Method and robotic milking device for milking dairy animals

Abstract

A robotic milking system (1) comprises a milking station (3), an animal identification device (7), a robotic milking device (5), a feeding device (6) for providing feed to the animal during the milking based on a feed allowance, and a controller (10) for the robotic milking system and for updating the feed allowance. The method comprises: selecting a future time period; determining a free capacity C of the robotic milking system; determining a subgroup of animals to which the controller can continue to provide feed after the milking; milking and feeding the animals accessing the milking system; and if an animal from the subgroup still has a feed allowance after the milking, feeding the animal additionally depending on the determined free capacity C. By additionally feeding the selected animal not in all cases but depending on the free capacity, additional feed is only given when the robot does not need to immediately milk the subsequent animal, which reduces the milking capacity less.

Description

[0001] In a first aspect, the present invention relates to a method for automatically milking and feeding a group of dairy animals by means of a robotic milking apparatus, the robotic milking apparatus being arranged in an area where the dairy animals can move freely and comprising: a milking site for receiving dairy animals; an animal identification device for determining the animal identity of the dairy animals; a robotic milking device for fully automatically extracting milk from the dairy animals during milking; a feeding device for providing feed to the dairy animals at the milking site during the milking based on the feed allowance of the dairy animals; and a controller configured to control the robotic milking system and update the feed allowance of each dairy animal, wherein the method comprises the following steps: selecting a future time period; determining a sub-group of the group of dairy animals, to which the controller can continue to provide feed after the milking; milking dairy animals accessing the milking system and feeding them during milking; and if dairy animals from the sub-group still have feed allowance after milking, then additionally feeding the dairy animals.

[0002] Such robotic milking systems are known per se. They almost always have a concentrate feeding device which provides concentrate feed to the dairy animals during milking. Not only are robotic milking devices ideally suited to providing individualized concentrate feed portions to dairy animals due to their always provided animal identification device, but concentrate feed is not only usually necessary for adequate rationing of dairy animals, but is also a very attractive feed, so that a sufficiently high frequency of visits to the robotic milking device can be guaranteed.

[0003] A disadvantage of robotic milking equipment is that, in view of the cost, it is desirable to optimize and usually maximize the milking capacity. Since most dairy animals are milked faster than they consume the portion of concentrate feed allocated to them, the dairy animals are at risk of being underfed. This applies in particular to the special but very common situation where the robotic milking equipment is the only place where concentrate feed is available. Otherwise, not only is the milking equipment not very effective in attracting, but the feed shortfall may then be made up in another concentrate feed station, so that the milking capacity is not affected thereby.

[0004] To compensate for this disadvantage, it is natural to ensure that (some or all) dairy animals always receive the portion of concentrate feed allocated to them in the robotic milking facility. A3 / A4 / A5 have the option of "priority feeding" by which the user can specify for one or more dairy animals that they should always be fully fed. The disadvantage of this is that the overall milking capacity of the robotic milking plant is reduced, since additional feeding is carried out at each milking, even if at that moment no free robot capacity is available due to full occupancy. The following animal must therefore then wait an unnecessarily long time.

[0005] The present invention aims to at least partially eliminate the above-mentioned disadvantages and in any case to provide a robotic milking device which reduces the risk of insufficient feed without initially reducing the milking capacity. Therefore, the present invention provides a method as claimed in claim 1, in particular a method for automatically milking and feeding a group of dairy animals with the help of a robotic milking device, the robotic milking device being arranged in an area where the dairy animals can move freely and comprising: a milking site for receiving dairy animals; an animal identification device for determining the animal identity of the dairy animals; a robotic milking device for fully automatically extracting milk from the dairy animals during milking; a feeding device for providing feed to the dairy animals at the milking site based on the feed allowance of the dairy animals during milking. feed; and a controller configured to control the robotic milking system and update the feed allowance for each dairy animal, wherein the method comprises the following steps: selecting a future time period; determining an idle capacity C of the robotic milking system as an expected value of the total time that the milking system will not be occupied by dairy animals during the future time period; determining a subgroup of the group of dairy animals, to which the controller can continue to provide feed after the milking; milking dairy animals accessing the milking system and feeding them during milking; and if dairy animals from the subgroup still have feed allowance after milking, feeding the dairy animals additionally depending on the determined idle capacity C.

[0006] Here, the present invention uses the following principle. A robotic milking device (hereinafter also referred to as a "robot") that is arranged in an area where dairy animals can move freely is not always visited at a constant frequency. The area does not need to be a milking parlor, but can also be a pasture, or even a piece of public land without a fence, as long as the dairy animals can freely visit the robot. It is known per se that, for example, dairy cows usually rarely visit the robot at night. It may even be the case that at some times the robot is idle, while at other times the robot is continuously occupied. The degree of occupancy of the robot can also change depending on the number of dairy animals, for example because the livestock has dried up or has been removed from the herd for health reasons, or livestock has been added. In addition, milk production changes during the lactation period, so that the associated milking time can also change. In summary, the robot sometimes has no spare time, but sometimes has a lot of spare time.

[0007] It is then advantageous to use the time when the robot is less occupied to provide some of the livestock that visit the robot with some or all of the feed allocated to them according to their feed allowance. During busy periods, the robot does not need to set aside additional time to provide feed. It is therefore also possible that the dairy animals "recover" more or less of any feed shortage during such quiet times. Since the robot provides additional surplus feed depending on the idle capacity, this will not happen if there is no (more) idle capacity, so the robot will not lose milking capacity. Therefore, it may be the case that the idle capacity is not used or reduced due to unexpectedly long times spent on multiple milkings, or due to a malfunction, or due to a long period of time without dairy animals visiting the robot. In this case, the robot will be able to provide additional feed only to some of the dairy animals from the subgroup, or in exceptional cases there is no such livestock to be provided with additional material. Nevertheless, the present invention can reduce the risk of insufficient feed without losing milking capacity. It should be noted that "insufficient feed" is understood here as an excessive "feed allowance". In addition, feeding is carried out based on the feed allowance so that when the feed allowance is consumed, feeding stops. It is well known that the feed allowance per unit time increases with the recommended ration for that unit time and decreases with the ration eaten at each feeding event. For example, if a dairy cow receives 9 kg of concentrate feed per day, the feed allowance increases by 0.375 kg / hour. It is simple for the robot to update the feed allowance because the dairy animal is identified in the robot, the visit time is recorded, and the amount of concentrate feed provided and usually also eaten is recorded.

[0008] It is important to note that only a subgroup of dairy animals is considered for such additional feeding actions. Thus, dairy animals that are less in need of such actions can be properly prevented from using up the sometimes scarce available spare capacity. Thus, the risk of insufficient feed for the entire herd can be reduced, and of course also the risk of insufficient maximum feed. In some cases, the following situation may occur: for example if the livestock in the lactation period is a much smaller group than the robot can handle, then the subgroup contains the entire herd of dairy animals, but it is important not only that the subgroup is not fixed but can be changed, but also that dairy animals from this subgroup are not necessarily given additional feed after milking, because this depends on the spare capacity of the robot.

[0009] Furthermore, if the dairy animal in question still has a feed allowance after milking, the robot only needs to continue to provide concentrated feed. Thus, it may be the case that the milking cup connection action takes an unexpectedly long time, or that the milking cups separate early during the milking process, so that the milking time is prolonged, and so on. In this case, dairy animals that normally have a remaining feed allowance after milking can eat their entire feed allowance during a longer milking period. And, for various calculations, the expected value of the feed allowance after milking is used. For example, this is calculated by looking at the expected average visit time of the dairy animal based on historical visit times, then calculating the feed allowance belonging to the expected average visit time, and then if the feed allowance has such a value, then looking at how much the dairy animal eats on average. The remainder is the expected value of the feed allowance that has not yet been added. Furthermore, other calculation methods may be followed.

[0010] It should also be noted here that during a visit to the robot, the dairy animals are not actually necessarily milked. It is always possible that dairy animals from the subgroup visit the robot but do not yet need to be milked. However, it can be advantageous, for example in the case of sufficient free capacity C and if the animals still have feed allowance or a certain minimum feed allowance, not to drive the dairy animals away from the robot but to still provide concentrated feed. Moreover, the time of such additional feeding actions is then included in the free capacity C, in the sense that C then has to be reduced.

[0011] Specific embodiments are given in the dependent claims and in the following part of the description.

[0012] In some embodiments, the step of determining the subgroup comprises the user inputting into the controller the identities of only a portion of the group of dairy animals. This may be useful in situations where, for example, only a few animals should be guaranteed their concentrate ration, such as animals recovering from an illness. The controller then selects from this subgroup the dairy animals that must be given extra concentrate feed, in any case until the spare capacity is used up. The user may then dynamically adjust the subgroup at any time (e.g. when the dairy animals have sufficiently recovered, or if a new case has occurred, etc.).

[0013] In some embodiments, the step of determining said subgroups is performed automatically by the controller. This has the advantage that some activities are not handled by the managers of the group of dairy animals and creates the possibility that the feed provision of dairy animals can be dynamically optimized.

[0014] In a practical embodiment, the step of determining the subgroup comprises: for each dairy animal, assigning a selection priority based on a selection criterion including at least the expected value of the feed allowance after milking; and selecting the dairy animal with the highest selection priority in the number that meets the predefined selection criterion. As already pointed out above, it is not equally important to prevent a feed shortage for each dairy animal in the group, for example, because the feed shortage of one animal almost never becomes large, while the feed shortage of another animal may become large. Therefore, a selection priority can be assigned to each dairy animal, wherein a high selection priority means a high chance of selection in the subgroup, and vice versa. For example, the selection priority can be expressed in terms of the expected value of the remaining feed allowance after milking, wherein a large value means a higher priority. The value can also be increased proportionally, or, for example, rise much more than the value. In addition, instead of an absolute remaining feed allowance, a relative remaining feed allowance, for example relative to the daily ration, can be used. It can also be specified that the remaining feed allowance must be greater than a predefined threshold, etc. Therefore, differences in daily feed rations can be taken into account.

[0015] In particular, the selection criteria include that the sum of the eating times of each dairy animal from the subgroup, calculated by the controller and belonging to the expected value of the corresponding feed allowance after milking, is at most equal to the idle capacity or a predefined mathematical operation of the idle capacity. In this way, the subgroup can be matched with the available idle capacity so that as many dairy animals as possible can benefit from the possibility of consuming their entire feed allowance in the robot without losing milking capacity. The eating time can be determined, for example, based on historical visits to the robot. Its duration can be stored together with the amount of concentrated feed eaten. The expected value of the remaining feed allowance can then be divided by the eating rate determined thereby, whereby the additional eating time for each dairy animal can be determined. Again, other ways of determining the eating time are also possible, such as target test measurements, which can, for example, take into account any slower eating at or after the end of milking.

[0016] Here, the sum of the calculated additional feeding times can be simply equalized to the available idle capacity, and then the idle capacity can be filled with these feeding times, for example (but not necessarily) based on the selection priority and in the order of the selection priority. Therefore, the subgroup can be matched with the available idle capacity. In particular, the mathematical operation is the idle capacity multiplied by a value between 0.8 and 1.5. Therefore, less than the full idle capacity can be used for additional feeding, such as between 80% and 100%. This takes into account any unexpected events, such as milking interruptions or otherwise requiring a long time. In this way, milking and feeding then have some flexibility. Secondly, this can avoid the robot from remaining idle while waiting for the next visit. It is also possible to retain a capacity greater than the idle capacity, such as between 100% and 150%. In theory, this will lead to a shortage of robot capacity and therefore a loss of milking capacity, but this can take into account the fact that it may not be possible that all dairy animals selected in the subgroup will actually visit the robot in the future time period. In this way, there will always be enough dairy animals visiting the robot and being given their full feed allowance. The degree of overcapacity (i.e. multiplication by a value greater than 1) can actually be determined from, for example, the measured visit distribution. A herd of dairy animals that is visited regularly needs to have fewer dairy animals than a herd of dairy animals that is visited irregularly. Furthermore, this situation may require an earlier dynamic adjustment of the sub-herds with respect to the used spare capacity.

[0017] In some embodiments, the selection priority is assigned based in part on the probability that the dairy animal will be milked in the future time period. As described above, there is naturally no guarantee that the dairy animal with a high chance of obtaining the remaining feed allowance will actually visit the robot in the future time period. For example, the dairy animal in question may have just been milked, so that it will not visit the robot in the next few hours. Then there is no point in assigning a high selection priority. According to this embodiment, this can be taken into account in the following way: the selection priority is adjusted according to the chance of visiting the robot during the future time period. This chance can be determined based on historical visit frequencies, etc. It should be noted that only robot visits when milking is ready, or every arbitrary visit, can be included here. Because even if the dairy animal is not ready for milking, dairy animals with a large feed deficit can be fed additionally.

[0018] In some embodiments, the selection priority is partly based on one or more livestock characteristics other than the expected feed allowance after milking, in particular the number of days of lactation and / or the number of lactations and / or the amount of milk produced. With the help of such livestock characteristics, the manager or controller of the group of dairy animals can take into account the characteristics related to feeding. For example, for dairy animals in the increasing lactation stage, that is, dairy animals with less than the (expected or average) peak production day of lactation, it is more important to provide sufficient feed or at least prevent feed shortage than other dairy animals. Another consideration may be to treat the most prolific dairy animals in this way, for example, because they face a greater risk of (very large) feed shortage, or from an economic point of view, they are more important to farmers. In addition, other characteristics such as the above-mentioned health characteristics may also be used as relevant factors in determining the selection priority.

[0019] In a particular embodiment, the robotic milking device comprises an upstream waiting area, wherein the controller is configured to determine the livestock identity of the dairy animals in the waiting area, and wherein the selection priority is assigned partly depending on the presence of dairy animals in the waiting area. In such a robotic milking device, it is possible to better predict which dairy animals will soon participate in a visit to the robot. In this way, the selection priority and therefore the subgroup can be adjusted accordingly. For example, the chance of visiting the robot can be determined more accurately, and dairy animals that do not have a high expected value for the remaining feed allowance after milking can be added to the subgroup. After all, it is better to give these dairy animals their full feed allowance than to wait unnecessarily for dairy animals that did not appear based on information from the waiting area.

[0020] In a further advantageous embodiment, the robotic milking device further comprises a dairy animal locating device, which is configured to determine the position of the dairy animal in the area, and the selection priority is assigned partly based on the determined position in the area, in particular based on the distance from the determined position to the robotic milking device, more particularly the change in position or distance. In these embodiments, for example, the controller can better estimate for each dairy animal the probability that this dairy animal will visit the robot in a future time period. In any case, dairy animals that are far away from the robot or whose distance increases will have a lower probability (and therefore obtain a lower selection priority) than dairy animals whose distance is small or rapidly decreasing. Therefore, it can be better prevented that part of the idle capacity is unnecessarily reserved for dairy animals that may not appear in the future time period.

[0021] The method for determining the idle capacity C is not particularly limited. For example, the idle capacity is determined based on historical values, such as the measured idle time and / or the access frequency of the robotic milking device, and / or the planned cycle of cleaning or maintenance of the robotic milking device, in each case determined according to the time of day. The idle time is, for example, the time between the dairy cow leaving the robot and the next visit of the next dairy cow. And the time when the robot is unavailable for cleaning or maintenance is also idle time. Maintenance and / or cleaning can be planned and unplanned. Planned events can be considered more easily than unplanned events, however, for unplanned events, such as historical averages can still be considered. In addition, optionally or additionally, the access frequency can also be viewed, and the idle capacity can be calculated or adjusted based on the frequency and the average expected milking duration or in another suitable way. It should be emphasized that the access frequency varies according to the time of day, but this can also vary during the year, for example due to the influence of grazing, etc.

[0022] The future time period for determining the calculation of idle capacity etc. must also not be particularly limited. In particular, the future time period is the time period until a planned cleaning or maintenance in the future, or a predefined fixed time period, such as 1 hour. In a longer time period, such as typically in the time period until a planned cleaning, in general, more idle capacity can be used for additional feeding than in a short time period. However, in a shorter or at least closer time period, it is possible to better and more accurately calculate and estimate which dairy animals need how much time for additional surplus feed. In practice, it will be easy to determine which time period gives the desired result.

[0023] In an important embodiment, the method also comprises repeating the method according to the invention after a predefined number of milkings or visits of the dairy animal to the robotic milking device, or after a predefined time. However, the free capacity C will always change over time and of course after milking, with or without additional feeding. Therefore, the time taken for milking may be shorter or longer than expected, the frequency of visits may be different from expected, etc. It is then advantageous to adjust the calculation after each milking or each visit of the dairy animal to the robot, or after the expiration of a predefined time such as 10 minutes or 1 hour. The calculation includes not only the free capacity, but also the subgroups, the selection priorities, etc. In fact, if a dairy animal with a high selection priority visits the robot in a future time period (early or late), and for example happens to also have a milking event that is so long that the full feed allowance can be given, then not only in any case in the instantaneous "future" time period (now in use), the dairy animal should be removed from the subgroup or its selection priority should be reduced, but also further subgroups must be adjusted for the now available free capacity, etc. It is then advantageous to repeat the method with an updated instantaneous situation. Furthermore, it is useful or even desirable, for example for all dairy animals, to stop the additional feeding if the time used for the additional feeding exceeds the expected free capacity. The capacity is then "consumed" in the current "future" time period. It should be noted that the term "future time period" applies before the calculation. Once the associated milking and feeding starts, or more precisely, once the relevant time period has started, this time period is no longer in the future. For the sake of clarity, the term "future time period" is still used for this time interval.

[0024] In addition to the features of additional feeding after milking, additional feeding can also be provided during milking. For example, at least for livestock from a subgroup of dairy animals that do not receive additional feed after milking, energy-rich supplements or otherwise more energy-rich feed can be provided during milking, so that the entire feed allowance can be provided during milking. Therefore, propylene glycol or other additives can be added to the concentrate feed. Therefore, initially all livestock can be fully fed when there is a spare capacity C, wherein other subgroups of dairy animals receive additional feed after they are milked. If there is no spare capacity C, additional feed can also be provided to all livestock in this way. However, the present invention still brings advantages relative to a system that works with such "auxiliary" feeding. Because the method according to the present invention uses less such additives, and livestock can be fed in a simpler way.

[0025] The invention also relates to a robotic milking apparatus as claimed in claim 14, in particular a robotic milking apparatus for milking a group of dairy animals, the robotic milking apparatus comprising: a milking site for receiving dairy animals; an animal identification device for determining the animal identity of the dairy animals; a robotic milking device for fully automatically extracting milk from the dairy animals during milking; a feeding device for providing feed to the dairy animals at the milking site during milking based on the feed allowance of the dairy animals; and a controller configured to control the robotic milking system and update the feed allowance of each dairy animal and to perform the method according to the invention, wherein the robotic milking apparatus is arranged in an area where the dairy animals can move freely. Again, it is emphasized that the robot is arranged in an area (pasture, milking parlor, nature reserve) where dairy animals, such as cows or goats, can freely access the robot. In any case, it is simple to prevent feed shortages in the case of driving dairy animals to scheduled milking events of the milking apparatus. However, the principle of spare capacity is not used here, since milking is planned more precisely.

[0026] Specific embodiments of the robotic milking device (also referred to as robot) largely correspond to specific embodiments of the method. In particular, the robot further comprises an upstream waiting area, and the controller is configured to perform the method according to the invention, wherein the controller is configured to determine the livestock identity of the dairy animal in the waiting area and to assign a selection priority partly depending on the presence of the dairy animal in the waiting area.

[0027] In a further embodiment, the robotic milking apparatus further comprises a dairy animal locating device configured to determine the position of the dairy animal in the area, and the controller is configured to determine the position of the dairy animal in the area and to assign a selection priority based partly on the determined position in the area, in particular on the distance from the determined position to the robotic milking apparatus.

[0028] For the described embodiment, the advantages already mentioned for the method also apply, so that for the sake of brevity they do not need to be repeated here.

[0029] The present invention will be explained in more detail below with reference to the accompanying drawings. The accompanying drawings show:

[0030] Figure 1 shows a schematic top view of a robotic milking device according to the present invention,

[0031] Figure 2 Schematically shows the sequence of steps of the method according to the invention, and

[0032] Figure 3 Shows Figure 2 Improvement of the method.

[0033] Figure 1 A schematic top view of a robotic milking device 1 according to the invention, also referred to as robot in the following, is shown. The robotic milking device comprises a central part 2 and a milking station or milking stall 3 and an accommodation area 20 with cows A and B.

[0034] The central part comprises a controller 10 and a robot arm 4 with a milking device 5. The milking shed 3 comprises a feed trough 6 and an animal identification device 7. Symbols 8a to 8d denote antennas of an optional animal positioning system, and 9 denotes an animal ID tag. Also optional is a waiting area 21 with an entrance gate 22, a waiting area animal identification device 23 and an exit gate 24.

[0035] The robotic milking device 1 comprises a central part 2, in which for example vacuum pumps, milking cups and other equipment required for robotic milking are arranged. A controller 10 is also provided for controlling the device and a robotic arm 4. The robotic arm carries a milking device 5, i.e. four milking cups for milking four teats of cows A and B in a milking shed 3. Furthermore, other numbers of milking devices / milking cups are possible, such as two milking cups for goats, and alternative configurations of the milking part of the robotic milking device can also be provided, such as a milking cup magazine and a robotic arm with a gripper which takes the milking cups from the magazine in each case and brings them to the teats with the help of a teat detection device (not shown here).

[0036] The robotic milking device 1 also comprises a feeding device, here represented as a trough 6. When a cow A or B enters the milking shed 3, it is identified by the animal identification device 7 based on the animal ID tag 9 worn around her neck. The information pertaining to the animal ID enables the robot 1 to milk the cow appropriately and individually and to give the cow an appropriate ration of concentrate feed in the trough 6. The controller 10 can put the concentrate feed from a storage container (not shown here) into the trough, from which the cow can eat during milking. In order to prevent waste, and also because the cow cannot eat very quickly, the ration is divided into a number of portions. It may even be the case that during a milking event, the cow cannot completely eat the portion of concentrate feed allocated to it or the "feed allowance". In particular, in the case where the robot 1 is the only site where the cow A, B can eat the concentrate feed, this may lead to insufficient feed, which is detrimental to milk production and even to the health of the cow. Typically, the robot is now the only place where concentrate feed is given, not only because the robot 1 is provided with an animal identification device 7, whereby the robot can provide concentrate feed specifically to each animal, but also because if concentrate feed is given elsewhere, the attracting effect of the concentrate feed in the robot will be weakened.

[0037] Therefore, in the prior art, it is possible to specify in the robot 1 for each cow the feed that must be provided after the milking of the corresponding cow until the feed allowance is consumed. However, this means that for a short or long period of time, the robot is only busy with feeding, while other cows are waiting to be milked. The milking capacity of the robot 1 is therefore reduced.

[0038] The present invention overcomes this drawback by understanding that the frequency of visits of the robot varies during the day. This means that sometimes there is no time for feeding, but there is such a feeding time during quiet moments. When it is quiet and the robot is not milking the cows, if the cows that are not milked but still have a feed allowance are actually given extra feed, at least partially in any case, no capacity is wasted.

[0039] In the embodiment shown, this occurs as follows. The controller 10 has information about the frequency of visits or occupancy levels of the robot 1 according to the time of day. Based on this information, the controller 10 can estimate how much time is available for additional feeding over a period of time because the robot 1 is not occupied. This available time is referred to here as "idle capacity C". The idle capacity C also changes during the day in a manner opposite to the occupancy level. Typically, during certain periods of the day, this idle capacity C is not enough to provide additional food for all cows, and the controller 10 must make a choice. Then, a subgroup of cows is given the opportunity to obtain additional feed during a visit to the robot 1, while another complementary subgroup is not given this opportunity. In the figure, each cow is represented by the letter A (provided with additional feed) or B (not provided with additional feed). It should be noted that these subgroups can also vary with the time of the day, but not necessarily.

[0040] If cow A visits the robot 1, she will be milked as usual and fed during milking. But cow "A" also gets additional feed, for example, until they have eaten their entire feed allowance. In addition, less than the entire remaining feed allowance, such as half, may also be provided. However, if cow "B" arrives, she is milked and fed during milking, but when milking is finished, she must leave the milking shed 3 even if she still has a remaining feed allowance.

[0041] Farmers can determine which cows are given "A" status and which are given "B" status. For example, if one or more cows are recovering from illness, or are producing particularly large amounts of milk, and therefore require large amounts of concentrate feed that the cows may not be able to eat in time, the farmer may designate those cows as "A".

[0042] However, the invention is not intended to be optimal in such a case of division into subgroups A and B in a fixed manner. In any case, when there is a lot of free capacity C, many cows can get extra feed. The subgroup of "A" cows will then be relatively large. But at busy milking times, there will be almost no free capacity C, and the following situation may occur. At the beginning of the selected time period, it happens that one or more cows "A" arrive who are exactly in this state. These cows will use up the available free capacity, and the most needed "A" cows will then be missed.

[0043] In order to further improve the advantages of the present invention, the possibility is created to allow the controller 10 to automatically determine the "A" and "B" cow subgroups. This will be referred to below Figure 2 Let me explain in more detail.

[0044] exist Figure 1 In the embodiment shown, an optional waiting area 21 is also shown, which has an entrance gate 22 and an exit gate 24. Entry and exit to the waiting area 21 is regulated by a waiting area livestock identification device 23. When a cow is ready or almost ready for milking and arrives at the entrance gate 22, she is allowed to enter the waiting area 21 after being identified by the device 23. If she is not yet ready (or far from ready) for milking, she is not allowed to come in. When the milking shed 3 becomes free for milking, the exit gate 24 and the entrance door of the milking shed (not shown separately) are opened, and the cow can enter the milking shed for milking.

[0045] The advantage of such a waiting area being known per se is that the controller 10 can better assess whether there are waiting cows and therefore whether free capacity C is expected. This is the case if the waiting area is empty; this is not the case if the waiting area is full. Furthermore, the controller 10 then knows in advance which cows are to be milked next. Again, this can be useful for a more detailed selection of cows that can get extra feed and cows that will not, even if all cows in the waiting area have an "A" status. For example, due to a malfunction or some slow milkings, the free capacity C may unexpectedly be very limited. Based on these developments, the controller 10 can also change the classification of "A" and "B" cows.

[0046] In the example shown, there are three cows, one with an "A" status and two with a "B" status, in the waiting area 21. In this simple example, the controller 10 will be able to provide additional feed to the one "A" cow, but not to the "B" cow.

[0047] The invention can be further improved by means of a livestock positioning system, here in the form of four antennas 8a to 8d. These antennas receive signals from livestock ID tags 9 equipped for this purpose (for the sake of clarity, only one cow is shown, but tags are provided on all cows), and thereby determine the position of the cow in the accommodation area 20 / waiting area 21, for example by means of triangulation or similar methods. From the change in position, the controller can determine the path of the cow. Therefore, the controller 10 can determine with a certain degree of confidence whether the cow will likely enter the waiting area 21 or the robotic milking shed 3 in the next period of time. For example, cow "A" at the bottom left in the figure is away from the robotic milking shed 3 and the waiting area 21, and is obviously not walking in that direction. Although the state is "A", the controller will not assign a high selection priority to this cow. This will refer to Figure 2 Let me explain in more detail.

[0048] Figure 2 The sequence of steps of the method according to the invention is schematically shown.

[0049] In a first step 100, the controller 10 determines a future time period for milking and feeding the herd. This time period may be selected, for example, as the next hour, or a time period from, for example, 10pm to 4am (because at other times there may be no spare capacity), or a time period until a planned cleaning. The planned cleaning time period occurs two or three times a day, and no milking is performed during this time, so that in the immediate period the robot will likely be busy and no additional feeding will be selected for some time.

[0050] When selecting a time period, the controller then determines an expected value for the idle capacity C. This can be determined based on historical values, such as visit frequency, total occupancy level or time, literature or empirical values ​​(for new dairy herds), etc. The idle capacity C is the time during which the dairy animals can eat the extra feed provided.

[0051] The controller 10 then determines for which dairy animals this free capacity will be available during this time period, in other words, it then determines the subgroup of the herd of dairy animals to which it can continue to provide feed by means of the feeding device after said milking. This determination can be made in many ways, which can be arbitrarily selected by the user / farmer.

[0052] For example, livestock are assigned a selection priority depending on certain characteristics and one or more criteria. Such a selection priority indicates a ranking of the importance of providing or not providing additional feed to a particular livestock rather than another livestock. This ranking can be used to fill idle capacity.

[0053] Therefore, the needs and health status of dairy animals can be taken into account. If a dairy cow is sick, it can be very important to optimize feeding until she recovers completely. In addition, some dairy cows produce a lot of milk and therefore require a lot of feed (concentrated feed). The chance that they will not be able to consume all of this feed during milking (which only takes a few minutes) is large, so additional feeding is a suitable option. Providing optimal feed for livestock with good feed conversion ratios can also be important economically. Other reasons may be that livestock is in an increased lactation phase, and for these reasons, if there is still a feed allowance after milking, then optimally it must be fed or obtain additional feed. In general, lactation stage and lactation count are possible components of the criteria for determining selection priorities.

[0054] There are also reasons that change over time. In particular, the chance of a remaining feed allowance after milking is important for allocating dairy animals to one subgroup or another. In any case, if the expected value of the remaining feed after the expected milking time is very high, the chance of a large remaining feed allowance after milking is usually also large. This means that the animals may have a feed deficit and in any case be fed irregularly, which should preferably be avoided. It is also possible that the animals eat slowly and, for this reason, there may be a large feed deficit after milking. All of the above reasons and criteria, as well as other conceivable criteria, can be used to determine the selection priority.

[0055] Furthermore, preferably, the chances of a dairy animal being milked and visiting the robot 1 within the selected time period must also be taken into account. In any case, if an animal is milked just before the start of the time period and the duration is not very long relative to the milking interval, the chance that the dairy animal will not visit the robot is very high. Therefore, even if this animal has a very high selection priority, it may be meaningless to reserve idle capacity for her. On the contrary, the chances of a livestock with a very low selection priority that needs milking at the beginning of the time period visiting the robot are very high. It is then certainly useful to take this situation into account and, for example, allocate capacity so that a portion of any remaining feed allowance can be provided in any case.

[0056] Based on any part of the above information, the controller 10 thus determines the selection priority. Based in part on the access opportunities, the controller 10 then establishes a priority order of the animals for which capacity should be reserved, in other words, for which additional feed can be provided after milking. The controller can then allocate the free capacity C to the dairy animals. Thus, starting with the animal with the highest combination of selection priority and access opportunities, the entire free capacity can be allocated until the free capacity is fully allocated by reducing the remaining feed allowance of the animal divided by the expected value of the feeding rate (animal-specific or average), repeating this action for the animal with the next highest combination of selection priority and access opportunities, and so on. It is also possible to allocate less than the full capacity, for example, leaving some flexibility in the case where all animals actually visit the robot, even if some milkings take a little longer and there is a (longer) waiting time at the end of the time period. Conversely, assuming that not all expected animals will participate in a visit to the robot, it is also possible to reserve only more than the free capacity, so that in any case, if necessary, all participating animals can be given additional feed. For example, 80% and 150% of the free capacity C can be used for allocation to dairy animals.

[0057] The animals to which a part of the idle capacity is allocated form a subgroup of the herd of dairy animals to which the controller can continue to provide feed after said milking. These animals receive additional feed after their milking is complete if they still have a remaining feed allowance.

[0058] The planned time period can now be started and the dairy animals visiting the robot are milked and fed, with or without additional feed, depending on the subgroup to which they are assigned. Then, the animals are received in the robot 10, milked and fed (with or without additional feed) until all animals have been processed. The whole method is then repeated for a new time period.

[0059] Optionally, as an improvement, the method is as follows Figure 3 Here, the time period T, the free capacity C and optionally the selection priority and / or the access opportunity are determined again (steps 100, 110, 115 and 116), after which the subgroup of cows to receive additional feed is determined (step 120). The method can now also be repeated after each milking and the values ​​adjusted if necessary and possible. One or more of the following can be considered here:

[0060] – Adjust time periods: If some of the time has passed, then there is less time left. All time periods can be switched in time, such as switching to the next hour, etc.

[0061] – Adjust the idle capacity C: Subtract the duration of the visit or visits so far from the time period and offset it against the idle capacity of the shorter time period. Or consider the available time as a time period that has been switched in time but is equally long, for example by one hour.

[0062] – Update feed allowances for all livestock and adjust accordingly. For some livestock, absolute numbers increase faster than others, so the expected value of remaining feed allowances may also change.

[0063] – If an animal has actually been milked, then in principle it should be removed from the sub-herd for the current time period. Maybe there is still capacity left for another animal? And so on.

[0064] In general, it should be noted that it is possible that a milking event takes much longer than expected, for example because the connection takes longer than expected, the milking cup is detached and reattached, or another failure occurs during the milking process. Then, it may be the case that the animal has been able to eat its entire feed allowance and does not need to be included in the "A" subgroup of animals that can receive additional feed. Also, the milking of animals in another subgroup may naturally affect the sorting, selection priority and even the access opportunities of animals from one subgroup of animals that should receive additional feed.

[0065] It will be clear that in using Figure 3 When improved methods are used, feeding of dairy animals can be more flexible, dynamic and precise, thereby optimally supporting animal health and milk production.

[0066] The embodiments shown are not intended to be limiting but are merely for illustration and explanation of the present invention. The scope of protection is determined by the appended claims.

Claims

1. A method for automatically milking and feeding a group of dairy animals by means of a robotic milking device, the robotic milking device being arranged in an area where the dairy animals can move freely and comprising: - a milking station for receiving dairy animals; - livestock identification equipment, the livestock identification equipment is used to determine the livestock identity of the dairy livestock; - a robotic milking device for fully automatically extracting milk from the dairy animal during milking; - a feeding device for providing feed to the dairy animal at the milking site during the milking period based on the feed allowance of the dairy animal; as well as - a controller configured to control the robotic milking device and to update the feed allowance for each dairy animal, Wherein, the method comprises the following steps: -Select a future time period; - determining the idle capacity of the robotic milking device as an expected value of the total time during the future time period that the robotic milking device will not be occupied by dairy animals; - determining a subgroup of said herd of dairy animals, said controller being able to continue to provide feed to said subgroup after said milking; - milking said dairy animal accessing said robotic milking apparatus and feeding it during said milking; and If dairy animals from the subgroup still have feed allowance after milking, they are additionally fed depending on the determined free capacity.

2. The method of claim 1, wherein: The step of determining the sub-group comprises a user inputting into the controller the identity of only a portion of the group of dairy animals.

3. The method of claim 1, wherein: The step of determining the subgroup is automatically performed by the controller.

4. The method of claim 1, wherein: The step of determining the subgroup comprises: - for each dairy animal, assigning a selection priority based on selection criteria including at least the expected value of said feed allowance after milking; and -Selection of dairy animals with the highest selection priority in numbers meeting predefined selection criteria.

5. The method of claim 4, wherein: The selection criterion comprises that the sum of the eating times of each dairy animal from the sub-group calculated by the controller is at most equal to the spare capacity or a predefined mathematical operation of the spare capacity, the eating times belonging to the expected value of the feed allowance after milking.

6. The method of claim 5, wherein: The mathematical operation is to multiply the idle capacity by a value between 0.8 and 1.

5.

7. The method according to any one of claims 4 to 6, wherein: The selection priority is assigned based in part on the probability that the dairy animal will be milked within the future time period.

8. The method according to any one of claims 4 to 6, wherein: The selection priorities are assigned based in part on one or more livestock characteristics other than the expected value of the feed allowance after milking.

9. The method of claim 8, wherein: The one or more livestock characteristics include number of days in lactation and / or number of lactations and / or milk production.

10. The method according to any one of claims 4 to 6, wherein: The robotic milking apparatus comprises an upstream waiting area, wherein the controller is configured to determine an animal identity of the dairy animal in the upstream waiting area, and wherein the selection priority is assigned depending in part on the presence of the dairy animal in the upstream waiting area.

11. The method of claim 10, wherein: The robotic milking apparatus further comprises a dairy animal locating apparatus configured to determine a position of the dairy animal in the upstream waiting area, and wherein the selection priority is assigned based in part on the determined position in the upstream waiting area.

12. The method of claim 11, wherein: The selection priority is assigned based in part on the distance from the determined location to the robotic milking device.

13. The method according to any one of claims 1 to 3, wherein: The spare capacity is determined based on historical values.

14. The method according to claim 13, wherein: The historical values ​​comprise measured idle times and / or access frequencies of the robotic milking device, and / or planned cycles for cleaning or maintenance of the robotic milking device, in each case determined as a function of the time of day.

15. The method according to any one of claims 1 to 3, wherein: The future time period is a time period until a planned cleaning or maintenance in the future, or a predefined fixed time period.

16. The method according to claim 15, wherein: The predefined fixed time period is 1 hour.

17. The method according to any one of claims 1 to 3, further comprising repeating the method according to any one of the preceding claims after a predefined number of milkings or visits of dairy animals to the robotic milking device, or after a predefined time.

18. A robotic milking device for milking a group of dairy animals, comprising: - a milking station for receiving dairy animals; - livestock identification equipment, the livestock identification equipment is used to determine the livestock identity of the dairy livestock; - a robotic milking device for fully automatically extracting milk from the dairy animal during milking; - a feeding device for providing feed to the dairy animal at the milking site during the milking period based on the feed allowance of the dairy animal; as well as - A controller configured to control the robotic milking device and to update the feed allowance per dairy animal and to perform the method of any one of claims 1 to 9 and 13 to 17.

19. The robotic milking apparatus of claim 18, further comprising an upstream waiting area, wherein: The controller is configured to perform a method as claimed in any one of claims 1 to 17 and within the scope dependent upon claim 10 .

20. A robotic milking apparatus as claimed in claim 19, wherein: The robotic milking device is arranged in an area where the dairy animal can move freely, and also includes a dairy animal positioning device, which is configured to determine the position of the dairy animal in the area, and wherein the controller is configured to perform the method as described in any one of claims 1 to 17 and within the scope of claim 11 or 12.

Images