Packaging machine for packing products into packages

Abstract

Packaging machine for packaging products in packages, the packaging machine comprising a transfer device for transferring a product to a package, a recognition device for recognizing data indicative of a relative position of the inserted product and the package, a correction device for changing a relative position of an inserted product and the package, and a control unit, wherein the control unit is designed to control the correction device based on the indicative data for changing the relative position of the inserted product and the packaging.

Description

[0001] The present invention relates to a packaging machine for packaging products in packages according to claim 1 and to a method for changing a relative position of a product placed in a package and the package in a packaging machine according to claim 9. State of the art

[0002] Packaging machines for packaging products in packages are well known in the art. These typically include a transfer device, such as an insertion station, for inserting the products into packages such as packaging trays. The products can be, for example, one or more slices of sausage or cheese, or larger product pieces such as cheese cubes can be inserted into packages as a product.

[0003] The packaging with the inserted product is then usually sealed by placing a sealing film on the packaging with the product inside and sealing it at a predefined sealing seam, thus creating a gas-tight connection between the packaging and the sealing film.

[0004] However, misalignments between the product and the packaging can occur when inserting the product. For example, the packaging may not exactly match the assumed position of the packaging, or the product itself may not be inserted in the desired position. Particularly with disc-shaped products, this can result in one or more discs of the product protruding or overlapping the packaging, obscuring the intended seal. This can lead to incomplete and therefore inadequate sealing of the packaging during the subsequent sealing process, resulting in rejects.

[0005] To minimize waste, it is common practice to manually monitor the correct insertion of products into the packaging and, if necessary, to manually align the products within the packaging through operator intervention. However, this approach is prone to errors, meaning operators can inadvertently overlook or misjudge product misalignments. While this approach can reduce waste, it still involves an error rate that depends primarily on the production speed and operator attention. Task

[0006] Based on the known state of the art, the technical problem to be solved is to specify a packaging machine for packaging products in packages and a method for changing a relative position of a product placed in a package and the package in a packaging machine, which achieve a reduced reject rate. Solution

[0007] This object is achieved according to the invention by the packaging machine for packaging products in packages according to claim 1 or the method for changing a relative position of a product inserted into a package and the package in a packaging machine according to claim 9. Advantageous developments of the invention are covered in the subclaims.

[0008] The packaging machine according to the invention for packaging products in packages comprises a transfer device for transferring a product to a package, a recognition device for recognizing data indicative of a relative position of the inserted product and the package, a correction device for changing a relative position of an inserted product and the package, and a control unit, wherein the control unit is designed to control the correction device based on the indicative data for changing the relative position of the inserted product and the packaging.

[0009] The indicative data can, in particular, be or include image data or position data of the packaging and / or the inserted product, so that preferably an identification of the position of the product relative to the position of the packaging is possible based on the indicative data (particularly preferably based solely on the indicative data). Preferably, the three-dimensional position of the product in space relative to the position of the packaging in space can be recognized by the recognition device, or the indicative data are indicative of this position.

[0010] According to the invention, the correction device is designed such that it can move at least the product or the packaging or both, for example by lifting the product and moving the packaging, or by changing the position of the product relative to the packaging during the lifting.

[0011] The control unit may be designed as a computer with a processor and associated memory and may have suitable programming to use the indicative data to generate control commands for the correction unit and thus to change the relative position of the inserted product and the packaging.

[0012] In the following, packaging is understood to mean a packaging tray or at least any type of packaging that is not yet closed or sealed, into which a product (such as one or more slices of sausage or cheese or chunky products such as cheese cubes) can be inserted.

[0013] This packaging machine enables reliable subsequent correction of the position of the inserted product within the package, further minimizing waste due to a subsequent, improper sealing or closing process. At the same time, the susceptibility to errors is reduced, as manual operator intervention is no longer required.

[0014] It can be provided that the control unit is designed to control the correction unit based on the indicative data and a target relative position of the inserted product and packaging.

[0015] The target relative position can either specify a target relative position of the product relative to the packaging in a plane, for example, in the transport plane or a plane of the packaging parallel to it, or it can specify a three-dimensional relative arrangement of the product relative to the packaging. By comparing the target relative position and the indicative data, it can then be determined whether and to what extent the product's position relative to the packaging needs to be adjusted, and the correction unit can be controlled. This can be achieved using known deterministic algorithms that do not use a standard network, thus reducing waste with less programming effort.

[0016] It can be provided that the control unit is designed to determine a necessary change in position of an inserted product and to control the correction device based on the necessary change in position.

[0017] A necessary position change is understood to be the change in the position of the inserted product that moves the product from its current position to a predetermined position, such as the target relative position of the inserted product relative to the packaging, so that the packaging can be reliably closed in a subsequent sealing or closing step. To determine the necessary position changes, one can, for example, take into account the minimum position change required to achieve reliable closing or sealing of the packaging. It is not absolutely necessary that this necessary position change moves the product into an exactly ideal position, but only changes its position to the extent that reliable closing is possible. This embodiment also further avoids waste.

[0018] In one embodiment, the control unit comprises a neural network trained to determine the necessary position change.

[0019] The neural network can, in particular, be a neural network designed to recognize patterns in images or corresponding indicative data indicating the relative position of objects. The neural network can be trained to determine the necessary position change by processing the indicative data, for example, by comparing the indicative data with known, correctly positioned products in packaging, for example, by forming a difference image between the recognized product and correctly positioned products in packaging known to the neural network and processing it by the neural network. The necessary position change can then be determined as an output through the processing in the neural network.Because neural networks are trained using a large amount of training data for pattern recognition, this can further increase the accuracy in determining the necessary position change and also the accuracy in detecting incorrectly positioned products, thus further reducing waste. The use of a neural network is particularly advantageous for deformable products, such as sausage slices, because an incorrectly positioned product is often bent or deformed, making determining the necessary position change more difficult. This can be taken into account during training of the neural network, so that the effects of incorrect positioning of a product relative to the packaging are also taken into account when determining the necessary position change.

[0020] The detection device may comprise a camera or a laser scanner.

[0021] The camera can be a two-dimensional camera or a 3D camera, which is designed to capture two-dimensional or three-dimensional images. Determining two-dimensional images generates less data to process, whereas determining three-dimensional images can be advantageous when the position of the product relative to the packaging in space must be determined. A laser scanner is particularly advantageous for determining data on the three-dimensional, relative arrangement of the product and the packaging. These configurations of the detection device ensure reliable detection of the actual position of a product, which makes it possible to subsequently determine any necessary position change even more precisely, thus further reducing waste.

[0022] In one embodiment, the correction device comprises a multi-axis robot. The multi-axis robot can, in particular, be a collaborative robot and / or a six-axis collaborative robot, or it can be designed as a linear unit with an adjustable gripper. This embodiment enables precise and reliable relative movement of the product to the packaging, allowing incorrectly positioned products to be repositioned with great precision, thus further reducing waste.

[0023] The packaging machine may comprise a transport device for transporting packages with inserted products along a transport direction, and wherein the transport device runs at least partially along the correction device. The transport of the packages in the transport device can, in particular, be continuous, so that the relative position of the product and the package can be changed during continuous operation, which increases the machine's throughput while simultaneously reducing waste.

[0024] The packaging machine may comprise a displacement device for displacing the correction device relative to the transport device along at least one dimension.

[0025] It can be provided that the correction device can be moved by the traversing device along the transport device, in particular parallel to the transport direction, at a speed that is equal to the transport speed of the packaging with the inserted product. This embodiment reduces the relative movement between the correction device and the packaging with the inserted product, which enables highly precise adjustment of the product position, thus further reducing waste.

[0026] According to the invention, a method for changing a relative position of a product inserted into a package and the packaging in a packaging machine is further provided, wherein the packaging machine comprises a transfer device for transferring a product to a package, a recognition device for recognizing data indicative of a relative position of the inserted product and the packaging, a correction device for changing a relative position of the inserted product and the packaging, and a control unit, wherein the control unit controls the correction device based on the indicative data for changing the relative position of the inserted product and the packaging.

[0027] This process can reduce waste due to incorrectly positioned products.

[0028] It can be provided that the control unit controls the correction unit based on the indicative data and a target relative position of the inserted product and the packaging. This embodiment allows, in particular, dimensionally stable products to be reliably repositioned.

[0029] The control unit can determine the necessary position change of the inserted product and control the correction device based on the required position change. By determining the necessary position changes, which can particularly define a two-dimensional or three-dimensional translation of the product, reliable repositioning is possible, thus reducing waste.

[0030] The control unit can comprise a neural network trained to determine the necessary position change, with the neural network determining the necessary position change. Especially with shape-changing products, neural networks can detect incorrect positioning of the product relative to the packaging and reliably determine the necessary position change, thereby further reducing waste.

[0031] The detection device may comprise a camera or a laser scanner and / or the correction device may comprise a multi-axis robot.

[0032] These devices enable precise determination of the position of the product relative to the packaging and precise changes to the position of the product relative to the packaging, which can further reduce waste.

[0033] The packages with inserted product can be transported along a transport direction by a transport device of the packaging machine, and the transport device can run at least partially along the correction device. The transport device can, in particular, be operated continuously, which increases the throughput of the packaging machine while simultaneously reducing waste.

[0034] The packaging machine can comprise a displacement device for displacing the correction device relative to the transport device along at least one dimension, and the method can comprise a displacement or correction device dependent on a position of the package in the transport device. This can cause the correction device to move along with the packages and products, thus reducing the relative movement between the correction device and the product, which can improve the result of repositioning a product.

[0035] All embodiments described here can be combined with each other. Short description of the characters Fig. 1 shows a packaging machine according to an embodiment. Fig. Figure 2 shows schematically the change in the position of the product relative to the packaging. Fig. Figure 3 shows a flow chart of a training process for a neural network to determine the necessary position change. Detailed description

[0036] Fig. 1 shows a schematic view of a packaging machine 100 for packaging products 131 in packages 130.

[0037] The packaging machine can, in principle, be designed in a conventional manner and, in particular, comprise a transport device 105 for transporting the packages 130 and a transfer device 107 for transferring products 131 to the packages 130. The configurations of the transport device 105 and the transfer device 107 are, in principle, arbitrary.

[0038] It can be provided that the transport device 105 is designed as a conveyor belt (such as an endless rubber conveyor belt or a conveyor chain) for at least single-row and optionally multi-row transport of packages 130. Packaging 130 is understood below to mean a packaging trough or at least any type of not yet closed or sealed package into which a product 131 (such as one or more sausage slices or cheese slices or chunky products such as cheese cubes) can be inserted.

[0039] The transfer device 107 can, for example, comprise a conveyor belt or other product conveying device that is inclined substantially in the direction of transport T of the transport device. This conveyor belt can comprise one or more drivers and / or retaining elements (not shown here) so that the product 131 does not accidentally slide down towards the transport device 105 due to the inclined plane. Alternatively or additionally, it can be provided that a coefficient of static friction between the product 131 and the contact surface of the transfer device with the product is selected to be as large as possible, so that a transition to sliding of the product on the transfer device is avoided as far as possible.

[0040] Downstream of the transfer device 107 in the transport direction T of the packages, a sealing station 104 can be arranged, which can apply a sealing film 141 to a package 130 filled with product and, for example, bond it to the package 130 by applying heat, thereby sealing the package. The design of such a sealing station is generally known to those skilled in the art.

[0041] In order to be able to detect any misalignments of the product relative to the packaging, as shown schematically here with the packaging 132, in which the product lies crookedly in the packaging and protrudes from it, the invention provides that the packaging machine 100 comprises a detection device 101 downstream of the transfer device 107. The detection device 101 is designed to detect or generate indicative data for a relative position of the inserted product and packaging. The indicative data can be image data, for example. For example, the detection device 101 can comprise a camera directed in the direction of the packaging transported in the transport device, which camera can record images of the packaging and the product inserted therein.The images can represent the indicative data, or the indicative data can be obtained through further processing of the images (e.g., changing the contrast or scaling). The camera can, for example, be designed as a camera for capturing two-dimensional images, so that a two-dimensional image of the packaging with the inserted products is created from above. Alternatively, the recognition device can comprise a 3D camera, which can preferably also generate a three-dimensional image of the packaging and the product inserted therein from above the transport device. This three-dimensional image can then represent the indicative data.

[0042] Alternatively, the recognition device 101 may comprise a laser scanner or a device designed to scan the three-dimensional structure of the packaging and the inserted product. This generates indicative data in the form of data representing the contour of the packaging and the inserted product. Since other information, such as the color of the packaging and the product, is not important for determining correct positioning between the product and the packaging, this embodiment may be particularly preferred, since processing of the corresponding data, for example, with the aid of a neural network, can be carried out particularly reliably.

[0043] According to the invention, the packaging machine 100 further comprises a control unit 180, which can be embodied, for example, as the central control unit of the packaging machine and can generally be configured as a computer with a processor and associated memory. The control unit is further configured to receive the indicative data from the detection device.

[0044] The packaging machine further comprises a correction device 102. The correction device 102 is designed to effect a change in the relative position / arrangement between a product 131 inserted into the packaging and the packaging 130. For this purpose, the correction device 102 can comprise, for example, a gripping element or a suction cup or a similar device with which the product can be gripped and lifted and moved at least in a plane parallel to the transport device. Alternatively or additionally, it can also be provided that a spatial movement of the inserted product is possible through the correction device 102.While the movement of the product is basically described here, it can also be provided additionally or alternatively that the correction device 102 lifts the product and can move the packaging relative to the product or that the product and packaging can be moved relative to each other by the correction device.

[0045] The design of the correction device is fundamentally unrestricted, as long as it can be used to change the relative position between the packaging and the product. However, preferred embodiments of the correction device include at least one collaborative and / or multi-axis robot, allowing the position of the product relative to the packaging to be changed as seamlessly as possible with high precision. The robot can be designed, in particular, as a six-axis robot. The use of collaborative robots or so-called co-bots can also be advantageous, as these can operate particularly efficiently, completely independently, and automatically.

[0046] According to the invention, it is further provided that the control unit 180 is designed to control the correction device 102 depending on the indicative data such that a position of an inserted product, such as the incorrectly positioned product here, in the packaging is changed before the packaging with the inserted product is fed to the sealing station. In particular, it can be provided that the control unit comprises a deterministic algorithm or a trained neural network in order to determine a necessary change in the position of a product from its actual position in the packaging, which can be determined from the indicative data, to a desired relative position that the product should assume before the packaging is sealed. The desired relative position can, in particular, be a desired relative position between the packaging and the product.This target relative position can either be a specific position or encompass a range that specifies the permissible positioning of a product within the packaging, particularly within the packaging cavity. This range can, for example, be the entire packaging cavity, as long as it is ensured that the inserted product does not protrude beyond the edge of the packaging cavity, as this could cause problems during subsequent sealing.

[0047] The control unit 180 is preferably designed to determine the necessary position change using the algorithm or neural network 181 and can derive control commands for the correction device therefrom in order to control the correction device based on the control commands, so that the correction device 102 changes the position of the product relative to the packaging.

[0048] Since a change in the position of a product within a package requires a certain amount of time, one embodiment can provide for the transport device to operate in cycles, so that packages with the product inserted therein come to a standstill relative to the correction device, and the correction device can make any necessary changes to the position of one or more products in the packages in each cycle. Since there is no relative movement between the correction device as a whole and the products, simpler control of the correction device is possible.

[0049] Alternatively, it can be provided that the correction device is movable along a direction by means of a displacement device 106, wherein the direction can preferably comprise at least one direction parallel or substantially parallel to the transport direction T. It is preferred if the displacement device 106 can move the correction device 102 parallel to the transport direction T and at the same speed. This also reduces or completely compensates for the relative movement between the packaging and the inserted product and the correction device 102. The displacement device can in particular comprise one or more rails on which the correction device can be movably arranged. In one embodiment, the displacement device can extend alongside the transport device or, alternatively, can be arranged above the transport device.

[0050] Alternatively or additionally, it can also be provided that the displacement device 106 can also move the correction device in a direction transverse to the transport direction T. This allows either multiple lines (each with its own transport devices) of different packaging machines to be served or, within a packaging machine, for example, even packages transported in multiple rows can be reliably reached by the correction device without the correction device itself having to be designed with long lever arms. This can improve the stability of the correction device and / or the accuracy with which the correction device can change the position of packages relative to the product inserted therein.

[0051] It can be provided that the path along which the correction device can carry out changes to products placed in packages is selected in such a way that a change in the position of products placed in several consecutive packages or in several packages transported in the same row is also possible before they reach the sealing stations 104.If, for example, the average time for changing the position of a product within a package is one second, and it is intended that up to five packages transported in a row transversely to the transport direction are processed by the correction device, it can be provided that the length of the transport path that is accessible to the correction device and / or the transport speed of the packages in the transport device are selected such that the correction device has at least six seconds available to reposition products in the packages.This ensures that even in the event that, for example, a temporary malfunction in the transfer device or another device of the packaging machine 100 results in a significant number of incorrectly positioned products in packages, this deviation can be reliably reduced, thus also reducing waste.

[0052] In general, it can be provided that an available interaction length of the correction device with packaging within the transport device is selected such that it is at least equal to the product of the transport speed of the packaging in the transport device with the average time required to change the position of a product and the maximum number of products to be repositioned in a given group of packaging or is greater than this product.

[0053] Fig. 2 shows a plan view of a package 230 and a product 240 inserted into a packaging recess 231 of the package 230, as well as a process of correcting the position of the product 240 relative to the package 230.

[0054] In the left illustration of the Fig. 2 that the product 240 is not arranged in the packaging recess 231, but partially protrudes from it. The packaging recess is to be understood here as a depression that extends in an area below the edge 233 of the packaging 230, so that the product 240 partially rests on the edge 233 and partially extends into the packaging recess 231. The product 240 is in the Fig. The embodiment shown in Figure 2 is a flexible product, such as a slice of cheese or a slice of sausage. However, this is not to be understood as limiting. Resting at different heights, on the one hand on the edge 233 and on the other hand in the packaging recess 231, causes a deformation of the product 240, so that the shape of the product 240 deviates from its circular shape, which is assumed here only as an example.

[0055] In the left illustration of the Fig. 2, the product 240 rests on the edge 233 to such an extent that it at least partially covers a provided sealing seam 232. The provided sealing seam 232 is the part of the package in which Fig. 1, the sealing film is to be bonded to the packaging. If the product partially rests on this intended sealing seam 232, the packaging cannot be sealed reliably in this area, and the packaging will either not be sealed at all or will be sealed with inadequate quality, or will generally be closed.

[0056] With the packaging machine according to the invention according to the Fig. 1 and the indicative data, the control unit 180 can now determine a necessary change in position Δr of the product 240 from its initial position in the left representation of the Fig. 2 into a target relative position 241 in which the product is positioned in the packaging trough 231. This necessary position change Δr can then be used to control the correction device in order to change the position of the product 240 relative to the packaging 230 and thus transfer it into the packaging trough. It can be provided that the necessary position change Δr is determined based on the current position 240 of the product and precisely a predetermined position 241 of the product within the packaging trough. However, this places high demands on the change in the position of the product, although even a slight deviation from this would not have a significant negative impact on the sealing result as long as the product is completely arranged within the packaging trough.Therefore, it can alternatively be provided that the necessary position change Δr is determined such that every point of the product lies within a predetermined range (which can be understood as a target relative position), such as an area corresponding to the packaging recess. However, due to the possible deformation of the product, this does not necessarily equate to a translational movement in the image plane shown here. Instead, the determined, necessary position change Δr must also take into account that the shape of the product changes during the displacement, in particular that the expansion of the product in the image plane shown increases when it is completely displaced into the packaging recess.

[0057] This circumstance can be taken into account by suitable deterministic algorithms, which, for example, also include a model of the physical behavior of the product during deformation, so that the necessary change in position can be determined.

[0058] Alternatively, a specially trained neural network can be used to determine the necessary position change, which is trained to determine necessary transformations between two states. The first state here is the actual position of the product relative to the packaging according to the left representation of the Fig. 2. The second state is a position of the product completely within the packaging cavity as the target position, whereby the necessary change in position is to be determined taking into account a possible deformation of the product or the transformation is determined by the neural network, which transfers the position of the product 240 into the target position.

[0059] Corresponding neural networks, which can be used to determine transformations between two states, for example, between two images, taking into account certain properties, such as the physical properties of the objects within the image, are generally known to those skilled in the art. These neural networks can use the indicative data discussed above as input, for example, an image of the packaging with the product inserted inside, or corresponding point clouds captured using a laser scanner and depicting the three-dimensional structure of the packaging and the inserted product.

[0060] Furthermore, information on the product's target position relative to the packaging can optionally be used as input data. One advantage of using neural networks is the ability to enable the necessary position change even without precisely specifying a specific target position of the product within the packaging and without specifically specifying a physical model to describe the product's behavior under deformation. This is done by selecting the neural network's training data in such a way that the relevant information is implicitly provided to the neural network during training. The neural network is trained in such a way that it determines the necessary transformation, whereby a possible deformation of the product is also implicitly taken into account due to the training of the neural network.

[0061] This shows Fig. 3 a flow chart of a training of a neural network according to an embodiment to determine the necessary position change or the necessary transformation in order to change from the position of the product according to the left representation in Fig. 2 in the right representation of the Fig. 2 to be determined.

[0062] The procedure for training a suitable neural network according to the Fig. 3 begins with a first step 301, in which a training data set is provided. This training data set comprises a large number of indicative data, which can be formed, for example, from images of packaging and inserted products in different relative positions. This can involve several thousand, or tens of thousands, or hundreds of thousands of images, whereby the training data set can be obtained in a known manner from a smaller number of indicative data, such as images, by modifying the corresponding images, for example using generally known image processing programs, for example by rotating images or changing their contrast or color. A corresponding training data set can also be specifically recorded by creating a large number of images of packaging with products inserted therein in different positions.

[0063] The training dataset preferably, but not necessarily, includes statistically evenly distributed arrangements of products relative to the packaging. Some arrangements are faulty, for example, the product protrudes beyond the intended packaging recess or is otherwise not fully inserted into the packaging, making sealing impossible. Furthermore, the training dataset also includes indicative data that is indicative of the correct positioning of the product relative to the packaging.

[0064] Preferably, the training dataset consists of indicative data for exactly one type of packaging and exactly one type of product (e.g. sausage slices), such as the one in Fig.2 shows the packaging trough with the corresponding edge. Alternatively, if the use of a neural network is intended to determine a necessary position change in connection with a package, for example in the form of a bag, a separate training data set can be used for this purpose.

[0065] First, the training data is processed to generate a set of reference data 302 and a validation data set 303. This can be done in a conventional manner. The validation data set preferably comprises a subset of the training data set, such as 100 images or 1% or 0.1% of the total number of images or indicative data contained in the training data set and the associated necessary position changes of the product relative to the packaging, and is not used during the normal training process.

[0066] The reference data set 302 includes an assignment of the remaining training data with the respective necessary position changes. Additionally, a target relative position of the product in the packaging can be specified for the validation data set and the reference data set. However, this is not mandatory. One advantage of using the neural network is that the necessary position change can be determined even without specifying a target relative position, since the neural network is trained to calculate a necessary position change for the received indicative data.

[0067] In a next step 304, the training data (without the previously separated validation data) is processed by the neural network. All available training data or a subset or true subset of the training data can be processed in step 304. The neural network can be initialized and its parameters set to certain initial values, which, for example, experience shows will ensure that the neural network determines at least partially accurate or approximately accurate predictions regarding the necessary position change.

[0068] By processing the training data in the neural network, the neural network generates predictions 305 regarding the necessary position change of the respective products from the indicative data of the training data set. Using images as indicative data, this can include, for example, the neural network predicting a first necessary position change for a first image and a second necessary position change for a second image. These predictions depend on the arrangement of the product in the packaging and do not have to be identical. In particular, the training data set can also contain indicative information in which the product is correctly positioned in the packaging. In such a case, the necessary position change to be predicted would be zero.

[0069] After the neural network has processed the training data in steps 304 and 305 and predicted the necessary position changes, a comparison with the reference data is made in step 306. In this process, the necessary position changes predicted by the neural network for the respective indicative data are compared with the actual necessary position changes from the reference data and, as is usual in the training of neural networks, differences or other quantities indicative of a deviation from the reference data are determined.

[0070] Subsequently, the parameters of the neural network are updated in step 307, with an attempt being made to modify the parameters of the neural network for the next training cycle in such a way that the prediction accuracy increases. This can be done using known methods, such as gradient descent methods or similar, and is generally known to those skilled in the art.

[0071] If the neural network or its parameters were updated in step 307, a validation step 308 can follow, in which the neural network, using its current parameters, determines predictions of the necessary position changes for the validation data set that was not used during training. If the comparison with the validation data set shows that the neural network predicts the necessary position changes with sufficient accuracy, training can be terminated, and in step 310, the neural network becomes a trained neural network that can then be used in the packaging machine.

[0072] If the comparison with the validation data set shows that the neural network is not yet sufficiently trained, a further training cycle can be carried out in step 309 according to steps 304 to 307 and optionally 308.

[0073] Alternatively or additionally, it may also be provided that validation does not occur after each training cycle, i.e., after each pass of a training data set through the neural network. For example, it may be provided that if it is determined in step 307 that the parameter values, or at least the value of a parameter, changes from a first to a subsequent training cycle by a value greater than a predetermined limit, no validation step is performed, and the next training cycle is started without a validation step. This process is repeated until the parameters of the neural network no longer change or change so little that their change is below a limit.Once this is achieved, the validation dataset can be used to check the quality of the training and rule out the possibility that the neural network was not generally optimized to predict necessary position changes, but was specifically trained on the indicative data of the training dataset. If the latter is the case, known methods can be used to reverse this specialization of the neural network.

[0074] Once the neural network has been trained, it can be used in the control unit according to the previous embodiments to determine the necessary position change based on the indicative data. This does not require a specification of a target relative position and / or a special physical model of the product, which can reduce the information required during production. In one embodiment, it can also be provided that the neural network is further trained based on the determined necessary position changes during normal operation. For this purpose, it can be provided that a check, in particular an automatic check, of the sealing result is carried out at least for the packages for which a necessary position change other than 0 was determined by the neural network. If the sealing result is good, ormeets specified quality criteria, this can be used to adjust the parameter values ​​of the neural network, just as negative seal results that do not meet the requirements can be used to adjust the parameters of the neural network.

Claims

[1] Packaging machine (100) for packaging products (131) in packagings (130), the packaging machine comprising a transfer device (107) for transferring a product to a packaging, a detection device (101) for detecting indicative data for a relative position of the inserted product and the packaging, a correction device (102) for changing a relative position of an inserted product and the packaging, and a control unit (180), wherein the control unit is configured to control the correction device (102) based on the indicative data to change the relative position of the inserted product (131) and the packaging (130). [2] Packaging machine (100) according to claim 1, wherein the control unit is configured to control the correction unit based on the indicative data and a target relative position (241) of the inserted product (131) and packaging (130). [3] Packaging machine (100) according to claim 1 or 2, wherein the control unit is configured to determine a necessary change in position (Δr) of an inserted product (131) and to control the correction device (102) based on the necessary change in position (Δr). [4] Packaging machine (100) according to claim 3, wherein the control unit (180) comprises a neural network (181) trained to determine the necessary change in position (Δr). [5] Packaging machine (100) according to one of claims 1 to 4, wherein the detection device (101) comprises a camera or a laser scanner. [6] Packaging machine (100) according to one of claims 1 to 5, wherein the correction device comprises a multi-axis robot. [7] Packaging machine (100) according to any one of claims 1 to 6, wherein the packaging machine (100) comprises a transport device (105) for transporting packages (130) with product (131) inserted along a transport direction (T) and wherein the transport device (105) runs at least partially along the correction device (102). [8] Packaging machine (100) according to claim 7, wherein the packaging machine (100) comprises a traversing device (106) for traversing the correction device (102) relative to the transport device (105) along at least one dimension. [9] Method for changing the relative position of a product (131) placed in a package (130) and the package in a packaging machine (100), the packaging machine (100) comprising a transfer device (107) for transferring a product to a package, a detection device (101) for detecting indicative data for a relative position of the product placed and the package, a correction device (102) for changing a relative position of the product placed and the package, and a control unit (180), wherein the control unit (180) controls the correction device (102) based on the indicative data to change the relative position of the product placed (131) and the package (130). [10] Method according to claim 9, wherein the control unit (180) controls the correction unit (102) based on the indicative data and a target relative position (241) of the inserted product (131) and the packaging (130). [11] Method according to claim 9 or 10, wherein the control unit (180) determines a necessary change in position (Δr) of the inserted product and controls the correction device (102) based on the necessary change in position (Δr). [12] Method according to claim 11, wherein the control unit (180) comprises a neural network (181) trained to determine the necessary position change and wherein the neural network determines the necessary position change (Δr). [13] Method according to any one of claims 9 to 12, wherein the detection device (101) comprises a camera or a laser scanner and / or wherein the correction device (102) comprises a multi-axis robot. [14] Method according to any one of claims 9 to 13, wherein the packagings (130) with product (131) inserted are transported along a transport direction (T) by a transport device (105) of the packaging machine (100) and wherein the transport device (105) runs at least partially along the correction device (102). [15] Method according to claim 14, wherein the packaging machine (100) comprises a traversing device (106) for traversing the correction device (102) relative to the transport device (105) along at least one dimension and wherein the method comprises a method or correction device (102) depending on a position of the packaging (130) in the transport device.

Images