Automatic milking system and method for determining the health status of an animal

Abstract

The automatic milking system includes an electronic control unit (1), a milking station (2), a product milk tank (3), a second milk tank (4), an animal identification system, a first sensor (7) configured to measure at least one indicator of animal illness, and a data memory (8) in which measurements performed by the first sensor (7) on milk from individual animals identified by the identification system are collected. The control unit (1) is configured to register a first time point (t1) at which milk from the animal is diverted to the second tank rather than being sent to the product milk tank (3), and to register a second time point (t2) at which milk from the animal transitions from being sent to the second tank to being sent to the product milk tank (3). If the period between the first time point (t1) and the second time point (t2) is within a predetermined number of days, the control unit (1) is configured to determine that the animal is in a sick state between the first time point (t1) and the second time point (t2).

Description

[Technical field]

[0001] The present invention relates to an automatic milking system comprising a control system including an electronic control unit, at least one milking station configured to milk an animal, a product milk tank for receiving milk from the milking station, a second milk tank configured to receive milk that is diverted from the milking station and not sent to the product milk tank, an animal identification system connected to the control unit and configured to provide the control unit with individual identification information of an animal to be milked in the automatic milking system, and at least one first sensor configured to measure at least one indicator of a disease of the animal.

[0002] The automated milking system may be a conventional system involving manual attachment of teat cups to the teats of the animal to be milked, or a robotic system in which a robot automatically attaches teat cups to the teats of the animal to be milked. One or more sensors for measuring disease indicators may be placed in the milk line from the milking station to the production tank. In the case of robotic systems in particular, a separate sensor may be provided in association with each teat cup.

[0003] The present invention also relates to a method for determining the health status of an animal milked in an automatic milking system. [Background technology]

[0004] In connection with the industrial milking of dairy cows on dairy farms, it frequently occurs that dairy cows suffer from mastitis, either clinical or subclinical. Early detection of animal diseases, especially mastitis, can be achieved by univariate or multivariate analysis of measurements of different milk characteristics. When signs of mastitis are detected, farmers often have to further investigate the milk or the animal in question to determine whether it is mastitis and whether it is clinical or subclinical mastitis. If the mastitis is clinical mastitis, antibiotics must be administered to the animal and the animal's milk must be diverted from the production milk for several days counting from the date of administration of the antibiotics.

[0005] After treating an animal with an antibiotic, it is of interest to know if the animal has recovered or not (chronic mastitis). A control system can be used to ascertain whether an individual animal in a herd has or has had a disease such as mastitis, especially clinical mastitis, and to determine the health status of the animal at a specific time after mastitis or other disease has been detected. It is the object of the present invention to propose such a control system. Summary of the Invention

[0006] The object of the present invention is to provide an automatic milking system, comprising: a control system including an electronic control unit; at least one milking station configured for milking an animal; a produced milk tank for receiving milk from the milking station; a second milk tank adapted to receive milk diverted from the milking station and not sent to the production milk tank; an animal identification system connected to the control unit and configured to provide the control unit with individual identification information of animals milked by the automatic milking system; at least one first sensor configured to measure at least one indicator of a disease in the animal, the first sensor being connected to a control unit; a data memory in which the measurements carried out by the first sensor on the milk from each individual animal identified by the identification system are collected, the time of each measurement also being registered and stored in the memory, the automatic milking system comprising a control unit, registering a first time point at which milk from an animal is not sent to a production milk tank but is diverted to a second tank, - configured to register a second time point at which milk from the animal is transitioned from being sent to the second tank to being sent to the production milk tank if the milk from the animal is diverted from the milking station to the second milk tank over a number of consecutive milking sessions after the first time point; -This is achieved by an automatic milking system, characterized in that if the period between the first and second time points is within a predetermined number of days, the control unit is configured to determine that the animal is in a sick state between the first and second time points.

[0007] The first time point may be triggered by the registration of the administration of an antibiotic to the animal. The administration may be determined and performed by a human or determined by an electronic control unit and performed either by a human or by a device for automatic administration of antibiotics to the animal, the device being controlled by the electronic control device. The identification system usually includes a tag placed on each animal and a tag reader placed at the milking station and connected to the electronic control unit.

[0008] The second tank should be interpreted broadly and may be a tank directly connected to the milking station, or a separate bucket, or a drain line for diverted milk.

[0009] According to one embodiment, the time period between the first and second time points is between 2 and 10 days. According to another embodiment, the time period is between 5 and 10 days. A shorter time period may indicate a lower severity, and a longer time period may indicate a higher severity. The control unit may be configured to determine different disease states depending on the length of the time period.

[0010] According to one embodiment, the control unit is configured to determine the animal's health state based on repeated measurements by the first sensor from said second time point to a predefined third time point. Preferably, the measurements are performed for each separate milking session. The results are collected in a data memory.

[0011] According to one embodiment, the control unit is configured to determine a status of recovery of the animal from the disease condition at a third time point.

[0012] According to one embodiment, the control unit is configured to determine a state of recovery from the disease state when the measurements from the first sensor stabilize within a predefined range between the second and third time points. If the somatic cell count is determined by the first sensor, the upper limit of the range may be 25% higher than the "normal" somatic cell count of the animal before the signs of disease are established. The "normal" somatic cell count is the cell count given by one or more previous historical measurements of the animal's somatic cell count stored in the data memory. The inventors have found that the third time point, at which stabilization should occur as an indication of recovery, is 3-4 weeks from the first time point.

[0013] According to one embodiment, the first sensor is a sensor configured to measure the electrical conductivity of the milk or a sensor configured to directly or indirectly determine the number of somatic cells in the milk of the animal.

[0014] According to one embodiment, the control unit is configured to compare previously stored data with the measurements received from the first sensor, and the control unit is configured to determine the health status of the animal based on the comparison. The previously stored data can be, for example, results of previous corresponding measurements on the animal related to a previous disease state of the animal, or results of previous corresponding measurements on other animals in the herd. The data is stored in a memory of the control system.

[0015] According to one embodiment, the automatic milking system comprises a plurality of further sensors for measuring a plurality of parameters other than or in addition to the illness indicator measured by the first sensor, the further sensors being connected to a control unit, the control unit being configured to perform a mathematical analysis of the measured parameters and based thereon determine an illness-indicating index indicative of illness based on inputs from the further sensors and the first sensor, and to generate an alert if the index exceeds a threshold value, The control unit is configured to determine the health status of the animal based on repeated measurements by the first sensor from the second time point to a predefined third time point if the first time point is within a predefined time period from a time point of generation of the alert for the animal. The further sensor may be configured to measure one or more of the following parameters: milk volume, blood lactate, dehydrogenase (LDH), activity, or body condition score. The mathematical analysis performed by the control unit 1 may be a multivariate analysis on the measurement results from the further sensor.

[0016] According to one embodiment, the control unit is configured to set the first time point based on a time point of generation of the alert.

[0017] According to one embodiment, the control unit is configured to set the second time point based on the time of generation of the alert. A third time point may also be set based on the alert, which according to one embodiment is 3-4 weeks after the alert.

[0018] According to one embodiment, the first time point corresponds to the time point of generation of the alert.

[0019] According to one embodiment, the milking system comprises a device for administering an antibiotic to the animal, the device being connected to a control unit, the control unit being configured to command the administration of the antibiotic to the animal within a predetermined time from the generation of an alert for the animal. The device for administration may be a device similar to that disclosed in US2019 / 000817.

[0020] According to one embodiment, the disease indicator is an indicator of mastitis.

[0021] According to one embodiment the indicator of disease is an indicator of clinical mastitis.

[0022] According to one embodiment, the milking system is characterized in that the period between the first and second time points is a first predetermined number of days, preferably within 5 to 10 days, during which the disease state is determined and the control unit is configured to determine that an antibiotic has been administered to the animal.

[0023] According to one embodiment, the milking system is characterized in that the period between the first and second time points is a second predetermined number of days, preferably within 2 to 4 days, during which a sickness state is determined and the control unit is configured to determine that no antibiotics have been administered to the animal.

[0024] The object of the present invention is also to provide a method for determining the health status of an animal in an automatic milking system, the automatic milking system comprising: a control system including an electronic control unit; at least one milking station configured for milking an animal; a produced milk tank for receiving milk from the milking station; a second milk tank adapted to receive milk diverted from the milking station and not sent to the production milk tank; an animal identification system connected to the control unit and configured to provide the control unit with individual identification information of animals milked by the automatic milking system; at least one first sensor configured to measure at least one indicator of a disease in the animal, the first sensor being connected to a control unit; a data memory in which the measurements carried out by the first sensor on the milk from each animal identified by the identification system are collected, the time of each measurement also being registered and stored in the memory, the method comprising: - registering a first time point at which milk from the animal is not sent to the production milk tank but is instead diverted to a second tank; - if the milk of the animal is diverted from the milking station to the second milk tank over a number of successive milking sessions after the first time point, registering a second time point at which the milk from the animal is transferred from being sent to the second tank to being sent to the production milk tank; - using the control unit to determine that the animal is in a diseased state between the first and second time points if the period between the first and second time points is within a predetermined number of days.

[0025] According to one embodiment, the method includes performing a mathematical analysis by the electronic control unit to determine a state of recovery of the animal from the disease state at a third time point.

[0026] According to one embodiment, the method includes determining a state of recovery from the disease state when a measurement from the first sensor stabilizes within a predetermined range between the second time point and the third time point.

[0027] According to one embodiment, the method comprises: - storing in a memory the data received from the sensor; - comparing such previously stored data with measurements received from the sensor and determining, by the control unit, the animal's health status based on said comparison.

[0028] According to one embodiment, the method includes the steps of: measuring a plurality of parameters other than or in addition to the sickness indicator measured by the first sensor; performing a mathematical analysis of the measured parameters and based thereon determining a sickness-indicating index indicative of sickness of the animal based on said measurements; and generating an alert if said index exceeds a threshold value; - if the first time point is within a predetermined time period from a time point of generation of the alert for the animal, determining the health status of the animal based on repeated measurements by the first sensor from the second time point to a predetermined third time point.

[0029] According to one embodiment, the method includes setting the first point in time based on a generation time of the alert.

[0030] According to one embodiment, the method includes setting the second point in time based on a generation time of the alert.

[0031] According to one embodiment, the method includes administering an antibiotic to the animal within a predetermined time from the time of generation of the alert for the animal. [Brief description of the drawings]

[0032] [Figure 1] 1 is a schematic diagram of an automatic milking system according to the present invention; [Diagram 2] 1 is a flow chart illustrating a first embodiment of the method of the present invention. [Diagram 3] 4 is a flow chart illustrating a second embodiment of the method of the present invention. [Figure 4] FIG. 13 shows somatic cell count / ml versus time when determined to be non-recovery. [Diagram 5] FIG. 1 shows somatic cell count / ml versus time when recovery was determined. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0033] 1 shows an automatic milking system comprising a control system including an electronic control unit 1, a milking station 2 configured for milking an animal, a product milk tank 3 for receiving milk from the milking station 2, and a second milk tank 4 configured for receiving milk diverted from the milking station 2 and not sent to the product milk tank 3. The product milk tank 3 is connected to the milking station 2 via a product milk line 5 and the second milk tank 4 is connected to the milking station via a second milk line 6.

[0034] The system also includes an animal identification system (not shown) connected to the control unit 1, configured to provide the control unit 1 with individual identities of the animals milked by the automatic milking system. The identification system typically includes an identification tag attached to each animal.

[0035] The milking system also includes at least one first sensor 7 configured to measure at least one indicator of an animal's illness, said sensor 7 being connected to the control unit 1. The first sensor 7 is arranged in the production milk line 5. Alternatively, the first sensor 7 is arranged in a bypass line from the milk production line 5. According to one embodiment, the first sensor 7 is a sensor capable of measuring the electrical conductivity of the milk. According to another embodiment, the first sensor 7 is a sensor capable of detecting the number of somatic cells in the milk. The milking system also includes a data memory 8 in which the measurements performed by the first sensor 7 on the milk from the individual animals identified by the identification system are collected. The time of each measurement is also registered and stored in the memory, so that a time log is created in the memory.

[0036] The milking system also includes a number of further sensors, indicated at 9, 10, 11, for measuring a number of parameters other than or in addition to the sickness indicators measured by the first sensor 7. The further sensors 9, 10, 11 are connected to the control unit 1, which is configured to perform a mathematical analysis of the measured parameters of the further sensors 9, 10, 11 and based thereon to determine a sickness-indicating index indicative of sickness based on inputs from said further sensors and said first sensor and to generate an alert if said index exceeds a threshold value. The further sensors may be configured to measure one or more of the following parameters: milk volume, blood lactate, dehydrogenase (LDH). The further sensors may also alternatively include sensors for measuring activity or a physical condition score. The mathematical analysis performed by the control unit 1 may be a multivariate analysis on the measurement results from the further sensors 9, 10, 11.

[0037] The milking system also includes a device 12 for administering an antibiotic to the animal, which device is connected to the control unit 1 , which is configured to control the operation of the administration device 12 .

[0038] 2 shows a flow chart in which the method steps carried out throughout the operation of the automatic milking system of the present invention are carried out. The following steps are carried out: S1: A first time point t1 is registered and starting from the first time point t1, the milk of the animals is diverted from the milking station 2 to the second milk tank 4 instead of to the product milk tank. S2: Register a second time point t2 when the milk from the animal is transitioned from being sent to the second tank to being sent to the production milk tank. The second time point t2 is registered if the milk of the animal is diverted from the milking station for multiple consecutive milking sessions after the first time point t1. S3: Determine whether the animal was or is in a disease state: If the period between the first time point t1 and the second time point t2 is within a predetermined number of days, the control unit is configured to determine that the animal is or was in a disease state. S4: Perform mathematical analysis with the electronic control unit to determine the state of recovery of the animal from the disease state at the third time point t3. This step includes determining the state of recovery from the disease state. Recovery is considered to occur when the measurement from the first sensor stabilizes within a predetermined range between the second and third time points. The first sensor is a sensor configured to measure the conductivity of the milk or a sensor configured to directly or indirectly determine the number of somatic cells in the milk of the animal. Figure 5 shows the number of somatic cells / ml versus time when recovery is determined.

[0039] FIG. 3 shows another embodiment of the method, which includes the following steps. S101: The control unit 1 determines that a given animal has clinical mastitis based on measurements of the further sensors 9, 10, 11 and by applying multivariate analysis to the obtained values. S102: In response to a command from the control unit 1, an antibiotic is administered to the animal by the administration device 12. S103: On command from the control unit 1, register a first time point t1 at which the animal's milk is diverted from the milking station 2 to the second milk tank 4 instead of being sent to the production milk tank 3. The control unit 1 is configured to control the diversion of milk to the second milk tank 4, for example by controlling a valve 13, through which the second milk line 6 is diverted from the production milk line 5 (see Figure 1). S104: Determine and register a second time point t2 at which milk from the animal transitions from being sent to the second tank 4 to being sent to the product milk tank 3. The second time point t2 is registered when the animal's milk has been diverted from the milking station for multiple consecutive milking sessions after the first time point t1. S105: Determine whether the animal was or is in a disease state: If the period between the first time point and the second time point is within a predetermined number of days, preferably 5-10 days, the control unit 1 is configured to determine that the animal is or was in a disease state. S106: Perform mathematical analysis with the electronic control unit to determine the state of recovery from the disease state of the animal at a third time point t3. This step includes determining the state of recovery from the disease state. Recovered is considered when the measurements from the first sensor stabilize within a predetermined range between the second time point and the third time point. The first sensor is a sensor configured to measure the electrical conductivity of the milk or a sensor configured to directly or indirectly determine the number of somatic cells in the milk of the animal. Figure 5 illustrates the somatic cell count / ml versus time when recovery is determined. The somatic cell count after recovery can be up to about 20-25% higher than before the alert.

[0040] If the milk somatic cell count or conductivity is not stable within a predetermined range, it is considered not to have recovered. FIG. 4 shows an example where the somatic cell count at the third time point t3 is at the same high level as the somatic cell count at the first time point t1. It has been observed by the inventors that, if recovered, cows recover from diseases, particularly diseases such as mastitis, within 3-4 weeks from the first time point, regardless of antibiotic administration or other treatment. If recovery is not observed after 3-4 weeks, the disease is likely to be chronic.

Claims

1. 1. An automated milking system comprising: a control system including an electronic control unit (1), at least one milking station (2) adapted to milk the animals, a produced milk tank (3) for receiving milk from said milking station (2); a second milk tank (4) adapted to receive milk diverted from said milking station (2) and not sent to said product milk tank (3); an animal identification system connected to said control unit (1) and adapted to provide said control unit (1) with individual identities of animals milked by said automatic milking system; at least one first sensor (7) configured to measure at least one indicator of a disease in an animal, said sensor being connected to said control unit (1); a data memory (8) in which the measurements carried out by said first sensor (7) on the milk from each animal identified by said identification system are collected, the time of each measurement also being registered and stored in said memory (8), said control unit (1) comprising: - registering a first time point (t1) at which milk from an animal is not sent to said product milk tank (3) but is diverted to said second tank, - configured to register a second point in time (t2) at which milk from the animal transitions from being sent to the second tank to being sent to the product milk tank (3), if the milk of the animal has been diverted from the milking station (2) to the second milk tank (4) for a number of successive milking sessions after the first point in time (t1), the control unit (1) is configured to determine that the animal was in a sick state between the first time point (t1) and the second time point (t2) if the period between the first time point (t1) and the second time point (t2) is within a predetermined number of days, the control unit (1) is configured to determine the health state of the animal based on repeated measurements by the first sensor (7) from the second time point (t2) to a predetermined third time point (t3), the control unit (1) is configured to determine the state of recovery from the sick state of the animal at the third time point (t3), and the control unit (1) is configured to determine the state of recovery from the sick state when the measurements from the first sensor (7) stabilize within a predetermined range between the second time point (t2) and the third time point (t3).

2. 2. The automatic milking system according to claim 1, characterized in that the first sensor (7) is a sensor configured to measure the electrical conductivity of milk or a sensor configured to determine, directly or indirectly, the number of somatic cells in the milk of the animal.

3. 2. The automatic milking system according to claim 1, characterized in that the control unit (1) is configured to compare previously stored data with measurements received from the first sensor (7), and the control unit (1) is configured to determine the health state of the animal based on the comparison.

4. a plurality of further sensors (9, 10, 11) for measuring a plurality of parameters other than or in addition to the illness indicator measured by the first sensor (7), said further sensors (9, 10, 11) being connected to said control unit (1), said control unit (1) being configured to perform a mathematical analysis of the measured parameters and based thereon determine an illness-indicating index indicative of an illness based on inputs from said further sensors (9, 10, 11) and said first sensor (7) and to generate an alert if said index exceeds a threshold value, The automatic milking system according to any one of claims 1 to 3, characterized in that the control unit (1) is configured to determine the health state of the animal based on repeated measurements by the first sensor (7) from the second time point (t2) to a predetermined third time point (t3) if a first time point (t1) is within a predetermined period of time from the time of generation of the alert for the animal.

5. 5. The automatic milking system according to claim 4, characterized in that the control unit (1) is adapted to set the first time point (t1) based on the time point of generation of the alert.

6. The automatic milking system described in claim 4, characterized in that the control unit (1) is configured to set the first point in time (t1) based on the point in time at which the alert is generated, and the control unit (1) is configured to set the second point in time (t2) based on the point in time at which the alert is generated.

7. 7. An automatic milking system according to any one of claims 4 to 6, characterised in that it comprises a device (12) for administering an antibiotic to an animal, said device (12) being connected to said control unit (1), said control unit (1) being configured to command the administration of said antibiotic to said animal within a predetermined time from the time of generation of an alert for said animal.

8. An automatic milking system according to any one of claims 1 to 7, characterised in that the disease indicator is an indicator of mastitis.

9. 9. The automatic milking system of claim 8, wherein said indicators of disease are indicators of clinical mastitis.

10. 10. The automatic milking system according to any one of claims 1 to 9, characterized in that the control unit (1) is configured to determine that an antibiotic has been administered to the animal if the period between the first time point (t1) and the second time point (t2) is within a first predetermined number of days, for which a disease state is determined.

11. 11. The automatic milking system of claim 10, characterized in that the control unit (1) is configured to determine that no antibiotics have been administered to the animal if the period between the first time point (t1) and the second time point (t2) is within a second predetermined number of days, at which a disease state is determined.

12. 1. A method for determining an animal's health status in an automatic milking system, the automatic milking system comprising: a control system including an electronic control unit (1), at least one milking station (2) adapted to milk the animals, a produced milk tank (3) for receiving milk from said milking station (2); a second milk tank (4) adapted to receive milk diverted from said milking station (2) and not sent to said product milk tank (3); an animal identification system connected to said control unit (1) and configured to provide said control unit (1) with individual identities of animals milked by said automatic milking system; at least one first sensor (7) configured to measure at least one indicator of a disease in an animal, said sensor being connected to said control unit (1); a data memory (8) in which the measurements carried out by said first sensor (7) on the milk from each animal identified by said identification system are collected, the time of each measurement also being registered and stored in said memory (8), said method comprising the steps of: - registering a first point in time (t1) at which milk from an animal is not sent to said product milk tank (3) but is diverted to said second tank; - if the milk of the animal has been diverted from the milking station (2) to the second milk tank (4) for several successive milking sessions after the first time point (t1), registering a second time point (t2) at which the milk from the animal transitions from being sent to the second tank to being sent to the product milk tank (3); - determining by said control unit (1) that said animal is in a disease state between said first time point (t1) and said second time point (t2) if the period between said first time point (t1) and said second time point (t2) is within a predetermined number of days, The method includes a step of performing a mathematical analysis by the electronic control unit (1) to determine the state of recovery of the animal from the disease state at the third time point (t3), and a step of determining the state of recovery from the disease state when the measurement value from the first sensor (7) stabilizes within a predetermined range between the second time point (t2) and the third time point (t3).

13. 13. The method according to claim 12, characterized in that the first sensor (7) is a sensor configured to measure the electrical conductivity of milk or a sensor configured to determine, directly or indirectly, the number of somatic cells in the milk of the animal.

14. - storing in said memory (8) the data received from said sensors; - comparing such previously stored data with measurements received from the sensor and determining, by the control unit (1), the animal's health state based on said comparison.

15. measuring a number of parameters other than or in addition to the sickness indicator measured by said first sensor (7), performing a mathematical analysis of said measured parameters and determining based thereon a sickness-indicating index indicative of a sickness of the animal based on said measurements and generating an alert if said index exceeds a threshold value; - determining the health state of the animal based on repeated measurements by the first sensor (7) from the second time point (t2) to a predefined third time point (t3), if the first time point (t1) is within a predefined time period from a time point of generation of the alert for the animal.

16. 16. The method of claim 15, further comprising the step of setting the first point in time (t1) based on a generation time of the alert.

17. The method includes a step of setting the first point in time (t1) based on a generation time of the alert, 16. The method of claim 15, further comprising the step of setting the second point in time (t2) based on a generation time of the alert.

18. A method according to any one of claims 15 to 17, characterised in that it comprises the step of administering an antibiotic to the animal within a predetermined time from the time an alert for the animal was generated.

Images