, other possible implementations of the feed redistribution mechanism include buckets, skirts, plows, or some other type of feed redistribution mechanism. The feeding robot can be operated automatically by the livestock management system 20.

[0037] Animal 10 in livestock area 30 carries tag 51 (see Figure 4Animal data indicating the position of animals 10 in livestock area 30 relative to the feeding platform is collected using this tag. According to some embodiments, the tag is configured to collect activity data. For example, the tag may include a motion sensor or accelerometer configured to measure rate or velocity changes. If the tag is positioned on the animal's head or neck, it can be used to determine head movement and feeding activities such as chewing and / or swallowing. Different techniques for determining the feeding behavior of individual animals based on accelerometer data are known in the agricultural field. An example is described in the article "Development of an automatic classification system for eating, ruminating and resting behavior of cattle using an accelerometer," published in Grassland Science (ISSN 1744-6961), Vol. 54, No. 4, pp. 231-237, December 2008.

[0038] In some implementations, animal data indicates the activities of animal 10, such as activities associated with feeding. For example, animal data indicates chewing activities associated with feeding, animal posture, etc. Note that chewing activities associated with feeding should be distinguished from chewing activities associated with rumination, such as by analyzing chewing patterns or separately studying the animal's posture or whether the animal has its mouth open. Animal data may also indicate behaviors that may potentially interfere with the feeding of this animal or other animals, such as abnormal animal movement that may be a sign of aggression or stress.

[0039] Animal data, for example, is provided by a real-time location system (RTLS) installed in the livestock area 30. The RTLS 50 is a known type of system used to track the location of objects (such as animal 10) in real time using tags 51 attached to objects located in the livestock area 30 (e.g., […]). Figure 1 Animals 10 in the livestock area (of the facility). Therefore, if RTLS 50 is already available, the proposed technique can utilize animal data provided by an already installed RTLS 50. The following will combine... Figure 3 and Figure 4 RTLS 50 is described in more detail.

[0040] In alternative embodiments, tag 51 communicates directly with one or more of a plurality of tag readers (not shown) arranged in the livestock area 30. In these embodiments, tag 51 may be configured to record animal data, including activity data. In some embodiments, the location of animal 10 is determined based on relative distance to one or more of the tag readers. For example, the location may be determined based on the ability to wirelessly communicate with one or more access points using near-field communication. The animal data is then transmitted from tag 51 to one or more of the tag readers.

[0041] Now refer to Figure 2 Flowcharts and Figure 1 The livestock section 30 describes the proposed technology in more detail. Figure 2 This is a flowchart of an exemplary method for investigating the feeding behavior of animals 10 in livestock area 30. Figure 2 The method shown is performed, for example, by control device 100. Control device 100 is, for example, a livestock management system 20 (…). Figure 3 ) control equipment 100.

[0042] The method can be implemented as a computer program that includes instructions that, when executed by a computer (e.g., a processor in a control device), cause the computer to perform the method. According to some embodiments, the computer program is stored in a computer-readable medium (e.g., a memory or optical disc) that includes instructions that, when executed by a computer, cause the computer to perform the method.

[0043] The proposed method is based on the idea of ​​using feeding criteria that define normal feeding behavior to assess animal location and activity. This assessment reveals whether feeding behavior is normal or whether there is a risk of malnutrition. The criteria include, for example, a set of conditions or algorithms used to determine whether the location and activity of animal 10 in livestock enclosure 30 correspond to normal feeding behavior. Normal feeding behavior is defined as behavior that generally maintains good physical condition and milk production in animals of the same species. Physical condition can be assessed, for example, by a Body Condition Score (BCS), a means designed to assess an animal's body reserves or fat accumulation. Default normal feeding behavior is typically determined by studying a large number of animals in similar environments. However, because animals are individuals, what is considered "normal" may differ and change over time. Therefore, feeding criteria can be adjusted or adapted to suit a specific livestock enclosure 30.

[0044] This method is typically performed continuously while the animals reside in livestock area 30. The method is based on investigating animal feeding behavior by evaluating animal data using feeding criteria. In some embodiments, the method includes obtaining S1 one or more feeding criteria that define normal feeding behavior of the animal with respect to its activity and / or its position relative to the feeding station. Obtaining S1 herein refers to receiving or retrieving feeding criteria. For example, receiving feeding criteria from an external server. Alternatively, via user equipment 40 (… Figure 4 The user interface configuration and / or retrieval of feeding standards by reading default standards from memory.

[0045] Initial feeding standards can be pre-configured during manufacturing or programmed during the installation of the control device 100 configured to perform the method. Feeding standards define one or more conditions that must be met for feeding behavior to be considered "normal." Normal feeding behavior corresponds to feeding behavior that maintains the animal's good physical condition and well-being, and is defined by professionals who study the behavior of healthy animals. What is considered normal, of course, varies between different herds. Therefore, initial or pre-configured feeding standards typically define the normal condition for a general animal population. If feeding standards are configured by a farmer via a user interface, they can be based on the farmer's understanding of the normal condition for a particular herd. Feeding standards can, for example, define normal ranges or minimum / maximum values ​​for various parameters associated with feeding, such as location, movement, etc.

[0046] For example, the feeding criteria obtained to define normal feeding behavior define the number of visits to the feeding area, the time spent in the feeding area, the time between entering the feeding area 33 and approaching the feed station, the movement within the feeding area 33, the time spent at rest in the feeding area 33, or the occupancy rate in the feeding area 33 or a portion thereof. Feeding criteria will be described in more detail below with examples. Feeding criteria can be static or dynamic. In other words, in some implementations, they can be updated over time to better match the “normal” condition of a particular herd.

[0047] The animals are then monitored to investigate whether their feeding behavior is within normal limits. This is done using, for example, RTLS ( Figure 4This is accomplished using animal data collected from tags obtained by the animal 10. In other words, the method includes monitoring animal data (S2) for a period of time after feed is dispensed at the feeding station in the livestock area 30, the animal data being collected using tags carried by the animal 10. Monitoring (S1) herein refers to obtaining animal data collected at multiple different time points. For example, animal data is provided in real time by RTLS. Alternatively, animal data is received from time to time. For example, animal data is stored in tags and retrieved from the tags when an animal approaches a tag reader. Tag readers are typically available around the feeding station 31 and / or in the feeding area 33.

[0048] Monitoring animal data after feed distribution typically refers to doing so immediately or shortly after feed distribution, while the feed is still available at the feeding station. In some implementations, animal data is monitored repeatedly at regular intervals (such as every 1-2 seconds) because new feed is usually distributed when most or part of the feed has already been consumed.

[0049] Animal data indicates animal activity and / or the animal's position relative to feed station 31. For example, animal data indicates a real-time position given by coordinates in a reference coordinate system, such as livestock area 30. RTLS 50 is capable of locating animal 10 with an accuracy of less than one meter (i.e., more precise). These positions may also be less accurate. In some embodiments, the position is represented by the identity of the nearest tag reader to animal 10. Activities indicated by animal data include, for example, animal movement, animal posture, chewing activity, animal speed, etc.

[0050] Animal behavior can be assessed based on monitored animal data indicating animal activity and / or the animal's position relative to the feeding station 31. For example, the animal's position and activity can be analyzed over time to determine how many times (and / or for how long) the animal has eaten. Additionally, the animal's posture can indicate whether it is able to eat (i.e., is likely or probably eating), such as whether its head is above the feed, for example, in a head lock. If an animal is stationary in the feeding area 33 (or within a distance of the feeding station 31 from which it can reach the feed 32) and is still performing activities associated with eating, it can be assumed that the animal is eating during that time period. Another example is that an animal that moves extensively throughout the feeding area 33 may be identified as being stressed while eating. In other words, in some embodiments, the method includes determining the feeding behavior of one or more animals in the S3 animals based on activity and / or position indicated by the monitored animal data. Various feeding behaviors can be determined based on animal data, such as the time of appearance at the feeding station, the time pattern of the time of appearance at the feeding station, the actual feeding time, the time pattern of the actual feeding time, the feeding or chewing rate, the stress level during feeding, the movement during feeding, the posture during feeding, and the position at the feeding station.

[0051] The animal data can then be analyzed directly using one or more predetermined criteria to determine whether the animal data corresponds to normal feeding behavior. For example, the location of each animal can be tracked to reveal the number of visits to feeding area 33. The number of visits indicates the animal's intention to feed. If an animal visits too infrequently, it may indicate poor health, herd-related problems, or some other reason. Lower-ranking animals may be afraid to enter feeding area 33. In other words, in some implementations, the method includes using one or more feeding criteria that define normal feeding behavior to assess S4a the animal location and animal activity indicated by the monitored animal data.

[0052] Alternatively, if feeding behavior has been determined in S3, predetermined criteria are used to analyze the determined feeding behavior. In other words, in these embodiments, the criteria do not define what is normal in terms of the animal's location or activity, but rather what is normal in terms of certain feeding behaviors determined in S3 based on location and activity. In other words, in some embodiments, the method includes using one or more feeding criteria that define normal feeding behavior to evaluate the feeding behavior determined in S4b. For example, when animal 10 has entered feeding area 33, the animal should reach the feeding station within a short period of time and remain there, feeding for a duration of up to 30 minutes. In some embodiments, one or more criteria define an acceptable ratio between the actual feeding time and the time spent in feeding area 33. The determined feeding behavior is then compared to this ratio to assess whether the individual animal's feeding behavior is normal. In other words, in some embodiments, one or more feeding criteria define an acceptable time between entering feeding area 33 and approaching the feeding station.

[0053] Feeding standards can specify normal feeding behavior for each individual animal, group, or entire herd. For example, the normal time an individual needs to be present at feed station 31 will naturally depend on how much feed they require. Feed intake can also be defined for different groups defined by parameters such as weight, lactation period, and age. In other words, in some implementations, one or more feeding standards include individual feeding standards for individual animals, feeding standards effective for a subset of animals, and / or common feeding standards effective for all animals.

[0054] An example of behavior that does not meet feeding criteria is that animal 10 eats too little compared to a predefined (potentially animal-specific or group-specific) feed intake. The actual amount to be compared to the normal amount can be revealed by the number of visits to feeding area 33 and / or the time spent in feeding area 33. In other words, in some embodiments, one or more feeding criteria defining normal feeding behavior define the number of visits to feeding area 33, and in some embodiments, one or more feeding criteria defining normal feeding behavior define the time spent in feeding area 33.

[0055] If an animal enters feeding area 33 but does not proceed to feed station 31 or is driven away from feed station by other animals, this will also be monitored and assessed through feeding criteria, as it may result in reduced feed intake. In other words, in some implementations, one or more feeding criteria that define normal feeding behavior define displacement within feeding area 33.

[0056] Animal behavior in feeding area 33 is also generally relevant here. Typically, animals will enter feeding area 33, move to feeding station 31, and remain stationary while feeding. Extensive movement may indicate the presence of queues, an animal fighting with another animal, or stress. This, in turn, may affect feeding and result in lower feed intake. In other words, in some embodiments, one or more feeding criteria that define normal feeding behavior define the time spent at rest in feeding area 33.

[0057] If assessments S4a and S4b reveal that animal data does not meet feeding criteria, there are usually one or more underlying factors. Further analysis of the animal data can be conducted to identify such factors. For example, if many animals 10 are eating too little, there is usually a general problem related to feed, feed access, or environmental issues. However, if only one or a few individuals exhibit deviated feeding behavior, the cause is usually individual-related. In other words, in some implementations, the method includes analyzing S5, the location and animal activity indicated by the monitored animal data, to identify potential factors contributing to the failure to meet feeding criteria when the monitored animal data fails to meet one or more feeding criteria.

[0058] For example, potential factors can be identified by analyzing the location of the animals. For instance, if there are too many animals in feeding area 33, overcrowding can cause stress and lead to changes in feeding behavior. Overcrowding can be easily detected by counting the animals per unit area and comparing that count to a reference value. Alternatively, based on the location of animal 10 relative to other animals, it is possible to determine whether there are more animals than the feeding space allows, which can be used as an indicator of overcrowding. In other words, in some embodiments, analyzing potential factors S5 includes detecting locations and / or activities indicating overcrowding in or within livestock area 30.

[0059] Alternatively, abnormal feeding behavior may be caused by one or more aggressive animals. Analysis of animal data can reveal correlations between the presence (or proximity) of certain individual animals. If abnormal feeding behavior frequently or always occurs around a particular animal, a potential factor may be that the particular animal is aggressive. In other words, in some embodiments, analyzing potential factors S5 includes detecting the location and / or activity indicative of aggressive behavior in one or more animals. Aggression and overcrowding can be analyzed together, as aggressive behavior at the feeding station may be caused by overcrowding.

[0060] Another cause of abnormal feeding behavior may be displacement around the feeding station 31. For example, many animals may attempt to feed in the same spot. In other words, in some embodiments, analyzing potential factors in S5 includes detecting the location and / or activity that indicates animal displacement. Displacement of individual animals may also be caused by specific animals, as some animals tend to avoid aggressive animals. Therefore, aggression and displacement can be analyzed together, since aggressive behavior at the feeding station can cause displacement.

[0061] Therefore, aggressive behavior at the feeding station can be caused by overcrowding. Aggression can also lead to displacement. Aggression can also be caused by animal hierarchies, with some animals being aggressive towards lower-ranking animals. Based on aggressive interactions at the feeding station, animals can be stratified, and dominant aggressive animals and lower-ranking animals can be identified. In conclusion, by analyzing the combined effects of overcrowding, aggressive behavior, and displacement, the root causes of the problem can be identified and mitigated.

[0062] If an animal's feeding behavior deviates from normal feeding behavior as defined by feeding criteria, remedial action should be taken. In other words, the method includes performing action S6 when monitored animal data fails to meet one or more feeding criteria that define normal feeding behavior for animal 10 regarding animal activity and / or the animal's position relative to feed station 31. For example, if some animals do not enter feeding area 33 frequently enough each day, an alert is issued to the farmer to check the health and condition of these animals 10. If a livestock management system is implemented, farmer interaction is not always required. The first attempt to address the problem can be made, for example, by triggering a robot to perform an action.

[0063] Therefore, abnormal feeding behavior is mitigated in different ways depending on its underlying cause. Measures can be implemented manually by the farmer, triggered automatically by an automated livestock management system, or a combination of both. In some implementations, performing the S6 action includes providing the user with S6a information about the failure to meet one or more feeding criteria. This information may be a message displayed on the user's device (…). Figure 4 Alternatively or additionally, the information may be audible or presented in any suitable manner.

[0064] This information typically includes information about the existence of a feeding problem and may also include information about what the problem is, such as low feed intake or stress. The information may also include indications of potential factors identified in S5 based on animal data. For example, the message might say, “Low feed intake in area X” and “This may be causing overcrowding in area X and underfeeding in area Y.” In other words, in some implementations, the information provided indicates one or more potential factors that have been identified in failure to meet one or more feeding criteria.

[0065] The information provided to the user may also include intervention recommendations. For example, the information may suggest removing overly dominant or underperforming animals from their respective feeding areas. Additionally, an overcrowding alarm may be issued for a specific portion or the entire feeding area 33. If only a portion of the feeding area 33 is overcrowded, the root cause may be uneven feed distribution or uneven animal flow. In other words, in some embodiments, the information provided includes instructions for the user to perform activities. Exemplary activities that may be proposed include: regrouping animals into other groups, separating a pair or group of animals that affect each other's feeding behavior, instructing for health checks, and performing livestock management.

[0066] If an automated livestock management system is used, some of the proposed activities can be performed automatically without any user intervention. For example, feed distribution and / or pushing or regrouping can be performed automatically via autonomous feeding robots or automated sorting gates. In other words, in some implementations, performing action S6 includes automatically performing activities associated with animals in livestock area 30 as described in S6b.

[0067] The feeding robot 1 typically operates according to a work plan that includes a schedule and a trajectory. The trajectory defines the route where the feeding robot 1 should travel and the corresponding speed, i.e., speed and direction of travel. The schedule defines when the feeding robot 1 should begin working along the trajectory. In other embodiments, the work plan defines the timing of sessions for the feeding robot 1. More specifically, a session herein refers to a work session that typically includes one round or trip within the livestock area 30. When the feeding robot 1 is not working, i.e., between work sessions, it typically recharges at a docking station.

[0068] The activities of S6b performed automatically by the feeding robot can be performed in different ways. In some embodiments, automatically performing the S6b activities includes triggering the feeding robot 1 to start and / or stop operation. In other words, the schedule or time for operating the feeding robot 1 is determined based on deviations in feeding behavior. This can be achieved by configuring the operating schedule or plan of the feeding robot based on deviations in feeding behavior. For example, if feed is dispensed more frequently, the animal may be induced to eat more frequently. In other words, in some embodiments, the action includes generating a trigger that causes the feeding robot to start or stop operation.

[0069] In some implementations, automatically performing the S6b activity includes controlling how or where the feeding robot 1 operates. More specifically, the method includes adjusting the trajectory of the feeding robot 1 as it operates within the livestock area. In this way, it is possible to control where the feeding robot 1 operates and its speed based on needs indicated by the animals' feeding behavior. Additionally, the activities of the feeding robot can be controlled, such as what it does and where it does it. For example, the feeding robot 1 can be controlled to have different operating modes, such as pushing feed and delivering (or simply moving) feed. Different operating modes can be activated in different locations based on needs indicated by feeding behavior. In other words, in some implementations, automatically performing the S6b activity includes adjusting the trajectory, speed, and / or operating mode of the feeding robot.

[0070] In some implementations, automating the S6b activity includes controlling a sorting gate. For example, if a sorting gate is arranged in the post-milking passage, the gate can be controlled to pick out aggressive animals or separate fighting animal pairs.

[0071] Feeding criteria can be static or dynamic. Dynamic criteria can be updated based on historical animal data. For example, if animal condition (represented by BCS) and milk production in a particular herd remain acceptable even when deviating from a certain normal average across many herds, then feeding criteria can be updated for that specific herd. In other words, as long as the animals feel well, the general goal is to maintain a certain feeding behavior. In other words, in some implementations, the method includes adjusting one or more feeding criteria (S7) based on location and activity indicated by monitored animal data.

[0072] If a deviation from feeding behavior is detected and good condition is verified, for example, through a health check (maintained BCS), then the feeding behavior can be considered normal for that particular herd. In other words, in some implementations, adjustment S7 is performed during the period of animal data monitoring when the deviation of milk yield S7a or body condition score S7b from a predefined reference value remains within a predefined tolerance level.

[0073] Figure 3 A control device 100 for investigating the feeding behavior of animals 10 in livestock enclosure 30, according to the second aspect, is shown in more detail. The control device 100 communicates with a real-time location system RTLS 50. In some embodiments, the control device 100 communicates with a user device 40 and / or a livestock management system 20.

[0074] User equipment 40 is a device that enables communication with a user (such as a farmer), such as a monitor, computer, tablet smartphone, etc. More specifically, user equipment 40 is configured to, for example, provide information to the user on a display. User equipment 40 may also include a speaker or other means for communicating with the user.

[0075] The livestock management system 20 is a system that helps farmers record and track their livestock. It can capture all events of the animals and track the most important dates in their lives. The livestock management system 20 can also be configured, for example, to automatically control a feeding robot 1 ( Figure 1 (and / or sorting gates (not shown)) to automatically perform livestock management.

[0076] In some embodiments, the control device 100 is a functional unit. Therefore, the control device 100 may be distributed among multiple physical control units, some of which may be located within the livestock area 30 and some of which may be located remotely from the livestock area 30. In some embodiments, the control device 100 is at least partially implemented in the feeding robot 1. In some embodiments, the control device 100 is included in a livestock management system.

[0077] Control device 100 includes hardware and software. Hardware includes, for example, various electronic components located on, for example, a printed circuit board (PCB). The most important of these components are typically a processor 101, such as a microprocessor, and a memory 102, such as an EPROM or flash memory chip. Software is typically software code running in a microcontroller. The control device 100 also includes a communication interface 103. Communication interface 103 is configured to transmit signals and / or data between control device 100 and other devices, such as the feeding robot 1 and RTLS 50. Communication interface 103 is configured for wireless communication using any suitable protocol, such as Bluetooth or IEEE 802.11. Communication interface 103 can also be configured for wired communication, for example, via a docking station. Communication interface 103 is configured, for example, to communicate with the control system 52 of RTLS 50. Figure 4 The control device 100 communicates with the user equipment 40. Specifically, the control device 100 is configured to acquire the tag 51 carried by the animal 10. Figure 4 Animal data collected. In some embodiments, communication interface 103 is configured to communicate with a server located remotely or locally.

[0078] More specifically, the control device 100 is configured to: monitor animal data for a period of time after feed is dispensed at the feeding station in the livestock area 30, the animal data being collected using tags carried by the animals 10 and indicating the animals' activity and / or their position relative to the feeding station (31); and to perform actions when the monitored animal data fails to meet one or more feeding criteria, the one or more feeding criteria defining normal feeding behavior of the animals (10) with respect to the animals' activity and / or their position relative to the feeding station (31). Additionally, the control device 100 may be configured to perform combined... Figure 2 One or more of the implementation schemes of the described method.

[0079] Figure 4 An example of an RTLS 50 that can be used by the proposed method and control device 100 is shown. The RTLS also includes a reader 54 that receives wireless signals from the tags 51 to determine their location. Wireless communication includes, but is not limited to, cellular radio, WiFi radio, Bluetooth radio, Bluetooth Low Energy radio, ultra-wideband radio, or any other suitable radio frequency communication protocol. The specific number and placement of the readers 54 will depend on the size and shape of the monitored tracking area 53 (e.g., livestock area 30).

[0080] In some embodiments, tag 51 also includes an orientation sensor, such as a triaxial accelerometer assembly or a gyroscope assembly, configured to generate data indicating the orientation of the sensor. Tag 51 may also include other sensors or components, such as object monitoring sensors. Object monitoring sensors may include a thermometer, heart rate monitor, vibration sensor, camera, microphone, or any other suitable device.

[0081] When using RTLS 50, the location of each tag 51 is tracked in real time within the tracking area 53 using multiple positioning techniques known in the art, such as time difference of arrival and received signal strength indication techniques. For this purpose, data from readers 54 is provided to a control system 52, which determines the instantaneous position of each tag 51 in the tracking area 53 in real time. The control system 52 can be implemented as a computer-based system capable of executing computer applications. An exemplary application of the control system 52 includes a real-time positioning function configured to determine the two-dimensional or three-dimensional position of the tag 51 within the tracking area 53. The control system 52 can determine the location of the tag 51, for example, using triangulation based on data provided by three or more readers 54.

[0082] In some embodiments, the control system 52 is configured to determine the movement of the tag 51, including, for example, the direction and amount of movement. In some embodiments, the control system 52 is configured to determine the orientation of the tag 51. In some embodiments, the control system 52 is configured to distinguish different activities of the animal 10 wearing the tag 51 based on the location, movement, and orientation of the animal tag within the tracking area 53. For example, the control system is configured to detect feeding and / or feeding-related activities, such as chewing activities associated with feeding.

[0083] The control system 52 may also have one or more communication interfaces. Communication interfaces may include, for example, modems and / or network interface cards. The communication interfaces enable the control system 52 to communicate with other computing devices (such as…) Figure 3 The control device 52 sends data to and receives data from other computing devices. The communication interface also enables the control system 52 to receive messages and data directly from the reader 54 or from the tag 51, or via another communication network. The communication network can be any network platform and can include multiple network platforms. Exemplary network platforms include, but are not limited to, WiFi networks, cellular networks, etc.

[0084] The terminology used in the description of the embodiments illustrated in the accompanying drawings is not intended to limit the described methods, control devices, or computer programs. Various changes, substitutions, and / or modifications may be made without departing from the embodiments of this disclosure as defined in the appended claims.

[0085] As used herein, the term “or” should be interpreted as mathematical OR, i.e., as inclusive disjunction, rather than as mathematical exclusive OR (XOR), unless otherwise expressly stated. Additionally, the singular forms “a,” “an,” and “the” should be interpreted as “at least one,” and thus may include multiple entities of the same kind, unless otherwise expressly stated. It should be further understood that the terms “comprising,” “including,” specify the presence of the stated features, actions, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, actions, integrals, steps, operations, elements, components, and / or groups thereof. For example, a single unit of a processor may perform the functions of several items recited in the claims.

Claims

1. A method for monitoring feeding behavior of animals in a tracking area of ​​a livestock enclosure (30), the livestock enclosure (30) comprising a feeding platform (31) and a feeding area (33) adjacent to the feeding platform (31), the method comprising: - Animal data is monitored for a period of time after feed is distributed at the feeding station in the livestock area using a real-time positioning system. The animal data is collected using tags (51) carried by the animals (10). The real-time positioning system includes a reader (54) that receives wireless signals from the tags (51) to determine the instantaneous position of each tag (51) in the tracking area (53) in real time. The tags are attached to the head or neck of the animals and include motion sensors or accelerometers. The animal data indicates the feeding activities of the animals and the position of the animals relative to the feeding station (31). - The feeding behavior of one or more of the animals is determined based on the feeding activities and locations indicated by the monitored animal data, wherein the feeding behavior includes the actual feeding time; - The identified feeding behavior is assessed using one or more feeding criteria that define normal feeding behavior, wherein the one or more feeding criteria define the ratio between the actual feeding time and the time spent in the feeding area, and the identified feeding behavior is compared with the ratio to assess whether the individual animal’s feeding behavior is normal. as well as - When monitored animal data fails to meet one or more feeding criteria, the one or more feeding criteria defining the normal feeding behavior of the animal (10) with respect to the animal's feeding activities and the animal's position relative to the feed station (31), wherein the action includes providing the user with information about the failure to meet one or more feeding criteria, wherein the action includes automatically performing activities associated with the animal in the livestock area, wherein the activities include one or more of the following: regrouping the animal into another group, separating a pair of animals or a group of animals that affect each other's feeding behavior, and instructing a health check.

2. The method of claim 1, wherein the one or more feeding criteria include individual feeding criteria for individual animals, feeding criteria effective for a subset of the animals, and / or common feeding criteria effective for all animals.

3. The method of claim 1, wherein the determined feeding behavior further includes one or more of the following: the time of appearance at the feeding station, the rate of eating or chewing, the pressure level during eating, the movement during eating, the posture during eating, and the position at the feeding station.

4. The method according to claim 1, wherein the method comprises: When the monitored animal data fails to meet one or more of the feeding criteria, - Analyze the location and animal activity indicated by the monitored animal data to determine potential factors that may have contributed to the failure of the animal data to meet feeding standards.

5. The method according to claim 4, wherein the analysis of potential factors includes: Detect location and / or activity that indicates one or more of overcrowding, aggressive behavior in animals, and animal displacement.

6. The method of claim 5, wherein the information provided indicates one or more potential factors identified by the one or more eating criteria that have not been met, wherein the information provided includes instructions to perform activities.

7. The method according to claim 1, wherein the method comprises: - Adjust one or more feeding criteria based on the location and activity indicated by the monitored animal data.

8. The method of claim 7, wherein the adjustment is performed during the period of animal data monitoring while the deviation of milk yield or physical condition score from a predefined reference value remains within a predefined tolerance level.

9. The method according to any one of the preceding claims, wherein, The feeding robot (1) operates along the feeding table according to a work plan including a schedule and trajectory, and performs actions including automatically adjusting the trajectory, speed and / or operating mode of the feeding robot.

10. A computer program comprising instructions that, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 9.

11. A control device for monitoring the feeding behavior of animals in a tracking area of ​​a livestock enclosure (30), the livestock enclosure (30) including a feeding platform (31) and a feeding area (33) adjacent to the feeding platform (31), wherein the control device (100) is configured to: - Animal data is monitored for a period of time after feed is distributed at the feeding station in the livestock area using a real-time positioning system. This animal data is collected using tags (51) carried by the animals (10). The real-time positioning system includes a reader (54) that receives wireless signals from these tags (51) to determine the instantaneous position of each tag (51) in the tracking area (53) in real time. The tags are attached to the animal's head or neck and include motion sensors or accelerometers. The animal data indicates the animal's feeding activity and its position relative to the feeding station (31). - The feeding behavior of one or more of the animals is determined based on the feeding activities and locations indicated by the monitored animal data, wherein the feeding behavior includes the actual feeding time; and - The identified feeding behavior is assessed using one or more feeding criteria that define normal feeding behavior, wherein the one or more feeding criteria define the ratio between actual feeding time and time spent in the feeding area, and the identified feeding behavior is compared to the ratio to assess whether the individual animal's feeding behavior is normal. - When monitored animal data fails to meet one or more feeding criteria, the one or more feeding criteria defining the normal feeding behavior of the animal (10) with respect to the animal's feeding activities and the animal's position relative to the feed station (31), wherein the action includes providing the user with information about the failure to meet the one or more feeding criteria, wherein the action includes automatically performing activities associated with the animal in the livestock area, wherein the activities include one or more of the following: regrouping the animal into another group, separating a pair of animals or a group of animals that affect each other's feeding behavior, and instructing a health check.

12. The control device according to claim 11, wherein the control device (100) is configured to perform the method according to any one of claims 2 to 9.

Images