Device for handling objects

Abstract

The invention relates to a device (8) for handling objects (14), comprising a grouping device (10) having at least one positioning robot (16; 16a, 16b, 16c) in order to position the objects (14) in a specified mutual end position (24). The grouping device (10) is associated with a position correction device (30) having the following features: - a camera (32) directed at the objects, and - an evaluation logic (34) in order to compare the position of each object (14) detected by the camera (32) with target values and generate a correction signal, from the comparison result, for a corrective actuation of the positioning robot (16; 16a, 16b, 16c), said evaluation logic (34) detecting and evaluating the corrective actuation resulting from the correction signal by means of the camera (32) and accordingly modifying the correction signal associated with the comparison result or leaving the correction signal unchanged.

Description

[0001] Device for handling objects

[0002] The present invention relates to a device for handling objects, comprising a grouping device with at least one positioning robot designed to position objects in a predetermined mutual end position. Such devices are found, for example, in the beverage industry to position beverage bottles in a group for a palletizable layer so that they can be further processed into a layered arrangement.

[0003] In such known devices for handling objects, in particular in devices for packaging and / or palletising piece goods such as parcels, containers or the like, these are often first conveyed on conveyor systems conveying in lines and are moved, aligned and / or assembled in a suitable manner in order to produce desired layer patterns which can then be stacked several times on top of one another, for example on pallets prepared for this purpose.

[0004] These processing steps can be particularly useful in systems for treating beverage containers. The piece goods in question can be, for example, packages, crates, cartons, containers, or clusters. To ensure that the aforementioned pallets are safe for transport, the assembled layer patterns, also known as assembled cycles, must meet certain requirements. Traditionally, preparatory measures are necessary to create such cycles. These include grouping or collecting the piece goods, which are initially transported regularly or step by step on a so-called allocation belt, on an intermediate conveyor belt, from where they are transferred in collected and / or grouped form to a layer-forming belt or layer-forming table.

[0005] In such devices for handling objects, especially piece goods, it can happen that, for example, after a format changeover to a different piece goods, an object, especially a piece goods with a new format, is not positioned correctly. For example, it may be slightly twisted or translationally offset in at least one direction relative to its target end position. Such positioning errors can also occur during the positioning of the object, for example, if an object is not gripped correctly by the gripper of a positioning robot.

[0006] It is therefore an object of the invention to provide a device of the above-mentioned type in which a misalignment of objects is corrected with a high positional accuracy.

[0007] This object is achieved by a device according to claim 1 and by a method according to claim 13. Advantageous further developments are the subject of the associated dependent claims. Preferred embodiments of the invention are also described in the description and in the figures.

[0008] In addition to the grouping device, the handling device can also have a conveyor device for delivering the individual objects and / or a group handling device for removing the grouped objects, which are preferably adapted to interact with the grouping device.

[0009] According to the invention, the grouping device is assigned a position correction device that contains a camera directed at the objects positioned or to be positioned by the grouping device, in particular covering the effective range of the positioning robot. Furthermore, the position correction device contains evaluation logic that is designed to compare the position of each object detected by the camera during positioning and / or its end position with target values ​​and to derive a correction signal for corrective actuation of the at least one positioning robot from the comparison result. Thus, if a deviation of the position from the target values ​​is detected during positioning or when the object is positioned in its end position, a correction signal for corrective actuation of the at least one robot is derived from the comparison result of the actual values ​​(current position) with the target values.The present invention goes one step further in correcting the position error by designing the evaluation logic to capture the corrective action of the positioning robot resulting from the correction signal using the camera and to evaluate this in turn with regard to compliance with the target values. Based on the evaluation, the correction signal assigned to the comparison result is then either retained if the target values ​​are met, or the correction signal is modified if, based on the current position data of the object received from the camera, it is determined that even the corrective action of the positioning robot does not lead to exact positioning of the object. The advantage of this self-monitoring is that the correction signals assigned to a comparison result become increasingly more precise.

[0010] Generally, each comparison result corresponds exactly to one positioning error, which can, however, be composed of various error types. For example, the position of an object can deviate from its target position both rotationally and translationally. It is essential that, depending on this comparison result, a correction signal is assigned, which is designed to correct the positioning error(s). The correction signal thus influences the movement and / or the movement sequence of the positioning robot when handling the object. By evaluating the corrective action of the positioning robot and making a corresponding change to the correction signal, very precise positioning results are obtained in a quasi-self-learning manner.

[0011] This self-learning effect can be further enhanced if the evaluation logic has a memory in which the comparison results and associated correction signals are stored with a timestamp. The evaluation logic is thus able to record how the correction signals associated with a comparison result have changed over time. This time history then enables the use of optimization algorithms, such as genetic algorithms, self-learning algorithms, or any type of artificial intelligence, to further optimize the associated correction signal.

[0012] Another possibility for evaluating the temporal sequence of the correction signals assigned to a comparison result is to generate a maintenance signal or warning signal if this temporal sequence conforms to or does not conform to predetermined principles. Within the framework of the inventive self-optimization capability of the positioning, it can be expected, for example, that the necessary changes to a correction signal for a given incorrect position, which is correlated with a comparison result, will become increasingly smaller. If it becomes apparent over time that the correction signal repeatedly needs to be changed quite radically, it can be assumed that certain parameters, in particular machine parameters, are present that do not allow precise positioning of the object. This could be due to wear and tear of components of the positioning robot or a fault in the electronics.This advantageous embodiment of the invention thus enables preventive detection of machine problems that prevent precise positioning of the object during grouping.

[0013] The evaluation logic can use, for example, an assignment algorithm or assignment tables to derive a correction signal from a comparison result. This assignment can preferably also take other parameters, such as temperature, humidity, or other machine parameters of the positioning robot, into account. This also leads to significantly more precise positioning, taking into account prevailing influences such as the ambient temperature or humidity or the age of the positioning robot's machine components.

[0014] The following parameters are ideal for machine parameters: the type of positioning robot, the age of the positioning robot, the age of components of the positioning robot, the maintenance status of the positioning robot or its components. All of these parameters can influence the accuracy of positioning and can be included in the correction signal when deriving a correction signal from the comparison result corresponding to a defined position error. The correction signal itself can be a complex control signal for the positioning robot or, for example, a simple control signal that causes the control of the positioning robot to access existing, stored movement sequences. The position setpoints that are used for comparison with the actual values, ieThe current object position can be used to determine the current position, for example, of the robot gripper, during the handling of an object and / or the final position of the object during group formation. This is relevant in that not only the final position of the object can be corrected after grouping, but also the movement sequences of the positioning robot during positioning, which in turn also has an impact on the accuracy of positioning the object in the final position.

[0015] Preferably, therefore, the correction signal is correlated with a control algorithm for controlling the movement sequence of the positioning robot, whereby changes in the correction signal can have an impact on the movement sequence of the positioning robot when positioning the objects.

[0016] The present invention is, in principle, suitable for a wide variety of applications. However, the invention is preferably used for a grouping device in a beverage line, where the task is to group beverage bottles or other beverage containers in such a way that they can be repackaged, for example, into beverage crates and other containers and packaging units. For example, individual objects can be arranged into groups, for example, in a 4 x 3 matrix, which can then be simultaneously transferred into a beverage crate by a multi-gripper in one movement sequence.

[0017] The handling device preferably comprises a conveyor device upstream of the grouping device, which is designed to convey the objects in a push-pull manner into the effective range of the at least one positioning robot. This supports the optimization of positioning accuracy according to the invention in that the push-pull conveyance minimizes a rotational offset of the objects and allows the gripper of the at least one positioning robot to grip the objects with greater angular precision. Any rotational offset is thus minimized, whereby the positioning optimization according to the invention can start at a completely different level of accuracy. The invention also relates to a method for grouping objects by means of a grouping device having at least one positioning robot. The method includes the following method steps, which are carried out successively one after the other.First, the position of each object is compared with its target position during and / or after positioning. From this comparison of the actual position with the target position, a correction signal for the corrective actuation of the at least one positioning robot is derived. This corrective actuation resulting from the correction signal is then recorded by the camera, and the recorded corrective actuation of the positioning robot is evaluated with regard to compliance with the target position of the object. Based on the evaluation, the correction signal assigned to the comparison result is then modified or not modified. The method according to the invention thus provides a self-learning optimization of an overall positioning process by checking, evaluating, and, if necessary, correcting the correction process for position errors of the object.By changing the correction signal assigned to the comparison result, the correction of a defined error type during positioning becomes increasingly more precise.

[0018] It is particularly advantageous if the comparison results and the associated correction signals are saved with a timestamp. The evaluation logic is then able to evaluate the temporal sequence of the correction signal changes. This temporal sequence can then be optimized, for example, using an optimization program such as a genetic algorithm, self-learning algorithms, or artificial intelligence. The optimized correction signal is then verified in its execution by the positioning robot.

[0019] Information about the state of the positioning robot can also be obtained from the time history of comparison results and associated correction signals. This can be seen when changing the correction signals to a defined comparison result, i.e., a defined incorrect position, does not result in increasingly small changes in the correction signal, but rather the correction signal must be repeatedly and substantially changed over time. This suggests that there must be some problem with the positioning robot, for example, high positioning tolerances, worn bearings, or worn-out positioning mechanisms. The invention can thus also be used to preventively maintain the positioning robot or its components in a state that always allows for precise positioning.Preferably, additional parameters such as ambient temperature, humidity, or specific machine parameters of the positioning robot are taken into account when deriving a correction signal from a comparison result. These parameters can also be incorporated by an optimization algorithm in such a way that the most accurate positioning results are obtained in a reproducible manner.

[0020] It is obvious to the person skilled in the art that the features of the device and the method described above can be combined with each other in any way, provided that combined features do not exclude each other.

[0021] The invention will now be described, by way of example, with reference to the schematic drawing, in which:

[0022] Fig. 1 is a plan view of a grouping device with the

[0023] Representation of two misalignments of objects, namely a rotational error and a translational positioning error, and

[0024] Fig. 2a and 2b a schematic plan view of the correction of the position of an object during positioning.

[0025] Fig. 1 shows a handling device 8 with a grouping device 10 and a conveyor device 12 arranged on its input side for objects 14, e.g. beverage bottles in the beverage and bottling industry. The handling device 8 is designed in particular for producing a grouping of several objects 14 into a palletizable layer or partial layer. The end of the conveyor device 12 projects into the effective range of a positioning robot 16 of the grouping device 10, which has a multi-linked arm 18 movable with several degrees of freedom, at the end of which a gripper 20 is articulated for gripping one object 14 at a time. The positioning robot 16 is controlled by a controller 22 of the grouping device 10. The objects 14 are positioned by the positioning robot 16 in a predetermined end position 24 relative to one another, in the present case a 4 x 4 matrix.From this end position 24, the entire object group is then grasped and further processed or transported, for example. The image shows two misalignments in the predefined matrix of the end position 24: a translational offset 26 of the object in the xy coordinate system of the matrix and a rotational position error 28.

[0026] The controller 22 contains a position correction device 30 connected to a camera 32 that covers the effective range of the positioning robot 16 and, in particular, the storage area of ​​the end position 24 of the objects. The position correction device 30 further contains an evaluation logic 34, which in turn includes a memory 36 for storing data.

[0027] The memory 36 stores target values ​​for the end positions of the object at the storage location 24. Furthermore, a large number of comparison results corresponding to different types of positioning errors are stored in the memory. These comparison results are stored with associated correction signals, which cause the positioning robot 16 to correct the positioning error. Changes to the correction signals associated with a comparison result are preferably stored over time, so that the evaluation logic 34 can optimize the correction signals using a genetic algorithm, self-learning algorithms, or K1, leading to increasingly precise positioning of the objects 14.

[0028] During current operation, the camera 32 captures all objects 14 in the intended end position 24. The evaluation logic 34 compares each current object position in the end position 24 with target values. In this case, the evaluation logic detects a translational misposition at point 26 and a rotational misalignment at point 28, which are correlated with corresponding defined comparison results from the target / actual value comparison. Based on the comparison result, the evaluation logic 34 now sends an assigned correction signal to the controller 22 of the positioning robot for correcting the objects in the misalignment at points 26 and 28. The correction signal and the corresponding corrective action of the positioning robot 16 correct the misalignment of the two incorrectly positioned objects 14 at points 26 and 28.

[0029] This corrective action of the positioning robot 16 and its gripper 20 resulting from the correction signal is recorded by the camera 32 and evaluated with regard to compliance with the target values ​​for the corresponding object positions. Based on the evaluation, the correction signal associated with the comparison result is modified, namely if the action of the positioning robot continues to lead to unsatisfactory position data of the objects at the positions 26, 28 to be corrected in the end position 24. The correction signal is retained if the corrective action of the positioning robot is successful, i.e., leads to precise positioning of the object according to the target values ​​within specified tolerances or limit values.

[0030] The evaluation logic 34 preferably contains an assignment unit, e.g., in the form of an assignment table and / or an assignment algorithm. Changes in the correction signal resulting from the adjustment of the corrective action of the positioning robot are written into this assignment unit, correlated with the comparison result, which is preferably stored in the memory 36 of the evaluation logic 34. The grouping device thus automatically detects, based on the detection and monitoring of the corrective action, an automatic change in the correction signals derived from an incorrect position, so that this system performs increasingly more precise positioning. Because the correction signal assigned to a comparison result can be continuously corrected, the positioning accuracy for correcting a defined incorrect position associated with a defined comparison result is continuously and automatically improved.

[0031] Furthermore, the evaluation logic can perform optimization based on the temporal history of changes in the correction signal using genetic algorithms or self-learning programs, which leads to an improvement in the correction signal resulting from a comparison result or correlated with the corresponding correction actuation, even without the positioning robot currently being activated. io

[0032] By examining the temporal histones of the correction signal changes, it is also possible to determine whether the positioning robot is no longer capable of precise positioning, for example, due to worn bearings, high tolerances, or general wear. In this case, the continuous change in the correction signals will not become increasingly smaller, but will not fall below a certain value. In this case, a maintenance signal can be issued based on the evaluation of the temporal sequence of the correction signal changes to service the positioning robot. The system can therefore also be used for overall system maintenance.

[0033] Figs. 2a and 2b show that a correction of the positioning of an object 14 can already be made during the transfer to the end position 24, i.e., during the positioning process. In all figures, identical and functionally equivalent parts are provided with identical reference numerals. In Figs. 2a and 2b, the controller 22 of the grouping device and all its associated components 30, 34, and 36 are present in the same way as in Fig. 1, but are not shown.

[0034] In the present case of Fig. 2a, the camera 32 detects a rotational misalignment of the object 14 during transport by means of the positioning robot 16a, which misalignment is designated by the reference numeral 40.

[0035] Using the same auto-correction mechanism as described above in connection with Fig. 1, the evaluation logic 34 will now cause the controller 22 of the positioning robot 16a to correct the rotational position of the object 14, as shown in Fig. 2b at position 42. The gripper of the positioning robot is thus slightly rotated to compensate for the incorrect angle of the object 14, so that it is again correctly aligned, as shown by the reference numeral 42. Thus, a check of the movement sequences when the positioning robot 16; 16a, 16b, 16c transfers the objects 14 into the end position 24 already takes place during the positioning process, which accordingly leads to significantly more precise positioning in the end position 24. The invention is not limited to the described embodiment, but can be varied as desired within the scope of the appended claims.

[0036] List of reference symbols:

[0037] 8 Handling device

[0038] 10 Grouping device

[0039] 12 Supply conveyor device - Conveyor belt, in particular for the supply of

[0040] Objects in collision

[0041] 14 object to be grouped

[0042] 16 Positioning robots in first embodiment

[0043] 16a, b ,c Positioning robot in second embodiment

[0044] 18 Articulated arm of the positioning robot

[0045] 20 grippers of the positioning robot

[0046] 22 Control of the positioning robot

[0047] 24 End position of the objects in a given grouping

[0048] 26 Incorrect position due to lateral offset in the setting plane

[0049] 28 Incorrect position due to rotation error

[0050] 30 Position correction device

[0051] 32 Camera for capturing objects during positioning and in the

[0052] End position

[0053] 34 Evaluation logic

[0054] 36 memory of the evaluation logic

[0055] 40 Rotational misalignment during positioning due to incorrect grip

[0056] 42 Correction of the rotational misalignment by corrective rotation of the gripper x, y positioning coordinates in the range provided for the end position of the group

[0057] Parking level

Claims

Claims:

1. Device (8) for handling objects (14) with a grouping device (10) for the objects (14) with at least one positioning robot (16; 16a, 16b, 16c) which is designed to position the objects (14) in a predetermined mutual end position (24), characterized by a position correction device (30) associated with the grouping device (10), which comprises the following features: - at least one camera (32) directed at the objects (14) positioned by the grouping device (10), - an evaluation logic (34) which is designed to compare the position of each object (14) detected by the camera (32) during positioning and / or its end position (24) with target values, and to derive a correction signal for the corrective actuation of the at least one positioning robot (16; 16a, 16b, 16c) from the comparison result, wherein the evaluation logic (34) is designed to detect the corrective actuation of the positioning robot (16; 16a, 16b, 16c) resulting from the correction signal by means of the camera (32), to evaluate this with regard to compliance with the target values ​​and, as a result of the evaluation, to modify or leave the correction signal assigned to the comparison result.

2. Device (8) according to claim 1, characterized in that the evaluation logic (34) has a memory (36) in which comparison results and associated correction signals are stored with a time stamp.

3. Device (8) according to claim 2, characterized in that the evaluation logic (34) is designed to evaluate the temporal sequence of the correction signals assigned to a comparison result in order to generate an optimized correction signal, in particular via genetic algorithms and self-learning algorithms.

4. Device (8) according to claim 2 or 3, characterized in that the evaluation logic (34) is designed to evaluate the temporal sequence of the correction signals assigned to a comparison result and to emit a maintenance signal if the sequence corresponds or does not correspond to predetermined regularities.

5. Device (8) according to one of claims 2 to 4, characterized in that the stored correction signals associated with the comparison results take into account further parameters such as temperature, air humidity and / or machine parameters of the at least one positioning robot (16; 16a, 16b, 16c).

6. Device (8) according to claim 5, characterized in that the machine parameters comprise at least one of the following parameters: - the machine type of the positioning robot (16; 16a, 16b, 16c), - the age of the positioning robot (16; 16a, 16b, 16c), - the age of components of the positioning robot (16; 16a, 16b, 16c), - the maintenance status of the positioning robot (16; 16a, 16b, 16c), - the maintenance status of components of the positioning robot (16; 16a, 16b, 16c).

7. Device (8) according to one of the preceding claims, characterized in that the target values ​​relate to the rotational position of the object (14) as well as its absolute position.

8. Device (8) according to one of the preceding claims, characterized in that the target values ​​include both the final position of the object (14) in the group and / or several positions of the object (14) during its handling by the positioning robot (16; 16a, 16b, 16c).

9. Device (8) according to one of the preceding claims, characterized in that the grouping device (10) is designed to group beverage bottles and / or containers with beverage bottles.

10. Device (8) according to one of the preceding claims, characterized in that the correction signal is correlated with a control algorithm for controlling the movement sequence of the positioning robot (16; 16a, 16b, 16c), and that the change in the correction signal has an effect on the movement sequence of the positioning robot (16; 16a, 16b, 16c) when positioning the objects (14).

11. Device (8) according to one of the preceding claims, characterized in that the evaluation logic (34) has an assignment table and / or an assignment algorithm for assigning a correction signal to a comparison result.

12. Device (8) according to one of the preceding claims, characterized in that it has a conveyor device (12) connected upstream of the grouping device (10), which is designed to convey the objects in abutting manner in the effective range of the at least one positioning robot (16; 16a, 16b, 16c).

13. Method for grouping objects (14) by means of a grouping device (10) having at least one positioning robot (16; 16a, 16b, 16c), comprising the following successive method steps: - the position of each object (14) is compared with its target position during positioning and / or after positioning, - a correction signal for corrective actuation of the at least one positioning robot (16; 16a, 16b, 16c) is derived from the comparison result, - the camera (32) detects the correction actuation of the positioning robot (16; 16a, 16b, 16c) resulting from the correction signal, - the detected correction operation of the positioning robot (16; 16a, 16b, 16c) is evaluated with regard to compliance with the target position of the object (14) and - As a result of the evaluation, the correction signal assigned to the comparison result is modified or not.

14. Method according to claim 13, characterized in that the comparison results and associated correction signals are stored with a time stamp.

15. Device according to claim 14, characterized in that the temporal sequence of the correction signals assigned to a comparison result is evaluated, e.g. by genetic or self-learning algorithms, in order to generate an optimized correction signal therefrom.

16. Method according to claim 14 or 15, characterized in that the temporal sequence of the correction signals assigned to a comparison result is evaluated, and that a maintenance signal or alarm signal is emitted if the temporal sequence of the correction signals corresponds to predetermined regularities or not.

17. Method according to one of claims 13 to 16, characterized in that for deriving the correction signals from the comparison results, further parameters such as temperature, air humidity and / or machine parameters of the at least one positioning robot (16; 16a, 16b, 16c) are taken into account.

18. Method according to one of claims 13 to 17 using a device (8) according to one of claims 1 to 12.

Images