Method for controlling automatic screw screwing machine

Abstract

The invention relates to a method for controlling an automatic screw screwing machine, comprising: a screw screwing unit for automatically screwing a screw into a component; and a control unit for controlling the screw-screwing unit using a screw-screwing program stored in the control unit, the screw-screwing program defining a time-varying process of the rotational speed required to drive the screw during the screw-screwing process, and optionally, the screw-screwing program being stored in the control unit, the screw-screwing program defining a time-varying process of the rotational speed required to drive the screw during the screw-screwing process. The screw screwing program also defines a time-dependent course of a feed force to be exerted on the screw during the screw screwing process, during which the control unit records a data set relating to the screw screwing process, and the control unit determines the time-dependent course of the feed force to be exerted on the screw during the screw screwing process. The data set comprises the following screw screwing parameter changing processes along with time: the actual rotating speed of the screw, the torque applied to the screw, the feeding position of the screw and the feeding speed of the screw. The control unit transmits the recorded data set to a remote data processing unit, in which the screw screwing process is evaluated by means of artificial intelligence on the basis of the transmitted data set.

Description

Technical Field

[0001] The present invention relates to a method for controlling an automatic screw-driving machine, the automatic screw-driving machine comprising: a screw-driving unit, the screw-driving unit being used to automatically screw a screw into a component; and a control unit, the control unit being used to control the screw-driving unit based on a screw-driving program stored in the control unit, the screw-driving program defining a time-varying process of a rotational speed required to drive the screw during the screw-driving process, and optionally, the screw-driving program also defining a time-varying process of a feed force to be applied to the screw during the screw-driving process, wherein during the screw-driving process, the control unit records a data set related to the screw-driving process, the data set comprising a time-varying process of the following screw-driving parameters: an actual rotational speed of the screw, a torque applied to the screw, a feed position of the screw and a feed speed of the screw. Background Art

[0002] Such methods are well known. In this respect, an experienced screw expert can identify based on the recorded data set what type of screw-driving process was performed, for example whether a metric screw element (with standard thread, fine thread and / or screw locking device) or a self-drilling screw element was performed, whether the self-drilling screw element was screwed into a metal sheet or a blind hole, and what type of material (e.g. plastic, aluminum, steel) the self-drilling screw element was screwed into. If the screw expert has access to a sample data set generated by a trouble-free screw-driving process, the screw expert can infer from the deviating data set the possible causes that have caused or will cause a troublesome screw connection in the future.

[0003] It is understood that the quality of such data set analysis and fault diagnosis depends not only on the experience of the screw expert, but also on whether a screw expert is available in the first place. If the end customer has purchased an automatic screwdriver, but does not have the financial and / or human resources to hire their own screw expert, the end customer may, for example, enter into a service contract with the manufacturer of the automatic screwdriver, which service contract includes regular and / or on-demand quality checks or fault diagnosis. In the case of a regular service appointment and / or if a fault has already occurred, the manufacturer's service technician will then go to the end customer and repair the automatic screwdriver and, if necessary, try to correct the existing fault. It is understood that the further the end customer is from the manufacturer, especially if the manufacturer and the end customer are located in different countries or even on different continents, the greater the costs and time involved in such a service appointment. Summary of the invention

[0004] An underlying object of the present invention is to provide a more reliable, economical and user-friendly method for controlling an automatic screwdriving machine.

[0005] This object is met by a method having the features of claim 1 and in particular the control unit sends the recorded data set to a remote data processing unit in which the screwdriving process is evaluated with the aid of artificial intelligence based on the sent data set.

[0006] In this context, a remote data processing unit is to be understood as a data processing unit that is not in the same location, in particular not in the same place, as the automatic screwdriver. The data processing unit may be located in a different city, a different country or even a different continent than the automatic screwdriver. For example, the data processing unit can be operated by the manufacturer of the automatic screwdriver, while the automatic screwdriver is used by the end customer. Due to the physical distance between the automatic screwdriver and the data processing unit, the communication between the dynamic screwdriver and the data processing unit is preferably carried out via a remote data connection, which may for example include the Internet.

[0007] Due to the evaluation of the data sets recorded in the automatic screwdriving machine and sent to the data processing unit with the help of artificial intelligence, the screwdriving process performed by the automatic screwdriving machine can be analyzed independently of the personal experience and abilities of the service technician and thus often with greater accuracy and reliability. In addition, it is not absolutely necessary to send a service technician to the end customer, for example, to correct a faulty screwdriving process, optimize an inherently effective screwdriving process or implement a new type of screwdriving process. On the contrary, these tasks can be solved remotely, efficiently and reliably with the help of artificial intelligence, thus saving time and costs in a customer-friendly way.

[0008] Advantageous further developments of the invention can be seen from the dependent claims, the description and the drawings.

[0009] According to one embodiment, the artificial intelligence is formed by a machine learning algorithm, for example a decision tree algorithm, a random forest algorithm, a support vector machine (SVM) algorithm or a k-nearest neighbor (kNN) algorithm.

[0010] According to another embodiment, the artificial intelligence has been trained using multiple heterogeneous datasets. In this context, a heterogeneous dataset is to be understood as a dataset generated by a variety of screwdriving processes, such as a screwdriving process in which metric screws with different threads and with or without fasteners are screwed into threaded holes or a screwdriving process in which self-tapping screws are screwed into blind holes or metal plates made of different materials. Therefore, a heterogeneous dataset must be distinguished from a homogeneous dataset, which is obtained using a single type of screwdriving process, i.e., in which the same type of screw is always screwed into the same type of component.

[0011] In order to be able to provide large amounts of data to the artificial intelligence on the one hand, and to be able to analyze the performed screwdriving process at a later point in time on the other hand, it is advantageous if the recorded data sets are stored in the memory of the control unit for a predetermined period of time, for example days, weeks or months, or even permanently.

[0012] It is usually conceivable that after completion of each screwdriving process, the control unit sends the associated data set to the data processing unit, or that the sending of a plurality of data sets is carried out automatically at predetermined time intervals. However, this requires a permanent or at least temporary, in particular automatically established communication between the automatic screwdriving machine and the data processing unit, which is not necessarily desirable from the point of view of the user of the automatic screwdriving machine. Therefore, in many cases it is advantageous if the control unit sends the recorded data set to the remote data processing unit at the instruction of the user of the automatic screwdriving machine.

[0013] In order to improve the quality of the evaluation of the artificial intelligence, the control unit can, at the instruction of the user of the automatic screwdriving machine, send at least one first data set associated with a fault-free screwdriving process and at least one second data set associated with a faulty screwdriving process to a remote data processing unit. The artificial intelligence preferably identifies a fault that occurred in the faulty screwdriving process by evaluating the first data set and the second data set. The so-called artificial intelligence performs a target / actual comparison, on the basis of which a fault that occurred in the faulty screwdriving process can be more reliably identified. In uncertain cases, for example, if the artificial intelligence is less than 80% certain about the fault, an interaction with the user can take place, wherein the user has the option of providing additional data to the artificial intelligence. Feedback indicating the identified fault is then sent to the automatic screwdriving machine via the remote data processing unit, enabling the result of the evaluation to be made available to the user of the automatic screwdriving machine. The feedback can in particular include suggestions for correcting the fault so that the user can correct or adapt the screwdriving process accordingly.

[0014] According to another embodiment, the control unit can, at the instruction of the user of the automatic screwdriving machine, send a plurality of data sets and user-defined specifications for optimizing the screwdriving process to a remote data processing unit, the plurality of data sets being in particular a plurality of data sets associated with a trouble-free screwdriving process. The specifications for optimizing the screwdriving process can, for example, consist of an increased cycle time and / or a higher screw connection quality. The artificial intelligence can then be able to create a suggestion for at least one changed machine setting and / or for a changed screwdriving program based on the sent plurality of data sets and taking into account the sent optimized specifications. In uncertain situations, for example, if the artificial intelligence is less than 80% sure of the appropriate change, an interaction with the user can also be carried out here, wherein the user can choose to provide additional data to the artificial intelligence. The suggestion for at least one changed machine setting and / or for a changed screwdriving program can then be sent from the remote data processing unit to the control unit of the automatic screwdriving machine so that the user can make the suggested changes and perform the corresponding optimized screwdriving process.

[0015] According to yet another embodiment, a user of the automatic screwdriving machine is able to define requirements for a new type of screwdriving process and to input the requirements into the control unit. For example, the new type of screwdriving process may involve the use of different screws or the use of different materials to be screwed in. The control unit sends the user-defined requirements to a remote data processing unit, and the artificial intelligence determines a screwdriving program that matches the user-defined requirements, and then sends the determined screwdriving program from the remote data processing unit to the control unit of the automatic screwdriving machine in order to perform the new type of screwdriving process. As described above, during the execution of the new type of screwdriving process, a data set related to the screwdriving process can be recorded, and after the screwdriving process is completed, the recorded data set can be sent to the data processing unit for evaluation by the artificial intelligence so that the artificial intelligence can determine, if necessary, a recommendation for an optimized screwdriving program and / or optimized machine settings, which is then sent back to the control unit of the automatic screwdriving machine.

[0016] Another subject of the invention is a system for controlling an automatic screwdriving machine, the automatic screwdriving machine comprising: a screwdriving unit for automatically screwing a screw into a component; and a control unit for controlling the screwdriving unit based on a screwdriving program stored in the control unit, the screwdriving program defining the time variation of the rotation speed required to drive the screw during the screwdriving process, and optionally, the screwdriving program also defines the time variation of the feed force to be applied to the screw during the screwdriving process. The control unit is configured to record a data set related to the screwdriving process during the screwdriving process, the data set comprising the time variation of the following screwdriving parameters: the actual rotation speed of the screw, the torque applied to the screw, the feed position of the screw and the feed speed of the screw. The system also comprises a remote data processing unit, the remote data processing unit having artificial intelligence, the recorded data set being able to be sent by the control unit to the remote data processing unit, and in the remote data processing unit, the screwdriving process can be evaluated by means of the artificial intelligence based on the sent data set. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The invention will now be described in more detail purely by way of example with reference to various embodiments and the accompanying drawings, in which:

[0018] Figure 1A is a cross-sectional view of a first embodiment of an automatic screwdriving machine for automatically performing a screwdriving process;

[0019] Figure 1B is a cross-sectional view of a second embodiment of an automatic screwdriving machine for automatically performing a screwdriving process;

[0020] Figure 2 is a data set consisting of four twisting curves and generated by a screw connection with thread forming;

[0021] Figure 3 is a data set consisting of four screw curves and produced by metric screw connections; and

[0022] Figure 4 Schematically illustrates the automatic screwdriving machine of FIG. 1 in communication connection with a remote data processing unit. DETAILED DESCRIPTION

[0023] FIG. 1 shows a first embodiment of an automatic screwdriving machine 10 for automatically performing a screwdriving process.

[0024] The automatic screwdriving machine 10 comprises a base unit 12 which can be mounted at a processing device (not shown), for example at a robot arm. Figure 1AThe automatic screw-driving machine 10 is also referred to as a robot-binding automatic screw-driving machine 10 hereinafter.

[0025] The base unit 12 supports a screw-driving unit 14 which can be moved in a feed direction relative to the base unit 12 by means of a pneumatic or hydraulic feed cylinder 16 of the base unit 12. It is understood that the screw-driving unit 14 can alternatively also be fed by means of an electric drive.

[0026] The screwdriving unit 14 comprises a drive motor 20 which rotationally drives a screwing tool 24 via a drive shaft 22 extending in the feed direction and which is supported displaceably in the feed direction in a positioning head 26 .

[0027] To carry out the screwdriving process, the automatic screwdriving machine 10 is first positioned by the handling device at a small distance in front of two components (not shown here) to be screwed together, for example at a distance of about 20 mm in front of the two components.

[0028] The screw 28 for screw connection is fed to the positioning head 26 by means of a feeding device 30. In order to align and hold the screw 28 in a defined screwing position, the positioning head 26 has a holding mechanism 32, which includes two claw arms 34, which are arranged at the sides of the positioning head 26 in an opposite manner, and the two claw arms 34 are rotatably supported at the positioning head 26, each of which is oriented at right angles to the feeding direction around the rotation axis and can be pressed open to release the screw 28 against the return force of the spring.

[0029] After positioning the automatic screwdriving machine 10 in front of the components, the screwdriving unit 14 and thus the screw 28 are fed towards the components by actuating the feed cylinder 16 until the screw 28 is supported at one of the components.

[0030] Next, the screwdriving unit 14 is advanced by the feed cylinder 16 so that the screwing tool 24 engages the screw 28. The screwing tool 24 is preferably rotated at a low rotation speed here.

[0031] As soon as the screwing tool 24 engages the screw 28 , the rotational speed of the screwing tool 24 and thus of the screw 28 is increased to the desired rotational speed in order to screw the screw 28 into the component while applying a feed force predefined by the feed cylinder 16 .

[0032] It is to be understood that the automatic screwdriving machine 10 does not necessarily have to be robot-bound, but can also be manually guided and can have a handle 36 for this purpose. Figure 1BAn embodiment of such a manually guided automatic screwdriving machine 10 is shown in FIG. Figure 1A Similar to the robot-bound automatic screwdriver 10 of the present invention, the manual-guided automatic screwdriver 10 has a screwdriver unit 14, which includes a drive motor 20, which drives a screwdriver tool 24 in a rotational manner via a drive shaft 22, and the screwdriver tool 24 is supported in a positioning head 26 in a displaceable manner in the feed direction. In addition, Figure 1B The manual guided automatic screw driving machine 10 is also provided with a feeding device 30, which is used to feed the screw 28 to the positioning head 26 of the automatic screw driving machine 10. Figure 1A The robot binding type automatic screw driving machine 10 is opposite, Figure 1B The manually guided automatic screwdriving machine 10 does not have a feed cylinder 16, because the required feed force is applied manually here.

[0033] In both embodiments, the screw-driving unit 14 is controlled by the control unit 80 ( Figure 4 ) control, wherein the control unit 80 stores at least one screwdriving program, which defines the time course of the rotational speed required to drive the screwing tool 24 and thus the screw 28 during the screwdriving process. Figure 1A In the case of a robot-bound automatic screwdriving machine 10 , the screwdriving program also defines the temporal variation of the feed force to be applied by the feed cylinder 16 to the screwdriving tool 24 and thus to the screw 28 during the screwdriving process.

[0034] Furthermore, during the screwdriving process, the control unit 80 records a data set related to the screwdriving process, which data set includes the time course of the following screwdriving parameters: the actual rotation speed of the screwdriving tool 24 and thus the actual rotation speed of the screw 28 detected by means of a suitable sensor, the torque detected by means of a suitable sensor and applied to the screw 28, the feed position of the screwdriving tool 24 and thus the feed position of the screw 28 detected by means of a suitable sensor, and the time derivative of the detected feed position, i.e. the feed speed of the screw 28, also referred to herein as the depth gradient. Each recorded data set is stored in the memory 82 of the control unit 80 for a predetermined period of time, such as days, weeks or months, or is permanently stored for analysis purposes.

[0035] Figure 2An example of such a data set recorded during a screw-driving process of thread forming is shown, wherein a self-drilling screw 280 is screwed into at least one component. The data set comprises four screwing curves, which, from top to bottom, show the actual rotation speed of the screwing tool 24 and therefore the actual rotation speed of the screw 28 (in rpm), the torque applied to the screwing tool 24 and therefore the screw 28 (in Nm), the feed position of the screwing tool 24 and therefore the screw 28 (in mm) and the time derivative of the feed position, i.e. the feed speed or depth gradient (in mm / s).

[0036] Based on the recorded screwing curve, the screwing process can be divided into four stages in time, namely: stage 1, in this paper, stage 1 lasts 0.9 seconds, and in this stage, the screw 28 is moved toward the parts to be screwed together at a low rotation speed of 200 rpm until the end of the screw hits the surface of one of the parts; stage 2, in this stage, in this paper, the screw 28 still penetrates the parts relatively quickly at a low rotation speed of 200 rpm for about 0.2 seconds; stage 3, in this paper, stage 3 lasts about 0.7 seconds, in this stage, in this paper, the screw 28 penetrates the parts more slowly at an increased rotation speed of 500 rpm; and stage 4, in this paper, stage 4 lasts about 0.6 seconds, in this stage, the screw 28 is again slowly screwed further into the parts at a low rotation speed of 200 rpm until the detected torque reaches a maximum value of 0.6 Nm, and the screwing process is stopped by reducing the rotation speed to 0 rpm. The screwing process takes a total of about 2.4 seconds, and after the screwing process is completed, the screw 28 has penetrated into the parts at a depth of about 33 mm.

[0037] For comparison, Figure 3 1 shows a data set recorded during a screw-driving process in which a metric screw 28 is screwed into a corresponding threaded hole provided in a facing component through a pre-punched hole on the facing component. The data set also includes four screwing curves, from top to bottom, the curves show the actual rotation speed of the screwing tool 24 and therefore the actual rotation speed of the screw 28 (in rpm), the torque applied to the screwing tool 24 and therefore the screw 28 (in Nm), the feed position of the screwing tool 24 and therefore the feed position of the screw 28, which is called the simulated depth (in mm) and the time derivative of the feed position, i.e. the feed speed or depth gradient (in mm / s).

[0038] By comparison Figure 2 and Figure 3 The twist curve shown, it can be easily seen that Figure 3 The process of connecting metric screws shown is similar to Figure 2The screw connections with the thread forming shown differ significantly in some respects. For example, Figure 3 The total time for the metric screw tightening process is close to 4 seconds, which is significantly longer than the connected Figure 2 The duration of the screw-driving process for thread forming is related on the one hand to the fact that the metric screw 28 is 50 mm long and must therefore be screwed into the component to a depth of 50 mm, and on the other hand to the fact that the rotation speed increased during phase 3 is only 280 revolutions per minute. Another significant difference is that at the end of the screw-driving process, the metric screw 28 can be tightened with a maximum torque of 6 Nm, i.e. with a torque of approximately Figure 2 The self-tapping screws 28 are tightened with ten times the torque.

[0039] In other words, by comparing the recorded data sets, it is possible to determine whether these data sets are based on similar or different screwdriving processes. Based on the characteristics of the individual screwdriving curves of the data sets, it is even possible to infer whether the underlying screwdriving process is a thread-forming screw connection or a metric screw connection, how long the screw used is, what the pitch and shank diameter of the thread are, and in the case of self-drilling screws, what type of material the self-drilling screw is driven into.

[0040] like Figure 4 As shown, the automatic screwdriving machine 10 can be temporarily or permanently connected to a remote data processing unit 110 via a remote data connection 90, which can include, for example, the Internet 100. For example, the data processing unit 110 can be located at the manufacturer of the automatic screwdriving machine 10 and can be located in a different city, a different country, or even a different continent than the automatic screwdriving machine 10. Via the remote data connection 90, data sets of the above type are provided in FIG. Figure 2 1 and can be sent by the control unit 14 of the automatic screwdriving machine 10 to the data processing unit 110 for analysis purposes. Conversely, the data processing unit 110, after evaluating the sent data set, can send recommendations for correcting a faulty screwdriving process, an incorrect machine setting, an optimized screwdriving program or even a new screwdriving program to the control unit 14 of the automatic screwdriving machine 10.

[0041] Specifically, in order to evaluate the sent data set, the data processing unit 110 uses artificial intelligence 120, which can be formed by a machine learning algorithm, such as a decision tree algorithm, a random forest algorithm, a support vector machine (SVM) algorithm or a k-nearest neighbor (kNN) algorithm, and has been trained using multiple heterogeneous data sets, that is, trained using multiple data sets based on various screw driving processes, including metric screw connections and thread forming screw connections, each screw connection is achieved using screws with different pitches and different lengths of shank diameters and screws of different materials.

[0042] In this way, the artificial intelligence 120 is not only trained or specialized for a specific screwdriving process, but is also able to use the aforementioned features of the screwing curves of the recorded data set to not only correctly classify the underlying screwdriving process with a high probability, but also to correctly identify faults with a high probability in the case of an erroneously performed screwdriving process, particularly when in addition to the data set associated with the faulty screwdriving process, a data set associated with the corresponding fault-free screwdriving process is available.

[0043] For example, if a user of the automatic screwdriving machine 10 discovers a faulty screw connection, and the user is unwilling or unable to determine the cause of the faulty screw connection by himself, the user can send a data set associated with the faulty screw connection together with a data set associated with a non-faulty screw connection to the data processing unit 110. The artificial intelligence 120 of the data processing unit 110 identifies the fault that occurred during the faulty screwdriving process by comparing the sent data sets, and the data processing unit 110 sends feedback indicating the identified fault to the control unit 14 of the automatic screwdriving machine 10, wherein the feedback can in particular include suggestions for correcting the fault. Such suggestions for correcting the fault can include, for example, suggesting the use of a different type of screw or increasing or decreasing the maximum torque at the end of the screwdriving process. If the artificial intelligence 120 is uncertain when identifying the fault, for example less than 80% certainty, an interaction with the user can take place, wherein the user can choose to provide additional data to the artificial intelligence 120.

[0044] It is also conceivable that the user of the automatic screwdriving machine 10 wishes to optimize an existing screwdriving process, for example in the form of an increased cycle time and / or a higher screw connection quality. In order to solve this optimization task, the user can send a plurality of data sets to the data processing unit 110, which ideally result from a fault-free screw connection. The artificial intelligence 120 can evaluate the sent data sets so that a suggestion for at least one changed machine setting or for a changed screwdriving program can be created, taking into account the optimization specifications similar to those sent. In uncertain situations, for example, if the artificial intelligence 120 is less than 80% sure about the requested change, an interaction with the user can also take place here, wherein the user has the option of providing additional data to the artificial intelligence. The suggestion for at least one changed machine setting or for a changed screwdriving program can then be sent from the remote data processing unit 110 to the control unit 14 of the automatic screwdriving machine 10, so that the user can make the suggested changes and the corresponding optimized screwdriving process can be performed.

[0045] It is further conceivable that a user of the automatic screwdriving machine 10 defines requirements for a new type of screwdriving process and inputs the requirements into the control unit 14. For example, the new type of screwdriving process may involve the use of different screws or the use of components formed of materials to be screwed together. The control unit 14 may send the user-defined requirements to the remote data processing unit 110, and the artificial intelligence 120 determines a screwdriving program that matches the user-defined requirements, and then sends the determined screwdriving program from the remote data processing unit 110 to the control unit 14 of the automatic screwdriving machine 10 in order to perform the new type of screwdriving process. As described above, during the execution of the new type of screwdriving process, a data set related to the screwdriving process may be recorded, and after the screwdriving process is completed, the recorded data set is sent to the data processing unit 110 for evaluation by the artificial intelligence 120, so that the artificial intelligence can determine a recommendation for an optimized screwdriving program and / or optimized machine settings if necessary, which is then sent back to the control unit 14 of the automatic screwdriving machine 10.

[0046] Reference numerals list

[0047] 10 Automatic screw tightening machine

[0048] 12 Base Unit

[0049] 14 Screwdriving unit

[0050] 16 Feed cylinder

[0051] 20 Drive Motor

[0052] 22 Drive shaft

[0053] 24 Screwing tools

[0054] 26 Positioning head

[0055] 28 Screws

[0056] 30 Feeding device

[0057] 32 Holding mechanism

[0058] 34 Claw Arm

[0059] 36 Handle

[0060] 80 Control Unit

[0061] 82 Memory

[0062] 90 Remote data connection

[0063] 100 Internet

[0064] 110 Data processing unit

[0065] 120 Artificial intelligence.

Claims

1. A method for controlling an automatic screw-twisting machine (10), the automatic screw-twisting machine (10) comprising: A screw-tightening unit (14), the screw-tightening unit (14) being used to automatically screw a screw (28) into a component; and a control unit (80) for controlling the screw-driving unit (14) based on a screw-driving program stored in the control unit (80), the screw-driving program defining the time course of the rotation speed required to drive the screw (28) during the screw-driving process, and optionally also defining the time course of the feed force to be applied to the screw during the screw-driving process, During the screw tightening process, the control unit (80) records a data set related to the screw tightening process, the data set comprising a time-dependent process of the following screw tightening parameters: an actual rotation speed of the screw (28), a torque applied to the screw (28), a feed position of the screw (28) and a feed speed of the screw (28); It is characterized in that The control unit (80) sends the recorded data set to a remote data processing unit (110), in which the screwdriving process is evaluated based on the sent data set with the aid of artificial intelligence (120).

2. The method according to claim 1, characterized in that The artificial intelligence is formed by a machine learning algorithm, for example a decision tree algorithm, a random forest algorithm, a support vector machine (SVM) algorithm or a k-nearest neighbor (kNN) algorithm.

3. The method according to claim 1 or 2, characterized in that: The artificial intelligence (120) has been trained using multiple heterogeneous data sets.

4. The method according to at least one of the preceding claims, characterized in that The recorded data set is stored in a memory (82) of the control unit (80) for a predetermined period of time or permanently.

5. The method according to at least one of the preceding claims, characterized in that The control unit (80) sends the recorded data set to the remote data processing unit (110) under the instruction of the user of the automatic screwdriving machine (10).

6. The method according to at least one of the preceding claims, characterized in that The control unit (80), under the instruction of a user of the automatic screwdriving machine (10), sends at least one first data set associated with a trouble-free screwdriving process and at least one second data set associated with a troublesome screwdriving process to the remote data processing unit (110).

7. The method according to claim 6, characterized in that The artificial intelligence (120) identifies a fault occurring during the faulty screwdriving process by evaluating the first data set and the second data set.

8. The method according to claim 7, characterized in that The remote data processing unit (110) sends feedback indicating the identified fault to the automatic screwdriving machine (10), wherein the feedback particularly includes a suggestion for correcting the fault.

9. The method according to at least one of the preceding claims, characterized in that The control unit (80), under the instruction of a user of the automatic screwdriving machine (10), sends to the remote data processing unit (110) a plurality of data sets and user-defined specifications for optimizing the screwdriving process, the plurality of data sets being in particular a plurality of data sets associated with the trouble-free screwdriving process.

10. The method according to claim 9, characterized in that The artificial intelligence (120) creates a suggestion for at least one changed machine setting and / or for a changed screwdriving program based on the sent multiple data sets and taking into account the sent optimization specifications, which suggestion is sent from the remote data processing unit (110) to the control unit (80) of the automatic screwdriving machine (10) in order to perform an optimized screwdriving process.

11. The method according to at least one of the preceding claims, characterized in that: A user of the automatic screwdriving machine (10) defines requirements for a new type of screwdriving process and inputs the requirements into the control unit (80); The control unit (80) sends the user-defined requirement to the remote data processing unit (110); The artificial intelligence (120) determines a screwdriving program that matches the requirements defined by the user; and The determined screwdriving program is sent from the remote data processing unit (110) to the control unit (80) of the automatic screwdriving machine (10) in order to execute the new screwdriving process.

12. The method according to claim 11, characterized in that: The data sets related to the screwdriving process are recorded by means of the determined screwdriving program during the execution of a novel screwdriving process; The recorded data set is sent to the data processing unit (110) and evaluated by the artificial intelligence (120); The artificial intelligence (120) determines a recommendation for an optimized screwdriving procedure and / or optimized machine settings based on the evaluation; as well as The optimized screwdriving program and / or the optimized machine settings are sent from the data processing unit (110) back to the control unit (80) of the automatic screwdriving machine (10).

13. A system for controlling an automatic screw-driving machine (10), the automatic screw-driving machine (10) comprising: A screw-tightening unit (14), the screw-tightening unit (14) being used to automatically screw a screw (28) into a component; and a control unit (80) for controlling the screw-tightening unit (14) based on a screw-tightening program stored in the control unit (80), the screw-tightening program defining the time variation of the rotation speed required to drive the screw (28) during the screw-tightening process, and optionally, the screw-tightening program also defining the time variation of the feed force to be applied to the screw during the screw-tightening process, The control unit (80) is configured to record a data set related to the screw tightening process during the screw tightening process, wherein the data set includes a time-varying process of the following screw tightening parameters: an actual rotation speed of the screw (28), a torque applied to the screw (28), a feed position of the screw (28), and a feed speed of the screw (28); It is characterized in that The automatic screwdriving machine (10) further comprises a remote data processing unit (110), the remote data processing unit (110) having an artificial intelligence (120), the recorded data set being capable of being sent by the control unit (80) to the remote data processing unit (110), and in the remote data processing unit (110), the screwdriving process being capable of being evaluated based on the sent data set with the aid of the artificial intelligence (120).

Images