Method and device for intelligently selecting field of view of camera on machine tool

Abstract

The invention relates to a method (34) for monitoring the production of a component (14) using a machine tool (12), in which a plurality of cameras (20a, b) cover different fields of view (22a, b). The algorithm (26) may create a ranking of correlations of the records of the cameras (20a, b), with only one or more most correlative records displayed or highlighted on the monitor (28). The algorithm (26) can assign a higher correlation to a record that: a) depicts the known machine tool part mentioned in the status output or error output (30); b) having a high optical flow; and / or c) depicting the identified machine tool part that is moving. In the case of c), if the identified machine tool part moves from one field of view (22a, b) belonging to the recording to the next field of view, the recording can be assigned a higher correlation in sequence. Records assigned with lower relevance may be deleted or displayed and / or saved in a reduced size. The algorithm (26) may take the form of artificial intelligence. The invention also relates to a device (10) for carrying out such a method (34).

Description

Technical Field

[0001] The present invention relates to a method and an apparatus for manufacturing a component. Background Art

[0002] From DE 10 2017 121 098 A1, it is known to track a component to be manufactured through a production process. For this purpose, a plurality of cameras are provided in the production, wherein the imaging camera changes depending on the field of view of the camera where the component is located. The position of the component can be supported by a UWB positioning system.

[0003] It is also known to observe the production of parts in a machine tool by observing from different fields of view or viewing angles of a plurality of cameras. The user or observer of the production process can be located at a different location from the machine tool and act as a remote controller. However, it is difficult for an inexperienced user to find one or more relevant camera recordings from among the numerous camera recordings. In particular, in the event of a malfunction of the machine tool, the user wishes to quickly view one or more relevant camera recordings.

[0004] Invention Task

[0005] It is therefore an object of the present invention to provide a method and a device which significantly simplify the monitoring of a machine tool. Summary of the invention

[0006] According to the invention, this object is achieved by a method according to claim 1 and an apparatus according to claim 12. The dependent claims reflect preferred further developments.

[0007] The object is therefore achieved by a method for producing a component using a machine tool, wherein the following method steps are performed:

[0008] A) creating a recording of at least part of the machine tool by a plurality of cameras, wherein the recording is fed to an evaluation unit having an algorithm for evaluating the recording;

[0009] B) Selecting one or more relevant records by an algorithm and outputting the one or more relevant records to the monitor. The relevance of the records is determined by one or more of the following criteria:

[0010] a) records showing machine tool parts mentioned in a status output or an error output of the machine tool control unit (e.g. a sensor value or an axis position) are assigned a higher relevance than records showing machine tool parts not mentioned in a status output or an error output of the machine tool control unit;

[0011] b) records with high optical flow are assigned a higher relevance than records with lower optical flow; and / or

[0012] c) Records in which a moved machine tool part was identified are assigned a higher relevance than records in which a moved machine tool part was not identified.

[0013] The method according to the invention enables a user or observer of a machine tool to focus on an area of ​​interest of the machine tool. This enables even inexperienced users to perform remote maintenance of the machine tool. Errors can be analyzed and found more quickly and easily.

[0014] The method may be arranged to display only the most relevant records on the monitor. Alternatively, only a few relevant records may be displayed on the monitor. Alternatively, the most relevant records or the few relevant records may be highlighted on the monitor. For example, less relevant records may be displayed smaller on the monitor.

[0015] The camera is preferably designed as a video camera; the recording is then available as a video recording.

[0016] Method steps A) and B) can be repeated, in particular can be repeated a plurality of times. This enables the user to continuously track a region of interest in the machine tool.

[0017] Preferably, at least one of the cameras is directed toward an area of ​​the machine tool specified by a status output and / or an error output of the machine tool control device. In particular, the alignment can be accomplished by rotating and / or zooming the camera. By rotating and / or zooming, the field of view of the camera can be directed particularly well toward the area of ​​interest. If multiple cameras are directed toward the area, the area can be viewed from multiple perspectives. The most relevant recording can then be selected from the multiple perspectives.

[0018] Particularly preferably, when criterion c) is applied, the algorithm predicts the field of view of the camera to which the identified machine tool part will move next and assigns a higher relevance to the associated recording (of the next field of view) if the identified machine tool part has already left the current field of view. This enables the user to easily track the machine tool part.

[0019] Method step B) can be performed in real time, ie after method step A) with as little delay as possible. Alternatively, method step B) can be performed using stored logs, in particular for fault analysis of the machine tool.

[0020] To save resources, less relevant records may be deleted or transmitted or stored at a reduced resolution.

[0021] In a further preferred embodiment of the invention, the algorithm is constructed in the form of a machine learning algorithm. The machine learning algorithm is preferably trained using stored records that have been assigned to a specific state output or error output. Alternatively or additionally, the machine learning algorithm can be trained by selecting the cameras of one or more experienced users.

[0022] Monitoring of the machine tool is further simplified if, in addition to one or more logs displayed on the monitor, a status output or an error output from the machine tool control unit is also provided on the monitor, in particular.

[0023] Furthermore, the possibility of interacting with the machine tool can be provided. The machine data can be shown in the log and, if the situation requires, functions can be deactivated / activated directly in the log. Example: The machine tool stops due to a "transmission control" error. The operator is shown one or more relevant live logs directly and can restart the machine tool in the log if the error is a false alarm. This makes operation much faster than having to switch to the control device and search for the function.

[0024] The method according to the invention is particularly suitable for use on machine tools designed for laser processing, in particular laser cutting or laser welding, of components. Alternatively or additionally, the method can be used on automatic bending machines or storage systems.

[0025] The object according to the invention is also achieved by a device for carrying out the method described herein. The device comprises a machine tool, a camera (in particular in the form of a video camera), an evaluation unit with an algorithm, and a monitor. The features and advantages described for the method refer accordingly to the device and vice versa.

[0026] The machine tool preferably has a laser head for laser processing of components, in particular for laser cutting and / or laser welding.

[0027] Alternatively or additionally, the machine tool can have an automatic tool changer, in particular for a punching machine or a bending machine.

[0028] Further advantages of the present invention are derived from the description and the accompanying drawings. Similarly, the features mentioned above and the features still to be described can be used alone or in any desired combination according to the present invention. The embodiments shown and described should not be understood as an exhaustive enumeration, but should be understood as having exemplary characteristics for describing the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of the device according to the invention and the method according to the invention is shown. DETAILED DESCRIPTION

[0030] Figure 1The device 10 is shown with a machine tool 12 for producing a component 14. For producing the component 14, the machine tool 12 has, among other things, a laser head 16, here in the form of a cutting head. The production of the component 14 is controlled by a machine tool control 18.

[0031] The production of the component 14 is monitored by a plurality of cameras 20a, 20b, here in the form of video cameras. However, each camera 20a, b monitors its own field of view 22a, 22b. The fields of view are shown separately here, but can also overlap. The recordings created by the cameras 20a, b, here in the form of videos, are fed to an evaluation unit 24. The evaluation unit 24 can be part of the machine tool control 18. The evaluation unit 24 has an algorithm 26 for evaluating the recordings.

[0032] The evaluation unit 24 is configured to use the algorithm 26 to determine the most relevant multiple records, in particular the most relevant one record, and transmit them to the monitor 28. By selecting one or more of the most relevant records, it is much easier for the user to evaluate the production process. In particular, in the event of a production error, the user can focus on the relevant records on the monitor 28 without having to hide the records of irrelevant parts of the machine tool 12.

[0033] Furthermore, the evaluation unit 24 is designed to direct the cameras 22a, 22b to the region of the machine tool 20 specified by the status output or error output 30 of the machine tool control unit 18. For this purpose, the cameras 22a, 22b can be rotated by the evaluation unit 24 or the zoom of the cameras 22a, 22b can be set.

[0034] The algorithm 26 may be designed in the form of a machine learning algorithm. In particular, the machine learning algorithm may be trained based on the labeled stored records and / or based on user behavior when selecting records.

[0035] Algorithm 26 can evaluate the relevance of a record in the following way:

[0036] a) evaluating the log to determine whether it indicates the part of the machine tool 12 for which the machine tool control unit 18 outputs a state output and / or an error output 30;

[0037] b) determining the optical flow of the records and assigning a high correlation to records with a high optical flow; and / or

[0038] c) Identifying the moved parts of the machine tool 12 in the recording.

[0039] For example, when criterion c) is applied, the algorithm 26 may identify the laser head 16 in the field of view 22a and recognize that it will next appear in the field of view 22b based on the movement of the laser head 16 in the direction of the arrow 32. If the laser head 16 leaves the field of view 22a, the algorithm 26 may assign a high relevance to the field of view 22b so that the user can continue to follow the movement of the laser head 16 on the monitor 28.

[0040] Figure 1 The device 10 shown or Figure 1 The illustrated method 34 enables focused tracking of the most relevant records. Irrelevant records may be deleted or transmitted at a reduced resolution to save storage space, computing power, and / or bandwidth.

[0041] Referring to the accompanying drawings, the present invention generally relates to a method 34 for monitoring the manufacture of a part 14 using a machine tool 12, wherein a plurality of cameras 20a, b cover different fields of view 22a, b. An algorithm 26 may create a ranking of the relevance of the recordings from the cameras 20a, b and display or highlight only one or more of the most relevant recordings on a monitor 28. The algorithm 26 may assign a higher relevance to the following recordings: a) depicting a known machine tool part mentioned in a status output or an error output; b) having a high optical flow; and / or c) depicting an identified machine tool part that is moving. In the case of c), if the identified machine tool part moves from one field of view 22a, b of the recording to the next field of view, the recording may be successively assigned a higher relevance. Records assigned a lower relevance may be deleted or displayed and / or saved at a reduced size. The algorithm 26 may take the form of an artificial intelligence. The present invention also relates to an apparatus 10 for performing such a method 34.

[0042] Reference numerals list

[0043] 10. Equipment

[0044] 12 Machine Tools

[0045] 14 Components

[0046] 16 Laser Head

[0047] 18 Machine tool control device

[0048] 20a, b Camera

[0049] 22a, b Field of view

[0050] 24 Analysis and processing unit

[0051] 26 Algorithms

[0052] 28 Monitor

[0053] 30 Status output and / or error output

[0054] 32 Moving direction of laser head 16

[0055] 34 Methods

Claims

1. A method (34) for producing a component (14) using a machine tool (12), comprising the following method steps: A) simultaneously creating a plurality of recordings of at least parts of the machine tool (12) by means of a plurality of cameras (20a, b) having different fields of view (22a, b) and feeding the recordings to an evaluation unit (24) having an algorithm (26); B) selecting the relevant one or more records by the algorithm (26) in a computer-based manner and outputting the one or more records to a monitor (28), wherein: The relevance of the records created is determined based on one or more of the following criteria: a) the corresponding entry shows an area of ​​the machine tool (12) which is specified by a status output and / or an error output (30) of a machine tool control device (18); b) The corresponding recording shows a relatively high optical flow; c) The corresponding log shows the part of the machine tool (12) identified in the log and moved by the machine tool.

2. The method according to claim 1, wherein: The method steps A) and B) are repeated, in particular a plurality of times.

3. The method according to claim 1 or 2, wherein: At least one of the cameras (20a, b) is focused, in particular by rotation and / or zooming, on a region of the machine tool (20) specified by a status output and / or an error output (30) of the machine tool control unit (18).

4. The method according to claim 2 or 3, wherein: The algorithm (26) predicts, under criterion c), into which next field of view (22a, b) the identified part of the machine tool (12) will move, wherein the recording of the next field of view (22a, b) is evaluated with high relevance if the identified part leaves the current field of view (22a, b).

5. A method according to any one of the preceding claims, wherein: Method step B) is performed using the stored records.

6. A method according to any one of the preceding claims, wherein: Records with low relevance are deleted or transmitted and / or stored at a reduced resolution.

7. A method according to any one of the preceding claims, wherein: The algorithm (26) is constructed in the form of a machine learning algorithm.

8. A method according to any one of the preceding claims, wherein: A status output and / or an error output (30) of the machine tool control unit (18) is output for one or more records displayed on the monitor (28).

9. A method according to any one of the preceding claims, wherein: Interaction options for controlling the machine tool (12) are output on one or more records displayed on the monitor (28).

10. A method according to any one of the preceding claims, wherein: The production of the component (14) includes laser processing by means of the machine tool (12).

11. The method according to claim 10, wherein: The manufacture of the component (14) includes laser cutting of the component (14).

12. A device (10) for performing the method (34) according to any one of the preceding claims, wherein: The device (10) comprises the machine tool (12), the camera (20a, b), the analysis unit (24) with the algorithm (26), and the monitor (28), wherein the algorithm (26) is configured to: determine the relevance of the recordings created by the camera (20a, b) based on one or more criteria a) to c) and output one or more relevant recordings to the monitor (28).

13. The device according to claim 12 in combination with claim 10, wherein The machine tool (12) has a laser head (16).

14. The apparatus according to claim 12 or 13, wherein: The machine tool (12) has an automatic tool changer.

Images