Method for automatically operating machine tool and machine tool

Abstract

A method for automatically operating a machine tool (10), characterized in that the method comprises the following steps: acquiring tasks, in particular machining tasks, given by an operator in a natural language by means of an acquisition device (26); determining a method step that can be carried out by the machine tool (10) from the acquired task, in particular the machining task, by means of a first processing device (30); the determined method steps are output to the interface (36) by means of the second processing device (34).

Description

Technical Field

[0001] The invention relates to a method for automatically operating a machine tool and a machine tool according to the preambles of the respective coordinated claims. Background Art

[0002] Machine tools, such as panel saws, are known on the market and include input and output devices with a graphical user interface (operator interface) through which the machine tool can be operated. The graphical user interface can be visualized, for example, on an input and output device in the form of a touch screen. Furthermore, basic language models such as GPT and LLAMA are well known and can process naturally written language and generate text output based on it. Voice control uses such language models to execute commands entered by the user in natural language. Summary of the Invention

[0003] The object of the present invention is to provide a method for automatically operating a machine tool and a machine tool, by which the operation of the machine tool can be simplified and the required expertise of an operator for carrying out the operation can be reduced.

[0004] This object is achieved by a method and a machine tool having the features of the independent claims. Advantageous developments are mentioned in the dependent claims.

[0005] The advantage of the present invention is that it significantly reduces operator training requirements compared to conventional machine tools, thereby expanding the range of possible operators. Furthermore, only qualified operators need to be trained at reduced expense, thus saving costs. Machine tool operation is also accelerated, and the frequency of operating errors is reduced. Even complex operating processes can be automated.

[0006] Specifically, this is achieved through a method for automated operation of a machine tool. Thus, the present invention does not automate the operational flow of the machine components of a machine tool, but rather the actual operation of the machine tool by the operator. Thus, there is an automated step between the processing request expressed by the operator and the method steps required to execute that processing request (e.g., the order and / or selection of the required method steps). The term "automated" is intended to encompass the fact that the operation of the machine is partially performed through computer-generated steps.

[0007] A machine tool is typically a machine tool that can be used to process a starting workpiece using one or more tools in several processing steps, which are usually carried out sequentially but sometimes simultaneously. However, a machine tool can also be a device that can automatically process workpieces using tools in several method steps. In this regard, a workpiece magazine in which workpieces are processed using tools, such as grippers or transport devices, can also be considered a machine tool in this context. A typical and particularly preferred example of such a machine tool is a sheet metal cutting machine, in particular a sheet metal cutting saw. Using such a machine tool, typically large-format sheets can be cut into workpieces, which are used, for example, in the production of furniture. Another typical and particularly preferred example is an automatic sheet metal magazine for such a sheet metal cutting machine.

[0008] The method of the present invention includes, as step a, collecting a task, particularly a processing task, specified by an operator in natural language using a collection device. The term "natural language" refers to the operator describing the task or processing task using, for example, commonly used words and / or complete sentences with a subject, predicate, and object, without necessarily using specific keywords (wherein the use of keywords simplifies collection). The task or processing task can be specified in spoken or written language. In the first case, the collection device is typically a microphone, while in the second case, it is typically a keyboard. A combination of these is also possible. It is also conceivable that the task or processing task can be specified in the form of gestures by the operator, which are then collected by a camera. The task or processing task can be defined, for example, by the desired end product to be processed, or by a workpiece or, in some cases, by some other, as unique as possible, target state of the machine tool itself. For example, a task or processing task may include reconfiguration of the machine tool, such as a tool change. It may also include modifications or setup of the machine tool, movements within the machine tool or within individual components of the machine tool, or changes in the machine tool's control state.

[0009] The method of the present invention further includes, as step b, automatically determining method steps that can be performed by the machine tool based on the acquired task or processing task using a first processing device. To this end, the first processing device typically includes a software-based, particularly artificial intelligence-based, language model that automatically assigns specific meanings to the acquired and recognized words and the context in which they were acquired and recognized. Using these assigned meanings, for example, by taking into account probabilities and / or using artificial intelligence, method steps stored in a database can be automatically selected and / or parameterized, and the method steps determined within this context. Thus, the method steps for solving the given task or processing task are determined.

[0010] Known optimization algorithms can also be used, which, for example, select method steps based on a predetermined optimization parameter or multiple predetermined optimization parameters. The sequence of method steps can also be automatically selected based on probabilities and / or can be determined by typical optimization algorithms based on predetermined optimization parameters. Examples of optimization parameters include speed, energy consumption, material usage, etc.

[0011] In the example of a panel saw, the operator can enter the following processing task, which is then collected by the collection device:

[0012] "I would like to make a strip of uncoated particleboard material that is 10 mm thick, 800 mm long and 200 mm wide. This strip should have on its upper side a continuous groove of rectangular cross-section transversely to its longitudinal direction, with a depth of 2 mm and a width of 10 mm. One such groove should be located at 200 mm along the strip's length, and another at 600 mm along its length."

[0013] The first processing device now automatically generates a proposal for corresponding method steps based on the processing task, which can be performed by the panel saw. The proposal may look like the following, for example:

[0014] It is required to load a raw board of uncoated particleboard material with a thickness of 10 mm from the board magazine (Plattenlager) into the machine tool.

[0015] Positioning the raw board relative to the saw assembly by means of the positioning device in order to perform the first longitudinal cut by the saw assembly.

[0016] Positioning the raw board relative to the saw unit with the aid of a positioning device in order to carry out a second longitudinal cut by the saw unit at a distance of 200 mm from the first longitudinal cut.

[0017] Rotate the slats 90° using the swivel mechanism.

[0018] Positioning the strip relative to the saw assembly by means of the positioning device in order to perform a first transverse cut by the saw assembly.

[0019] Positioning the strip relative to the saw unit with the aid of the positioning device so that the second transverse cut is performed by the saw unit at a distance of 800 mm from the first transverse cut.

[0020] Transport the finished slats to the milling equipment.

[0021] Positioning of the slats relative to the milling equipment.

[0022] A first transversely extending groove with a depth of 2 mm and a width of 10 mm is introduced by means of a milling device at a position 200 mm in the strip length.

[0023] Positioning of the slats relative to the milling equipment.

[0024] A second transversely extending groove with a depth of 2 mm and a width of 10 mm is cut into the strip at a length of 600 mm using a milling device.

[0025] Transport the finished slats to the output station.

[0026] The method according to the invention comprises, as step c, outputting the determined method steps to an interface, in particular by means of a second processing device. Outputting the determined method steps to an interface allows the determined method steps to be processed in a variety of ways.

[0027] As can be seen from the above description, the present invention reduces the discrepancy between the operator's knowledge of a task or machining task and the operator's known implementation possibilities. Thus, the operator can focus on the "what" (i.e., the "semantics") of machine tool operation, and the machine tool automatically determines the resulting "how."

[0028] In one refinement, the determined method steps are automatically transmitted from the interface to an output device, where they are output. The output device can be, for example, a display (e.g., a touchscreen), a loudspeaker, or a printer. Thus, an operator can check whether the method steps automatically determined by the first processing device are reasonable and whether they solve the given task or processing task. This improves the reliability of the machine tool's operation.

[0029] To this end, one refinement provides for the visualization of specific method steps on an output device as a predefined user interface. Thus, the specific method steps are displayed to the operator, for example, in the form of a graphical flow chart. For example, the operator can then modify individual method steps, for example by parameterizing them differently or changing their sequence.

[0030] In one refinement, specific method steps are automatically transferred from the interface to a third processing device, and converted by the third processing device into machine instructions that can be processed by a machine controller. This creates a direct and automated process from the operator's semantic description of a task or processing task to the programming of the machine tool's machine control for automatically executing and completing the specified task or processing task. It is understood that the operator will take into account changes to the specific method steps.

[0031] To this end, one refinement provides that the machine controller does not automatically execute generated specific machine instructions or does not automatically execute generated specific machine instructions without additional confirmation by the operator, and / or that the machine controller automatically executes generated specific machine instructions only after issuing a warning. This improves the safety of machine tool operation. For example, the present invention can provide that safety-related components of a machine tool can only be operated within a limited range, or that safety-related method steps cannot be executed arbitrarily. For example, the execution of a semantically specified task or processing task such as "remove material from the machine tool" may be prevented because it could endanger the safety of the operator.

[0032] In one refinement, the method steps in step b are automatically determined using a language model specific to the machine tool type. This improves the accuracy of the method step determination. For example, a language model for a panel saw machine tool might differ from a language model for a CNC machining center or a combined milling and drilling unit. It is also conceivable to use the language model for a combination of multiple machine tools, such as those with different machining methods. This is possible, for example, in the context of job preparation. Here, for example, the optimal sequence of method steps can be determined and recommended.

[0033] To this end, one refinement provides that the language model is automatically trained using previously acquired data, including operator voice input and the method steps generated therefrom and accepted by the operator. This also increases the accuracy of the method steps determined based on the specified task or processing task.

[0034] In one refinement, the method includes, before step b, the following steps: automatically checking whether the method steps can be generated from the acquired task or processing task with at least a predetermined degree of reliability, and automatically generating a message or question for the operator if the check indicates that the method steps cannot be generated with at least the predetermined degree of reliability. This also improves the operational reliability of the machine tool. For example, it is conceivable that the operator can predefine a required degree of reliability, for example, in the form of a percentage or a qualitative indicator such as "accurate," "very accurate," and "extremely accurate." To determine the actual degree of reliability, appropriate algorithms and / or artificial intelligence can be used, for example, by evaluating the voice quality and the uniqueness or ambiguity of the acquired words.

[0035] In one refinement, at least step b is performed by a processing device located remotely from the machine tool. This provides increased computing power for more precise processing of, for example, natural language, and facilitates improvements. Tasks or machining tasks specified in natural language, or "voice commands," can be processed in the processing device, which can be located, for example, in a dedicated customer data processing center or in the "cloud."

[0036] The present invention also includes a machine tool comprising an input device, at least one processing device, and an interface. In the machine tool according to the present invention, the input device is configured and designed to capture a task or processing task specified by an operator in natural language. The machine tool also includes a first processing device and a second processing device. The first processing device is configured and designed to generate machine instructions processable by a machine controller based on the specified task or processing task, and the second processing device is configured and designed to output the generated machine instructions to the interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The embodiments of the present invention are explained below with reference to the accompanying drawings. As shown in the drawings:

[0038] Figure 1 is a schematic plan view of a machine tool in the form of a panel cutting saw; and

[0039] Figure 2 Is used to run Figure 1 Schematic flow chart of the method for the machine tool shown. DETAILED DESCRIPTION

[0040] exist Figure 1 In the present embodiment, a machine tool, exemplarily in the form of a panel saw, is generally designated by reference numeral 10. The panel saw shown here is used to cut large raw workpieces (which may also be provided in stacks) into separate workpieces. These separate workpieces are typically used to manufacture furniture. In other embodiments (not shown), the machine tool can also be another machine, such as a CNC machining center, a milling machine, a drilling machine, or a combination of different types of machines.

[0041] The panel cutting saw 10 shown in the present example comprises a feed table 12, a machine table 14 and a removal table 16 consisting of several areas. A portal program slide 18 can be moved in a known manner in the transport direction 20. It comprises a plurality of clamps 22, which can usually engage the rear edge of a plate-like workpiece. In the machine table 14, there is usually a saw groove (not shown) which extends transversely to the transport direction 20. The saw group can be moved in the longitudinal direction of the saw groove. Above the machine table 14 there is a pressure bar (also not shown) which clamps the workpiece between the pressure bar and the machine table during the sawing process. The program slide 18 serves in particular to position the workpiece relative to the saw group.

[0042] The machine tool 10 includes a control and regulating device, generally designated by reference numeral 24. This control and regulating device may be implemented, for example, by one or more computers having one or more memories for program code, one or more microprocessors for processing the program code, and generally a series of interfaces for inputting and outputting data and information. Components of the control and regulating device 24 are typically located near the removal station 16, but components may also be located remotely, for example in another building.

[0043] The control and adjustment device 24 includes several devices and functional modules described below: First, the control and adjustment device 24 includes a collection device 26. This collection device 26 is configured and designed to collect natural language 28 from the operator. For example, the collection device 26 may be a microphone. Alternatively or additionally, the collection device 26 may be a keyboard or include a keyboard. Specifically, the collection device 26 is configured and designed to collect tasks, particularly machining tasks, specified by the operator to the machine tool 10 in natural language 28.

[0044] The term "natural language" is intended to encompass the operator's description of a task or machining task, for example, using commonly used words and / or complete sentences with a subject, predicate, and object. A task or machining task may be defined, for example, by a desired end product to be machined, by a workpiece, or, in some cases, by other target states of the machine tool 10 itself, or by machine parameterization or switching states.

[0045] The control and regulation device 24 also includes a first processing device 30, which is configured and designed to determine method steps that can be executed by the machine tool 10 based on the acquired task or processing task. To this end, the first processing device 30 is provided with a language model 32. In the present case, this language model 32 is, for example, specific to the type of machine tool 10. Thus, taking the machine tool 10 designed as a panel saw as an example, this involves a language model 32 specifically configured for panel saws.

[0046] Furthermore, the control and regulation device 24 includes a second processing device 34, which is configured and designed to output the method steps determined by the first processing device 30 to an interface 36. The interface 36 is connected to an output device 38, and the determined method steps are automatically transmitted from the interface 36 to the output device 38. The determined method steps are output at the output device 38. For example, the output device 38 may include a touch screen on which the determined method steps are visualized as a predetermined user interface.

[0047] A third processing device 40 is also part of the control and regulation device 24. This third processing device 40 is also connected to the interface 36. Therefore, the specific method steps are also automatically transmitted from the interface 36 to the third processing device 40. The third processing device 40 is configured and designed to convert the method steps transmitted from the interface 36 into machine instructions that can be processed by a machine controller 42 of the machine tool 10 into an automated sequence of the specific method steps. To this end, the machine controller 42 is connected, for example, to various sensors and actuators of the machine tool 10.

[0048] The machine controller 42 has a safety module 44, which is configured and designed so that the machine controller 42 does not automatically execute the generated specific machine instructions or does not automatically execute the generated specific machine instructions without additional confirmation by the operator, and / or the machine controller 42 automatically executes the generated specific machine instructions only after issuing a warning, for example on the output device 38.

[0049] The third processing device 40 and the output device 38 are also functionally connected to an optimization module 46, wherein in the present case, the output device 38 also serves as an input device, for example by being designed as a touch screen. The optimization module 46 is configured and designed to automatically train the language model 32 using previously acquired data, wherein the previously acquired data includes the operator's voice input and the method steps generated thereby and accepted by the operator.

[0050] Finally, the reliability module 48 should be mentioned, which is connected in particular to the first processing device 30. This reliability module enables the first processing device 30 to automatically check whether a method step can be generated from a captured task or processing task with at least a predetermined reliability. If the check indicates that a method step cannot be generated with at least a predetermined reliability, a message and / or question is automatically generated for the operator and displayed on the output device 38.

[0051] It should be understood that the above-mentioned various devices and modules do not necessarily need to be implemented in the form of separate hardware components, but can be realized, for example and possibly, by software modules of a computing device in the form of one or more computers.

[0052] Refer to the following Figure 2 Instructions for running Figure 1 The method of the machine tool 10 is shown:

[0053] The method begins in a start function block 50. In function block 52, a task or processing task specified by an operator in natural language is acquired by acquisition device 26. In function block 54, method steps that can be executed by the machine tool are determined based on the acquired task or processing task by first processing device 30. In function block 56, the determined method steps are output to interface 36 by second processing device 34. In function block 58, the determined method steps are automatically transferred from interface 36 to output device 38 and output there.

[0054] In function block 60, the determined method steps are automatically transferred from interface 36 to third processing device 40. In function block 62, the transferred method steps are converted by third processing device 40 into machine instructions that can be processed by machine controller 42. In function block 64, the generated specific machine instructions are not automatically executed, or are not automatically executed without additional confirmation by the operator, and / or are automatically executed only after a warning is issued. The method ends in end function block 66.

Claims

1. A method for automatically operating a machine tool (10), characterized in that The method comprises the following steps: a. Acquiring a task given by the operator in natural language, in particular a processing task, by an acquisition device (26); b. determining method steps capable of being performed by the machine tool (10) from the collected task, in particular the processing task, by means of a first processing device (30); c. Outputting the determined method steps to an interface (36), in particular by means of a second processing device (34).

2. The method according to claim 1, characterized in that The determined method steps are automatically transmitted from the interface (36) to a device (38) and output at the device (38).

3. The method according to claim 2, characterized in that The determined method steps are visualized on the output device (38) as a predefined user interface.

4. The method according to at least one of the preceding claims, characterized in that The determined method steps are automatically transmitted from the interface (36) to a third processing device (40), and the determined method steps are converted by the third processing device (40) into machine instructions that can be processed by a machine controller (42).

5. The method according to claim 4, characterized in that The machine controller (42) does not automatically execute the generated specific machine instructions or does not automatically execute the generated specific machine instructions without additional confirmation by the operator, and / or the machine controller (42) automatically executes the generated specific machine instructions only after issuing a warning.

6. The method according to at least one of the preceding claims, characterized in that The method steps in step b are determined automatically using a language model (32) specific to the type of machine tool (10).

7. The method according to claim 6, characterized in that The language model (32) is automatically trained using previously collected data, including operator speech input and method steps generated thereby and accepted by the operator.

8. The method according to at least one of the preceding claims, characterized in that Before step b, the method also includes the following steps: automatically checking whether the method steps can be generated from the collected task, in particular the processing task, at least with a predetermined reliability, and when the check shows that the method steps cannot be generated at least with a predetermined reliability, automatically generating information or questions for the operator.

9. The method according to at least one of the preceding claims, characterized in that At least step b is performed by a processing device (30) arranged remote from the machine tool (10).

10. A machine tool (10), comprising a collection device (26), at least one processing device (30) and an interface (36), characterized in that: The acquisition device (26) is configured and designed so that tasks, in particular processing tasks, given by an operator in natural language can be acquired by the acquisition device (26); the machine tool (10) includes a first processing device (30); the first processing device (30) is configured and designed so that method steps that can be executed by the machine tool (10) are determined from the acquired tasks, in particular processing tasks; and the machine tool includes a second processing device (34); the second processing device (34) is configured and designed so that the second processing device can output the determined method steps to an interface (36).

11. The machine tool (10) according to claim 10, characterized in that The machine tool is designed to carry out the method according to any of the dependent claims 2 to 9.

Images