WO2023031759A1

WO2023031759A1 - Animal husbandry system

Info

Publication number: WO2023031759A1
Application number: PCT/IB2022/058061
Authority: WO
Inventors: Paulus Jacobus Maria Van Adrichem; Yan Li; Sonu Ranjit JACOB; Karel Van Den Berg; Adrianus Cornelis Maria MEEUWESEN
Original assignee: Lely Patent N.V.
Priority date: 2021-09-02
Filing date: 2022-08-29
Publication date: 2023-03-09

Abstract

The present invention relates to an animal husbandry system, wherein a group of animals (2) can move about freely in an area (1), the system comprising monitoring and analyzing means (11) suitable for determining and tracking the identity and the position of each individual animal (2), wherein lameness detection means operatively connected to the monitoring and analyzing means (11) are provided, suitable for the early detection of lameness in an animal (2), the lameness detection means being arranged to take into account at least one of the following parameters measured with the aid of the monitoring and analyzing means (11): - lying down time of the animal (2) in a certain time interval; - distance covered by the animal (2) in a certain time interval; - speed of movement of the animal (2) in a certain time interval (top speed and/or average speed and/or speed stability); - time spent by the animal (2) in zones of the area having a particular animal density characteristic. The use of monitoring and analyzing means (11) enables an automatic and accurate early lameness detection in animals (2) moving about freely in an area (1). The early detection of lameness in animals (2) can prevent serious health complications for the animals (2) in question.

Description

Animal husbandry system

The present invention relates to an animal husbandry system.

Such systems are widely known in the art. Monitoring the animals in the system and gathering relevant animal data plays an important role in such systems.

WO201 4/118788 discloses an optical monitoring system for livestock in which various activities and parameters may be monitored and measured to determine the health state of the population of animals, in particular chickens. WO2015/104250 divulges a method and system of livestock rearing in which the illumination in a shed is optimized. WO2011/039112 shows an animal behavior monitoring system in which several animal movement patterns are derived from recorded image data. WO2018/174812 describes a video monitoring system using thermal imagers for individually identifying animals carrying an identification tag. None of these documents addresses the detection of lameness in animals.

WO201 2/078054 is concerned with an apparatus for monitoring a milking animal during milking using thermographic cameras to detect certain conditions, such as lameness. WO2017/030448 relates to a method and apparatus for automatically evaluating an animal, in which lameness is detected by analysing the shape of the spine of a cow using a 3D camera. EP3335551 discloses a method of monitoring livestock inside of a building using a camera arranged above the livestock for gathering tracking data relating to a motion of a tracked individual animal. The motion data thus obtained can be compared to stored motion data or motion patterns related to or associated with e.g. lameness of an animal. The comparison with stored reference data or motion patterns which extend over a period of time renders the system rather inert. There is no disclosure of any concrete, practical motion data or pattern to be used for detecting lameness.

There is a need for an improved system with a speedy, yet accurate lameness detection.

It is an object of the present invention to provide such an improved system. The invention achieves the object at least in part by means of a system according to claim 1 , in particular an animal husbandry system, wherein a group of animals can move about freely in an area, the system comprising monitoring and analyzing means suitable for determining and tracking the identity and the position of each individual animal, wherein lameness detection means operatively connected to the monitoring and analyzing means are provided, suitable for the early detection of lameness in an animal, the lameness detection means being arranged to take into account at least one of the following parameters measured with the aid of the monitoring and analyzing means:

- lying down time of the animal in a certain time interval;

- distance covered by the animal in a certain time interval;

- speed of movement of the animal in a certain time interval (top speed and/or average speed and/or speed stability);

- time spent by the animal in zones of the area having a particular animal density characteristic.

In this way, an animal-friendly and efficient system for the early detection of lameness is realized. Applicant has performed extensive research on visually detectable lameness indicators. The invention is based on the insight that the use of monitoring and analyzing means enables an automatic and accurate early lameness detection in animals moving about freely in an area. Moreover, this can be achieved in a very animal-friendly way, without the use of additional equipment carried by or attached to the animals. The early detection of lameness in animals can prevent serious health complications for the animals in question.

Suitable and advantageous embodiments are described in the dependent claims, as well as in the description below.

According to a first embodiment of the invention, the lameness detection means comprise a processor or a computer processing unit. This constitutes a simple, yet effective implementation.

According to a further embodiment, wherein the area has a number of activity areas connected by an alley, the activity areas comprise a resting area, a feeding and a drinking area, and in case the animals are dairy animals, a milking area. These areas constitute essential animal activity areas.

Advantageously, additional identification means are provided for identifying individual animals in the area. In this way, the animal identification and tracking accuracy of the system is enhanced.

In a further embodiment, the area has a resting area with cubicles, the lameness detection means being arranged to take into account the lying down time of the animal in a cubicle in a certain time interval measured with the aid of the monitoring and analyzing means. This lying down time has turned out to be highly indicative of an early stage of lameness.

In yet another embodiment, the lameness detection means are arranged to take into account the distance the animal moves in the alley between the activity areas in

2 a certain time interval measured with the aid of the monitoring and analyzing means. This parameter, one the one hand, is relatively easy to measure and, on the other hand, has turned out to be also a good indicator of an early stage of lameness.

Advantageously, the lameness detection means are arranged to take into account the number of times the animal moves at a maximum speed higher than a threshold speed in a certain time interval measured with the aid of the monitoring and analyzing means. This parameter has also turned out to be a good indicator of an early stage of lameness.

According to a further embodiment, in the area crowded zones are determined, which contain a number of animals exceeding a minimum number within a particular perimeter, the lameness detection means being arranged to take into account the time the animal spends in a crowded zone in a certain time interval measured with the aid of the monitoring and analyzing means. From extensive research it has been found that this is also a good indicator of an early stage of lameness.

In yet another embodiment, the lameness detection means are arranged to compare the parameters measured with the aid of the monitoring and analyzing means with a threshold which depends on historical and/or measured data relating to lame and to healthy animals. This has proven to give excellent results.

Advantageously, the threshold is an average herd value. In this way, relevant measurements are taken into account.

In accordance with a further embodiment, the lameness detection means are arranged to generate an attention signal for a specific animal in case the animal is found to be or to become lame. This is highly useful for the farmer. Moreover, animal health and welfare can be improved.

In yet a further embodiment, the system further comprises a treatment area and a selection gate for guiding an animal to the treatment area, the lameness detection means being arranged to automatically activate the selection gate in order to guide an animal to the treatment area in case the animal is found to be or to become lame. This constitutes a very user-friendly embodiment.

Advantageously, the monitoring and analyzing means are suitable for determining and tracking the body orientation of each individual animal. This is helpful in making the tracking process even more reliable.

Finally, the lameness detection means are also arranged to take into account the number of steps and/or step size (biggest and/or average) of the animal in a certain time interval measured with the aid of the monitoring and analyzing means. In this

3 way, the reliability of the lameness detection can be even further improved.

The invention will now be further explained with reference to the following Figure, which schematically illustrates the system according to the invention.

Figure 1 shows an area with an animal husbandry system according to the invention.

Figure 1 shows an animal husbandry system with an area here in the form of a stable or shed or barn 1 wherein a group of animals 2 can move about freely. In the example shown the animals are cows. Of course, the invention can also be applied to an animal husbandry system involving other animals, such as goats, pigs, horses, or even chickens, turkeys, etcetera. The stable 1 can also be a structure with, for example, a partly open roof. The system according to the invention can also be applied in a complete free stall barn or in an open area like a meadow or pasture with cattle fences.

In the example shown, the stable 1 has a number of activity areas for the animals 2: a feeding area 3, a drinking area 4, at least one resting area 5 with cubicles 6, a milking area 7 with a milking robot 8, and a treatment area 9. All these areas are connected by an alley 10.

A number of monitoring and analyzing means suitable for determining and tracking the identity and the position of each individual animal 2 is provided, here in the form of cameras 11 .

The cameras 11 are arranged for monitoring a region or a subarea or a number of subareas of the stable 1 . Of course, other (optical) sensors can in principle be used instead of the cameras 11 , or additionally. The cameras 11 are provided in such a way that, collectively, they are suitable for monitoring substantially the complete shed 1 , particularly all the animals 2. The cameras 11 can be provided in the shed 1 above the animals 2 (e.g. centrally, optionally only one camera can be used), or as shown in Figure 1 , on the walls of the shed 1. They are positioned and oriented in such a way that all relevant subareas of the area 1 can be monitored. The subareas may overlap. In case the invention is used in a pasture, the cameras 11 may be provided on fences, at a sufficient height above the ground. The chosen configuration obviously depends on the shape of the area or shed 1.

Suitable illumination means for the shed 1 (not shown) are also provided.

When keeping animals 2 in an area 1 , it is often desirable to monitor the animals 2 and to gather specific animal data. In accordance with the invention, data on lameness of an animal 2 can be efficiently and reliably gathered, although the animals 2 are allowed to move about freely in the area 1.

4 In accordance with the invention, lameness detection means operatively connected to the monitoring and analyzing means 11 are provided, suitable for the early detection of lameness in an animal 2, the lameness detection means being arranged to take into account at least one of the following parameters measured with the aid of the monitoring and analyzing means 11 :

- lying down time of the animal 2 in a certain time interval;

- distance covered by the animal 2 in a certain time interval;

- speed of movement of the animal 2 in a certain time interval (top speed and/or average speed and/or speed stability);

- time spent by the animal 2 in zones of the area 1 having a particular animal density characteristic.

The lameness detection means comprise a processor or a computer processing unit with processing means and storing means, operatively connected to the cameras 11. The lameness detection means can be regarded as a control unit (not shown), which can be located inside or outside or at a distance from the area or shed 1 . Video analyzing means, known as such, can be incorporated in the cameras 11 , or in the control unit I lameness detection means.

Thus, an animal-friendly and efficient system for the early detection of lameness is realized. Applicant has performed extensive research on visually detectable lameness indicators. The use of monitoring and analyzing means 11 enables an automatic and accurate early lameness detection in animals 2 moving about freely in an area 1. Moreover, this can be achieved in a very animal-friendly way, without the use of additional equipment carried by or attached to the animals 2. The early detection of lameness in animals can prevent serious health complications for the animals in question.

Additional identification means may be provided for identifying individual animals 2 in the area 1. In this way, the animal identification and tracking accuracy of the system is enhanced. This can be done in various ways, known as such. For example, Radio Frequency Identification with tags/transponders and an antenna with a reader can be used. Alternatively, image analysis can be used with the aid of the video monitoring and analyzing means 11 , e.g. recognizing the spotted pattern on the back of cows 2. Images of all cameras 11 can be combined to track each cow 2 in the whole barn 1. A further possibility involves resetting the position of an animal 2 every time it visits a certain known location, for example when a cow 2 visits the milking robot 8 in the milking area 7, and then tracking the animal 2 from that known location using the monitoring means 11. Knowing the exact location of the animals 2 in the area 1 has a lot of advantages:

5 individual animal data can be gathered.

The area 1 has at least one, or, as shown in the example in Figure 1 , several resting areas 5 with cubicles 6. The lameness detection means are arranged to take into account the lying down time of the animal 2 in a cubicle 6 in a certain time interval measured with the aid of the monitoring and analyzing means 11. This lying down time has turned out to be highly indicative of an early stage of lameness. The more time a cow 2 spends in a cubicle 6, the more likely it is that she is lame. Lame cows 2 tend to lie down a lot.

The lameness detection means may also be arranged to take into account the distance the animal 2 moves in the alley 10 between the activity areas 3, 4, 5, 7, 9 in a certain time interval measured with the aid of the monitoring and analyzing means 11. This parameter is relatively easy to measure and has turned out to be also a good indicator of an early stage of lameness. The less distance a cow 2 moves in the alley 10 between the activity areas 3, 4, 5, 7, 9, the more likely it is that she is lame. Lame cows 2 tend to avoid unnecessary movements.

Also, the lameness detection means can be arranged to take into account the number of times the animal 2 moves at a maximum speed higher than a threshold speed in a certain time interval measured with the aid of the monitoring and analyzing means 11. For example, the number of times a cow 2 moves at a speed exceeding 1 meter per second is counted in a given time interval. This parameter has also turned out to be a good indicator of an early stage of lameness. The number of quick displacements is lower for a lame cow 2, when compared to a healthy cow 2, for obvious reasons.

With the aid of the cameras 11 , in the area or shed 1 crowded zones are determined, which contain a number of animals 2 exceeding a minimum number within a particular perimeter. For example, a certain subarea of the area 1 is called a crowded zone when at least 4 cows are standing together within a 3 meter radius. This is illustrated in Figure 1 on the right hand side, where four cows 2 stand closely together. The lameness detection means are arranged to take into account the time the animal 2 spends in such a crowded zone in a certain time interval measured with the aid of the monitoring and analyzing means 11. From extensive research it has been found that this is also a good indicator of an early stage of lameness. The less time a cow 2 spends in crowded areas, the more likely it is that she is lame. Lame cows 2 tend to avoid crowded areas.

The lameness detection means are arranged to compare the parameters measured with the aid of the monitoring and analyzing means 11 with a threshold which depends on historical and/or measured data relating to lame and to healthy animals. This

6 has proven to give excellent results in determining lameness. For example, the lying down time of the animal 2 in a cubicle 6 in a 24 hour time interval can be compared to a threshold value of 16 hours. Healthy cows normally lie down for about 14 hours per day. In case a cow 2 lies down for more than 16 hours per day, she is likely to be lame.

For other parameters suitable thresholds can be found either in literature or by experimenting (e.g. trial and error).

In order to determine a suitable threshold, use can also be made of artificial intelligence techniques (known as such), such as machine learning or deep learning, e.g. using neural networks.

The threshold can be chosen as an average herd value. This is advantageous in that relevant measurements are thus taken into account.

The lameness detection means are arranged to generate an attention signal for a specific animal 2 in case the animal is found to be or to become lame. This is highly useful for the farmer, who can take appropriate action. Animal health and welfare are thus improved.

Optionally, the system comprises a treatment area 9 and a selection gate 12 for guiding an animal 2 to the treatment area 9, the lameness detection means being arranged to automatically activate the selection gate 12 in order to guide an animal 2 to the treatment area 9 in case the animal is found to be or to become lame. For example, an animal 2 leaving the milking robot 8 in the milking area 7 can be guided directly into the treatment area 9, as soon as it has been determined by the lameness detection means that it is lame. This constitutes a very user-friendly embodiment.

The monitoring and analyzing means 11 can be suitable for determining and tracking the body orientation of each individual animal 2. This is helpful in making the tracking process even more reliable, as it aids in tracking the animal's movements more precisely.

The lameness detection means can also be arranged to take into account the number of steps and/or the step size (biggest and/or average) of the animal 2 in a certain time interval measured with the aid of the monitoring and analyzing means 11. In this way, the reliability of the lameness detection can be even further improved.

Thus, a very efficient, yet animal-friendly system is realized. The use of monitoring and analyzing means 11 enables an automated, accurate early detection of lameness in animals 2 moving about freely in the area 1.

7

Claims

8 CLAIMS

1. An animal husbandry system, wherein a group of animals (2) can move about freely in an area (1), the system comprising monitoring and analyzing means (11) suitable for determining and tracking the identity and the position of each individual animal (2), wherein lameness detection means operatively connected to the monitoring and analyzing means (11) are provided, suitable for the early detection of lameness in an animal (2), the lameness detection means being arranged to take into account at least one of the following parameters measured with the aid of the monitoring and analyzing means (11):

- lying down time of the animal (2) in a certain time interval;

- distance covered by the animal (2) in a certain time interval;

- speed of movement of the animal (2) in a certain time interval (top speed and/or average speed and/or speed stability);

- time spent by the animal (2) in zones of the area having a particular animal density characteristic.

2. System according to claim 1 , wherein the lameness detection means comprise a processor or a computer processing unit.

3. System according to any one of claims 1 - 2, wherein the area (1) has a number of activity areas (3, 4, 5, 7, 9) connected by an alley (10), the activity areas (3, 4, 5, 7, 9) comprising a resting area (5), a feeding (3) and a drinking area (4), and in case the animals (2) are dairy animals, a milking area (7).

4. System according to any one of claims 1 - 3, wherein additional identification means are provided for identifying individual animals (2) in the area (1).

5. System according to any one of claims 1 - 4, wherein the area (1) has a resting area (5) with cubicles (6), the lameness detection means being arranged to take into account the lying down time of the animal (2) in a cubicle (6) in a certain time interval measured with the aid of the monitoring and analyzing means (11).

6. System according to any one of claims 3 - 5, wherein the lameness detection means are arranged to take into account the distance the animal (2) moves in the alley (10) between the activity areas (3, 4, 5, 7, 9) in a certain time interval measured with the aid of the monitoring and analyzing means (11).

7. System according to any one of claims 1 - 6, wherein the lameness detection means are arranged to take into account the number of times the animal (2)

8 9 moves at a maximum speed higher than a threshold speed in a certain time interval measured with the aid of the monitoring and analyzing means (11).

8. System according to any one of claims 1 - 7, wherein in the area (1) crowded zones are determined, which contain a number of animals (2) exceeding a minimum number within a particular perimeter, the lameness detection means being arranged to take into account the time the animal (2) spends in a crowded zone in a certain time interval measured with the aid of the monitoring and analyzing means (11).

9. System according to any one of claims 1 - 8, wherein the lameness detection means are arranged to compare the parameters measured with the aid of the monitoring and analyzing means (11) with a threshold which depends on historical and/or measured data relating to lame and to healthy animals.

10. System according to claim 9, wherein the threshold is an average herd value.

11. System according to any one of claims 1 - 10, wherein the lameness detection means are arranged to generate an attention signal for a specific animal (2) in case the animal (2) is found to be or to become lame.

12. System according to any one of claims 1 - 11 , wherein the system further comprises a treatment area (9) and a selection gate (12) for guiding an animal (2) to the treatment area (9), the lameness detection means being arranged to automatically activate the selection gate (12) in order to guide an animal (2) to the treatment area (9) in case the animal (2) is found to be or to become lame.

13. System according to any one of claims 1 - 12, wherein the monitoring and analyzing means (11) are suitable for determining and tracking the body orientation of each individual animal (2).

14. System according to any one of claims 1 - 13, wherein the lameness detection means are also arranged to take into account the number of steps and/or the step size (biggest and/or average) of the animal (2) in a certain time interval measured with the aid of the monitoring and analyzing means (11).

9