Method for controlling hand-held power tool and hand-held power tool

Abstract

The invention relates to a method (200) for controlling a hand-held power tool (100), comprising: receiving (201) sensor data of at least one operating parameter (119) of the hand-held power tool (100); receiving (203) an input value for a control parameter of the hand-held power tool (100) based on a user input of a user of the hand-held power tool (100); ascertaining (205) a first target value of the control parameter on the basis of the sensor data and the input value; executing (207) a state determination module (107) on the sensor data and ascertaining an operating state (A, B, C) of the hand-held power tool (100); and controlling (209) the hand-held power tool (100) on the basis of the first target value of the control parameter and the determined operating state (A, B, C).

Description

Technical Field

[0001] The present invention relates to a method for controlling a handheld power tool and also to a corresponding handheld power tool, which is configured to carry out the method. Background Art

[0002] Methods for controlling handheld power tools are known from the prior art. Summary of the Invention

[0003] The object of the present invention is to provide an improved method for controlling a handheld power tool and a handheld power tool.

[0004] This object is achieved by a method for controlling a handheld power tool and a handheld power tool as described in the independent claims. Advantageous embodiments are the subject matter of the dependent claims.

[0005] According to one aspect of the present invention, a method for controlling a handheld machine tool is provided, comprising:

[0006] receiving sensor data of at least one operating parameter of the handheld power tool;

[0007] receiving an input value based on a user input of a user of the handheld power tool, the input value being for a control parameter of the handheld power tool;

[0008] Obtaining a first target value of a control parameter based on sensor data and input values;

[0009] executing a state determination module and applying it to the sensor data and ascertaining an operating state of the handheld power tool; and

[0010] The handheld power tool is controlled based on the first target value of the control parameter and the determined operating state.

[0011] This achieves the following technical advantages: an improved method for controlling a handheld power tool is provided, wherein, in addition to user input by a user of the handheld power tool, operating states determined during operation of the handheld power tool are also taken into account when controlling the handheld power tool. To this end, the operating state of the handheld power tool is determined based on sensor data of at least one operating parameter of the handheld power tool by executing a state determination module on the sensor data. The handheld power tool is then controlled taking into account the determined operating state. Depending on various embodiments, the control of the handheld power tool can be precisely adapted to the current operating state. Control can also be performed independently of user input. For example, user input can be adapted to the current operating state, enabling control of the handheld power tool that is appropriate to the operating state to be achieved even if the user input would otherwise cause other control that is inappropriate for the operating state.

[0012] According to one embodiment, controlling the handheld power tool includes:

[0013] determining a second target value for the control parameter based on the determined operating state of the handheld power tool;

[0014] Finding an output target value of the control parameter based on the first target value and the second target value; and

[0015] The output target value is output to an actuator of the handheld power tool for controlling the handheld power tool.

[0016] This achieves the following technical advantage: the control of the handheld power tool is precisely adapted to the current operating state. To this end, the handheld power tool is controlled based on a first target value and a second target value of a control parameter. The first target value is based on a user input by a user of the handheld power tool. However, the second target value is determined while taking into account the determined operating state of the handheld power tool. An output target value for the control parameter is then determined while taking into account the first and second target values ​​of the control parameter. This output target value is ultimately output to an actuator of the handheld power tool for controlling the handheld power tool.

[0017] By taking into account the first and second target values, the output target value can be determined taking into account not only the user input by the user but also the respective current operating state of the handheld power tool for controlling the handheld power tool. This has the advantage that, in particular, if the user input by the user (e.g., by actuating a trigger switch) does not allow for optimal control of the handheld power tool for the respective current operating state, by taking into account the second target value adapted to the current operating state, the control can be adapted to the current operating state even in the presence of inappropriate user input.

[0018] By taking into account the first target value based on the user input, the control of the handheld power tool remains primarily under the control of the user. Only for certain operating states can the user input be overridden or better adapted to the current operating state in the form of an output target value that takes into account the second target value. Thus, the method enables control of the handheld power tool that is adapted to its respective current operating state, taking into account target values ​​of control parameters adapted to the respective current operating state of the handheld power tool in addition to the user input.

[0019] In the sense of the present application, a target value is a desired value of a control parameter.

[0020] According to one embodiment, obtaining the output target value includes:

[0021] defining the output target value as the minimum or maximum of the first and second target values; and / or

[0022] The output target value is defined as the product of the first and second target values.

[0023] This achieves the following technical advantage: By defining the output target value as the minimum or maximum value between the first target value and the second target value, the output target value of the control parameter is determined as precisely as possible, thereby enabling the output target value to be determined as simply as possible. Depending on the determined operating state of the handheld power tool and the type of control parameter, the output target value of the control parameter optimally adapted to the respective determined operating state can be determined by selecting the minimum or maximum value between the first target value and the second target value. This enables the control of the handheld power tool to be adapted as precisely as possible to the respective current operating state.

[0024] The control parameter may be, for example, the motor speed or torque of the motor of the handheld power tool. The operating state of the handheld power tool may, for example, describe the progress of the work of the handheld power tool. For example, if the handheld power tool is designed as an electric screwdriver, the operating state of the handheld power tool may be described as indicating that the screw to be screwed in has been screwed into the corresponding workpiece in a form-fitting manner.

[0025] The desired motor speed or torque input by the user, which is input by operating a trigger switch of the handheld power tool, may be too high for such a positive fit of the screw to be screwed in. Therefore, the second output target value calculated taking into account the corresponding determined operating state (which exhibits a correspondingly lower motor speed or a correspondingly lower motor torque) results in the output target value of the control parameter describing a reduced speed or a reduced torque of the handheld power tool motor compared to the user input. Thus, the output target value allows for control of the handheld power tool that is optimally adapted to the respective current operating state.

[0026] By defining the output target value as the maximum or minimum of the first target value and the second target value, the target value that best suits the current operating state can be selected as the output target value. In particular, compared to the input parameter, the target value can be limited to a numerical range that is suitable for the current operating state.

[0027] By multiplying the first target value and the second target value, the second target value can be used as a sensitivity relative to the first target value. The second target value allows the first target value, based on the user input, to be increased or decreased accordingly by a factor in the form of the second target value, relative to the respective current operating state. Thus, the user input, and the user's control of the handheld power tool based thereon, can be effectively adapted to the respective current operating state.

[0028] According to one embodiment, obtaining the output target value includes:

[0029] If the second target value is less than the predetermined threshold, the output target value is defined as the first target value; if the second target value is greater than or equal to the predetermined threshold, the output target value is defined as the predetermined target value.

[0030] This achieves the following technical advantage: a still more precise output target value can be provided that is optimally adapted to the respective current operating state of the handheld power tool. To this end, depending on the second target value relative to a predetermined threshold value, either the first target value based on the user input or the predetermined target value is used as the output target value.

[0031] Taking into account the predetermined threshold value enables simple and precise adaptation of the output target value to the respective current operating state. Depending on the type of the respective current operating state, the predetermined target value can be adapted, like the predetermined threshold value of the control parameter, so that the correspondingly generated output target value enables optimized control of the handheld power tool.

[0032] Therefore, the second value of the control parameter determined based on the corresponding determined operating state can be used as a switching value for the first target value input by the user. Based on the determination of the second target value relative to a predetermined threshold, the system switches between the first target value input by the user and the predetermined target value as the output target value.

[0033] According to one embodiment, obtaining the output target value includes:

[0034] If the second target value is less than the predetermined threshold, the output target value is defined as the product of the first target value and the first predetermined target value; if the second target value is greater than or equal to the predetermined threshold, the output target value is defined as the product of the first target value and the second predetermined target value.

[0035] This achieves the following technical advantage: a more accurate output target value can be provided. Based on the relationship between a second target value determined for the determined operating state and a predetermined threshold, the output target value is defined as the product of the first target value input by the user and the first predetermined target value or the second predetermined target value. The first and second predetermined target values ​​can be configured as constant target values.

[0036] As in the previous embodiments, the predetermined target values, like the predetermined threshold values, can also be adapted to the current operating state. The first and second predetermined target values ​​serve here again as sensitivity values, which, by multiplication with the first target value, correspondingly increase or decrease the first target value and thus adapt to the respective current operating state.

[0037] Here, the second target value determined based on the operating state is used as a switching value by switching between the product of the first target value and the first predetermined target value and the product of the first target value and the second predetermined target value based on the relationship of the second target value to the corresponding predetermined threshold value. This allows the output target value to be adapted as precisely as possible to the corresponding determined operating state.

[0038] According to one embodiment, obtaining the operating status includes:

[0039] An event time is predicted, wherein the handheld power tool transitions from one operating state to another operating state at the event time.

[0040] This achieves the following technical advantage: In addition to actually ascertaining an existing operating state, the state determination module can additionally or alternatively predict an event time that defines a transition between different operating states of the handheld power tool. Thus, by predicting the event time, the control of the handheld power tool can be adapted to the upcoming transition to another operating state. This allows for the most precise possible control of the handheld power tool by enabling the control to adapt to events or operating states that have not yet occurred and into which the handheld power tool will enter in the future.

[0041] According to one embodiment, the control parameter includes one or more of the following: motor speed, motor current, and motor power of a motor of the handheld power tool.

[0042] This achieves the following technical advantage: the handheld power tool can be precisely controlled while taking into account the output target values ​​of the control parameters. The motor speed, motor current, or motor power of the motor of the handheld power tool are reliable control parameters based on which the handheld power tool can be controlled.

[0043] According to one embodiment, the operating parameters include one or more of the following: motor current, motor position angle, motor rotational speed, voltage of a voltage source of the handheld power tool, movement of the handheld power tool and / or vibrations in the handheld power tool.

[0044] This provides the following technical advantages: the operating parameters provide meaningful measured values ​​for determining the operating state. By measuring the motor current, the motor position angle, the motor rotational speed, the operating voltage of the power source of the handheld power tool, or the movement of the handheld power tool and / or vibrations in the handheld power tool, information about the effective force can be obtained, based on which the operating state of the handheld power tool can be determined.

[0045] For example, in the above example, by measuring the motor current, it is possible to detect whether the screw to be screwed in has been screwed into the corresponding workpiece to be processed in a form-fitting manner. When the form-fit is achieved, the change in the motor current can be detected, as can the motor rotational speed or motor speed, so that the operating state can be accurately determined based on this. For example, it is also possible to detect from the motion signal whether a handheld power tool (such as a screwdriver) has moved to the next work location and has therefore completed the previous working phase (i.e., the previous tightening process). This allows the operating state to be reset in a timely manner according to the new working phase (i.e., the new tightening process).

[0046] According to one embodiment, the operating state includes one or more of the following: the load range of the handheld machine tool, the vibration intensity in the handheld machine tool and / or on the workpiece being processed and / or in the user of the handheld machine tool, the temperature in the handheld machine tool and / or on the workpiece, the operating mode of the handheld machine tool, the working progress of the handheld machine tool, the material of the workpiece, and the existence of a positive fit between the handheld machine tool and the workpiece.

[0047] This achieves the following technical advantage: various operating states that the handheld power tool can be in or can enter can be taken into account when controlling the handheld power tool in the form of second target values ​​for the control parameters. Thus, the method according to the present invention can take various operating states into account, thereby providing a widely applicable control method.

[0048] According to one specific embodiment, the state determination module includes a trained artificial intelligence that is trained to ascertain an operating state of the handheld power tool and / or to predict an event time based on sensor data of operating parameters.

[0049] This achieves the technical advantage that a state determination module that is as reliable and high-performance as possible can be provided, which is configured to identify a current operating state or to predict a future event time.

[0050] A method for generating a training data set for training artificial intelligence to determine an operating state of a handheld power tool and / or predict an event time point thereof is provided, wherein the method comprises:

[0051] providing a plurality of measured values ​​of an operating parameter of the handheld power tool, wherein the measured values ​​are each provided with a time stamp;

[0052] identifying an event time within the plurality of measured values, wherein the event time defines a time at which the handheld power tool transitions from a first operating state to a second operating state;

[0053] The measurement values ​​are provided with label values, which are suitable for marking whether each measurement value is assigned to an event time point and / or an event time range;

[0054] Arranging the plurality of labeled measurements in a time series based on timestamps of the measurements; and

[0055] Provide a training dataset consisting of a time series of labeled measurements of operational parameters.

[0056] This achieves the following technical advantages: an improved method for generating a training data set for training an artificial intelligence to determine operating states of a handheld power tool and / or to predict the timing of events thereof is provided. Accordingly, an improved training data set can be provided for optimally adapting the training of an artificial intelligence to determine operating states or predict the timing of events within the context of the method for controlling a handheld power tool according to the above-described embodiments.

[0057] To this end, measured values ​​of operating parameters of the handheld power tool are first provided, each of which is time-stamped. The measured values ​​can be based on multiple measurements of the operating parameters of the handheld power tool or on multiple measurements of similar handheld power tools. The corresponding measured values ​​can be recorded, for example, during operation of the handheld power tool or during operation of multiple handheld power tools and provided as measured values ​​for the corresponding measurements of the method for generating a training dataset. The measured values ​​of the operating parameters recorded during operation of the handheld power tool can be transmitted, for example, to a server architecture provided for this purpose, which is configured to provide the corresponding measurement set for the method for generating a training dataset.

[0058] Furthermore, based on a timestamp defining the time at which each measured value was recorded, an event time is identified within the plurality of measured values ​​as a time at which the handheld power tool transitioned from a first operating state to a second operating state. Furthermore, the measured values ​​are provided with a label value that identifies each event time. Furthermore, the plurality of labeled measured values ​​are arranged in a time series, and the time series of labeled measured values ​​is provided as a corresponding training data set.

[0059] Based on the training data set generated in this way (in which the labeled measured values ​​of the operating parameters are arranged in a corresponding time series with respect to their timestamps), an artificial intelligence can be optimally trained for determining operating states or predicting event times. The individual labels of the measured values ​​of the operating parameters allow for determining whether the handheld power tool was in the first or second operating state at the time of the individual measured values.

[0060] Furthermore, at least one tag of the measured value allows precise determination of the event time of the transition between two operating states. The measured values ​​are provided with tag values ​​corresponding to the measured value tags known from the prior art, wherein a corresponding tag is assigned to each measured value of the operating parameter as an identifier.

[0061] According to one embodiment, the method further comprises:

[0062] For each measured value of the operating parameter, determining the rotation angle of the motor of the handheld power tool; and

[0063] Taking into account the time interval required for performing a complete motor revolution and the timestamp, a revolution time is calculated for each measured value of the operating parameter, wherein the revolution time defines the time at which a complete motor revolution is completed for the measured value starting from the respective timestamp.

[0064] This can achieve the following technical advantage: the training dataset is further improved. For each measured value of an operating parameter, the rotation angle of the handheld power tool's motor at the time the measured value of the operating parameter was recorded is determined. Furthermore, a rotation time is calculated for each measured value of the operating parameter, wherein the rotation time describes the time at which a complete rotation is performed starting from the corresponding rotation angle of the corresponding measured value of the operating parameter. Since both the timestamp and the rotation angle are present, all other sensor signals / quantities can be selectively correlated with time or rotation angle. This improves the dataset because states and events can be present and evaluated both in the temporal progression and in the rotation angle changes.

[0065] Based on the rotation angle or rotation time assigned to each measured value, the event time can be accurately predicted. This is achieved by knowing the rotation time at which the motor of the handheld power tool performs another complete rotation for each measured value of the operating parameter with a corresponding time stamp. This allows for an accurate prediction of the event time for each measured value of the operating parameter and the corresponding associated time stamp, taking into account the number of motor rotations required to achieve the transition to the second operating state. This enables optimized training of artificial intelligence.

[0066] According to one embodiment, identifying the event time point includes:

[0067] receiving sensor data reflecting an operating state of the handheld machine tool;

[0068] Find the event time points in the time series of sensor data;

[0069] synchronizing the time series of the operating parameters with the time series of the sensor data; and

[0070] Based on the event time points of the time series of the sensor data, event time points are identified in the time series of the operating parameters.

[0071] This achieves the following technical advantages: the method for generating a training data set and the correspondingly generated training data set can be further improved. For this purpose, sensor data from another sensor is used to identify the operating state or determine the time of an event. The sensor data reflects the hand tool in the corresponding operating state and is suitable for identifying the time of an event. For this purpose, the sensor data also includes a time stamp that defines the time at which the respective sensor data was recorded.

[0072] The sensor data is arranged in a corresponding time series based on its time stamps, and event times are determined within this time series based on the corresponding time stamps. Subsequently, the time series of the measured values ​​of the operating parameter and the time series of the sensor data are synchronized based on the corresponding time stamps. Based on the event times within the time series of the sensor data, the event times are then identified within the time series of the operating parameter. The additional information in the sensor data allows for more precise determination of event times within the time series of the measured values ​​of the operating parameter. This allows for a more accurate training dataset.

[0073] According to one embodiment, the sensor data are data from an external sensor, in particular a camera sensor.

[0074] This achieves the following technical advantage: the method for generating a training data set can be further improved. The sensor data is based on data from an external sensor. This external sensor can be configured, for example, as a camera sensor. The camera data from the camera sensor can reflect the operation of the handheld power tool, during which measured values ​​of operating parameters are recorded. By reflecting the operation of the handheld power tool, different operating states can be clearly reflected in the sensor data.

[0075] This allows for the unambiguous identification of the event time when the handheld power tool transitions from a first operating state to a second operating state. For example, the event time may describe the time at which a form-fitting connection between the screw and the workpiece is achieved when the screw is screwed into the workpiece in the above example. This time can be unambiguously identified using camera data from a camera sensor that reflects the tightening process of the handheld power tool. The event time can be unambiguously identified by considering the timestamp of the camera data, which respectively identifies the recording time of the camera data.

[0076] Therefore, by synchronizing the time series of operating parameter measurement data and camera data, the time at which the form-locking between the screw and the workpiece is achieved can be precisely identified in the time series of the operating parameter measurement data. By assigning corresponding labels to the operating parameter measurement values ​​(marking the corresponding identified event time), artificial intelligence can be trained based on the operating parameter measurement data to determine different operating states of the handheld machine tool or predict event time points. This enables the creation of a precise training data set, based on which artificial intelligence can be optimally trained to determine operating states or predict event time points.

[0077] A method for training artificial intelligence to determine the operating state of a handheld power tool is provided, wherein the method comprises:

[0078] Providing a training data set generated by the method for generating a training data set according to one of the above embodiments; and

[0079] Based on the training data set, artificial intelligence training is carried out to determine the operating state of the handheld power tool and / or to predict the time of events in the handheld power tool.

[0080] This achieves the following technical advantages: an improved method for training an artificial intelligence is provided, which is suitable for training the artificial intelligence to recognize operating states of a handheld machine tool and predict event times for transitions between operating states of the handheld machine tool according to the method for controlling a handheld machine tool according to the present invention. By using the training data set according to the present invention, the training of the artificial intelligence is adapted to the application scenario of the artificial intelligence in controlling a handheld machine tool according to the method for controlling a handheld machine tool, thereby achieving optimized training results and optimally trained artificial intelligence.

[0081] The method according to one of the above-mentioned embodiments provides an artificial intelligence for ascertaining an operating state of a handheld power tool and / or for predicting the timing of an event.

[0082] This achieves the following technical advantage: a state determination module that is as reliable and high-performance as possible is provided for controlling a handheld machine tool according to the method for controlling a handheld machine tool according to the present invention, the state determination module being configured to identify respective operating states of the handheld machine tool or to predict event times for transitions between operating states. This is achieved by using the method for training an artificial intelligence according to the present invention so that the artificial intelligence is configured for optimized use in controlling a handheld machine tool according to the method for controlling a handheld machine tool.

[0083] According to one embodiment, the artificial intelligence is designed as an artificial neural network, in particular a network having at least one recurrent layer with an internal state memory.

[0084] This achieves the following technical advantage: a particularly reliable and high-performance artificial intelligence can be provided. The artificial neural network can be designed, in particular, as a long short-term memory (LSTM) network. LSTM networks are particularly suitable for identifying operating states of handheld power tools and predicting event times for transitions between different operating states. LSTM networks are known in the prior art for pattern recognition and behavior or event prediction based on historical data.

[0085] Implementations of long short-term memory (LSTM) networks are suitable as neural network layers and can be combined with other network layers in a network.

[0086] According to another aspect, a computing unit is provided which is configured to execute the method for controlling a handheld machine tool according to one of the above-described embodiments and / or the method for generating a training data set according to one of the above-described embodiments and / or the method for training an artificial intelligence according to the present invention and / or the artificial intelligence according to the present invention.

[0087] According to another aspect, a computer program product is provided, which includes instructions that, when the program is executed by a data processing unit, cause the data processing unit to execute a method for controlling a handheld machine tool according to one of the above-mentioned embodiments and / or a method for generating a training data set according to one of the above-mentioned embodiments and / or a method for training an artificial intelligence according to the present invention and / or an artificial intelligence according to the present invention.

[0088] According to another aspect, a handheld power tool is provided, comprising a computing unit and at least one sensor for ascertaining sensor data of at least one operating parameter of the handheld power tool, wherein the computing unit is designed to execute a method for controlling the handheld power tool, the method comprising:

[0089] receiving sensor data of at least one operating parameter of the handheld power tool;

[0090] receiving an input value based on a user input of a user of the handheld power tool, the input value being for a control parameter of the handheld power tool;

[0091] Obtaining a first target value of a control parameter based on sensor data and input values;

[0092] executing a state determination module on the sensor data and determining an operating state of the handheld power tool;

[0093] determining a second target value for the control parameter based on the determined operating state of the handheld power tool;

[0094] Finding an output target value of the control parameter based on the first target value and the second target value; and

[0095] An output target value is output to an actuating device of the handheld power tool to control the handheld power tool.

[0096] As a result, the following technical advantage can be achieved: an improved handheld power tool is provided, which is configured to carry out the method for controlling a handheld power tool having the above-mentioned technical advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] The embodiments of the present invention are described below with reference to the accompanying drawings.

[0098] Figure 1 : A schematic diagram of a handheld power tool according to one embodiment;

[0099] Figure 2 : Another schematic diagram of a handheld power tool, in which various functional processes of the handheld power tool are illustrated;

[0100] Figure 3 : the temporal trends of the operating parameters of the handheld power tool and the rotation angle of the motor of the handheld power tool;

[0101] Figure 4 : A schematic diagram of a handheld power tool according to one embodiment, wherein the handheld power tool is shown in different operating states;

[0102] Figure 5 : A schematic diagram of a drive control of a handheld power tool according to one embodiment;

[0103] Figure 6 : A flow chart of a method for controlling a handheld power tool according to one embodiment;

[0104] Figure 7 : Another flow chart of a method for controlling a handheld power tool according to another embodiment;

[0105] Figure 8 : Another flow chart of a method for controlling a handheld power tool according to another embodiment;

[0106] Figure 9 : A schematic diagram of an artificial intelligence configured for use in controlling a handheld machine tool;

[0107] Figure 10 : A flowchart of a method for generating a training data set for training artificial intelligence according to one embodiment;

[0108] Figure 11 : Another flow chart of a method for generating an artificial intelligence training data set for training artificial intelligence according to another embodiment;

[0109] Figure 12 : A flowchart of a method for training an artificial intelligence; and

[0110] Figure 13 : Schematic diagram of a computer program product. DETAILED DESCRIPTION

[0111] Figure 1 A schematic diagram of a handheld power tool 100 according to one embodiment is shown.

[0112] The exemplary handheld power tool 100 includes a motor 101 having a motor control unit 103. The handheld power tool 100 also includes a computing unit 105 on which a state determination module 107 is installed and executable. The computing unit 105 including the state determination module 107 is configured to execute the method for controlling the handheld power tool 100 according to the present invention. The handheld power tool 100 also includes a power supply 111 and a current measuring device 113. The handheld power tool 100 also includes a trigger switch 109, by means of which the handheld power tool 100 can be controlled by a user. The handheld power tool 100 also includes an adjustment device 115, by means of which different operating modes of the handheld power tool 100 can be adjusted. Finally, the handheld power tool 100 includes a tool 117, by means of which corresponding work processes can be performed by the handheld power tool 100.

[0113] The handheld power tool 100 can be configured, for example, as a screwdriver or a battery-operated screwdriver. For this purpose, the tool 117 can be configured, in particular, as a receptacle for a replaceable screwdriver bit.

[0114] The motor control device 103 shown may include in particular the relevant power section of the motor control device. The motor 101 may also include a corresponding transmission device. Figure 1 The transmission is not explicitly shown.

[0115] The motor 101 can be designed, for example, as a mechanically or electronically commutated DC motor. The corresponding transmission can be designed as a planetary transmission.

[0116] According to the invention, the power part of the motor control device 103 can convert the control signal into the voltage or current curve required for the motor 101, for example, by PWM pulse width modulation. To this end, the control signal can first be converted into a corresponding digital signal, and the corresponding converted signal can then be transmitted via a suitable data bus (for example I2C or SPI). In the case of an electronically commutated DC motor, a corresponding rotating magnetic field can be generated, which can be tracked synchronously with the rotation of the rotor. The motor control can implement voltage-oriented or speed-oriented regulation. In voltage-oriented regulation, the motor speed decreases with increasing load (torque) at an increased operating current. Information about the motor speed or rotation angle can be derived system-inherently from the phase shift. Additionally or alternatively, a continuous rotation angle sensor (in Figure 1 This rotor position signal can be transmitted to the computing unit 105 for controlling the handheld power tool 100. This signal transmission can be performed via PWM, I2C, SPI or in an analog manner.

[0117] According to one embodiment, an algorithm is implemented in the control unit instead of the rotation angle sensor. This algorithm derives the rotation angle from the measured motor current and voltage (including a signal component generated by the voltage induced by rotor feedback). In this embodiment, executing the algorithm functionally replaces the rotation angle sensor.

[0118] The power supply 111 may be provided by a plurality of battery cells, such as lithium-ion batteries, and may be protected from overcharging, overcurrent, and deep discharge by a corresponding battery management system.

[0119] According to one embodiment, the trigger switch 109 can be embodied as a potentiometer, which provides an analog control signal for controlling the handheld power tool 100 to the computing unit 103, the control signal corresponding to a linear actuation of the trigger switch 109. By actuating the trigger switch 109, a corresponding user input for controlling the handheld power tool 100 can be provided.

[0120] The current measuring device 113 can determine the battery current of the power supply 111, which is dominated by the motor current of the motor 101. The control unit typically has a current consumption of less than 200 mA. A low-ohmic resistor or a Hall effect sensor can be used as the measuring element. Using an amplifier circuit and level control, an analog signal proportional to the current can be provided to the computing unit 103, which serves as the control unit of the handheld power tool 100.

[0121] The setting device 115 can be implemented as a rotary potentiometer, a toggle switch or a double switching element. It is necessary to implement a threshold value required for the automatic function that can be influenced by the user.

[0122] The computing unit 101 may include a microcontroller with conventional circuits (voltage regulator, clock source, EMC measures) and communication devices (Bluetooth, 4G, WLAN). The microcontroller may include an analog-to-digital converter and a digital interface to generate or detect signals for triggering the switch 109, the setting device 115, the current measuring device 113, the supply voltage, the motor speed, and the control signals of the motor 101. The microcontroller (in Figure 1 The state determination module 107 (not explicitly shown) can be executed to carry out the method according to the present invention for controlling the handheld power tool 100.

[0123] The handheld power tool 100 can be configured as a screwdriver, a rotary impact screwdriver, or a simple battery screwdriver. Alternatively, the handheld power tool 100 can be configured as an electric drill, a percussion drill, a hammer drill, or a rock drill.

[0124] Figure 2 A further schematic diagram of the handheld power tool 100 is shown, in which individual functional sequences of the handheld power tool 100 are illustrated.

[0125] Figure 2 A further schematic diagram shows an embodiment of a handheld power tool 100 according to the present invention. The handheld power tool 100 comprises a motor 101 , a computing unit 105 arranged on a printed circuit board 127 , an energy source 109 and a tool 117 .

[0126] The illustrated graphical representation shows some operating modes of the handheld power tool 100. Figure 1 Some components of the handheld power tool 100 are shown, while other components or parts are not shown in order to keep the illustrated diagram as simple as possible. However, the illustrated handheld power tool 100 may include Figure 1 All parts shown in .

[0127] In addition to the above, it can be constructed as a microcontroller or include such a microcontroller and have a Figure 1 In addition to the computing unit 105 of the state determination module 107 shown, a rotational speed sensor 129 (by means of which the motor speed of the motor 101 can be measured), a vibration sensor 131 (by means of which the vibration of the handheld power tool 100 can be measured) and an inverter 133 are also installed on the circuit board 127.

[0128] The user can transmit user input 139 to the computing unit 105 via the trigger switch 109 by means of electrical signal transmission 147. The power of the handheld power tool 100 can be adjusted via the user input 139 (including, for example, the trigger level of the trigger switch 109). Furthermore, the user input 139 can determine the direction of rotation of the motor 101 (which, for example, defines the tightening or drilling direction) or the operating mode of the handheld power tool 100 (which, for example, describes a tightening process with or without an impact function).

[0129] Based on user input 139, computing unit 105 controls the control device of handheld power tool 100 and outputs corresponding control signals to inverter 133 via electrical signal transmission 147. Inverter 133 outputs corresponding electrical energy transmission 141 to motor 101. Motor 101 transmits force 135 to tool 117 via corresponding force-torque transmission 143, with which workpiece 137 can be machined.

[0130] According to the present invention, the method for controlling the handheld power tool 100, executed by the computing unit 105, uses measured values ​​of operating parameters, based on which the operating state of the handheld power tool 100 is determined. These operating parameters may be, for example, motor current, motor power, and the speed or torque of the motor 101. In the illustrated embodiment, the motor speed of the motor 101 is used as an operating parameter based on which the method according to the present invention determines the operating state of the handheld power tool 100. These motor speeds are measured by the illustrated speed sensor 129, which detects the movement 145 of the motor 101. Furthermore, in the illustrated embodiment, vibrations / movements 145 of the handheld power tool 100, the tool 117, or the workpiece 137 are used as operating parameters. The vibrations / movements are measured by the vibration sensor 131. The corresponding measurement signals from the speed sensor 129 and the vibration sensor 131 are transmitted to the computing unit 105 for further processing.

[0131] The vibration sensor 131 may be designed as an acceleration sensor, for example.

[0132] According to the present invention, to control handheld power tool 100, state determination module 107 is executed to analyze the measured rotational speed or detected vibration and determine the current operating state of handheld power tool 100. Based on the determined operating state, the control of handheld power tool 100 is adjusted accordingly.

[0133] For a more detailed description of the method according to the invention for controlling a handheld power tool 100 , reference is made to the following description of the figures.

[0134] The force transmission 135 of the handheld power tool 100 can be realized as a direct drive or via a gear mechanism.

[0135] In addition, Figure 1 and Figure 2 The drive of the handheld power tool 100 , which is only schematically shown in FIG. 1 , can include different drive options, such as a percussion mechanism, a hammer mechanism, or a chisel mechanism.

[0136] As already mentioned, the force of the motor 101 is transmitted to the drive of the handheld power tool 100 via a transmission. This transmission can be designed, for example, as a shift transmission with two, three, or more gears. The transmission can be connected to a slip clutch, which in turn has a direct connection to the drive. Alternative solutions without a slip clutch are also conceivable, in which the transmission is directly connected to the drive.

[0137] The vibration or movement detected by the vibration sensor 131 may, for example, include movement of the handheld power tool 100 triggered by a user of the handheld power tool 100. Furthermore, the movement or vibration may be caused by a motor, a transmission, or a drive. Alternatively, the movement or vibration detected by the vibration sensor 131 may be caused by movement of a bit on a screwdriver head or by the force exerted by the screwdriver on the workpiece 137.

[0138] Figure 3 The operating parameters of the handheld power tool 100 and the rotation angle α of the motor 101 of the handheld power tool 100 are shown. rot The trend over time.

[0139] Graph a) shows the temporal progression of an operating parameter 119 of the handheld power tool 100. In the illustrated embodiment, the operating parameter 119 describes the motor current of the motor 101 of the handheld power tool 100. In the illustrated embodiment, the handheld power tool 100 is designed as a screwdriver, and the illustrated progression of the operating parameter 119 shows the temporal progression of the motor current during the tightening of a self-tapping screw into a workpiece 137 made of wood or a similar material.

[0140] The temporal progression of operating parameter 119 describes a time series 123 consisting of a plurality of chronologically arranged measured values ​​121 of the motor current. Measured values ​​121 are recorded by corresponding current sensors within handheld power tool 100 during operation of handheld power tool 100 (i.e., while screwing a screw into a workpiece).

[0141] Graph a) illustrates a typical curve of motor current I when driving a self-tapping screw into wood or similar material. Motor current I reflects the torque output by motor 101 via a torque constant in Newton meters per ampere. The torque generated by the starting current is used to accelerate the rotor of motor 101 at the beginning of the tightening process. This results in a peak in motor current I at the beginning of time sequence 123. After the rotor has started, the speed remains largely constant, resulting in a nearly horizontal curve for motor current I.

[0142] In diagram a), this range is characterized by operating state A (in which a starting peak is first detected, followed by an almost horizontal profile of the motor current I). In this operating state A, more than 90% of the applied torque is converted into actual mechanical work on the screw or workpiece, thereby driving the screw into the workpiece.

[0143] In diagram a), two event times 125 , 126 are also marked, at which the operating state changes from A first to operating state B and then to a further operating state C.

[0144] In the region of event time 126 , the motor current I in operating state A shows a uniform increase. This is because the screw is screwed further into the workpiece, so that as the screwing depth increases, a continuously greater torque is required, to generate which an increasing motor current I is required.

[0145] Compared to the relatively gradual increase in operating state A, the motor current I increases more steeply starting at event time 126. This steeper increase in motor current I is caused by the conical screw head of the self-tapping screw bearing against the surface of the workpiece being machined. At event time 126, the conical screw head of the self-tapping screw contacts the surface of the workpiece. Therefore, operating state B is characterized by the conical screw head being screwed into the workpiece. Due to the conical shape of the screw head, a higher torque is required to drive the screw head into the workpiece, resulting in a steeper increase in motor current I.

[0146] Conversely, event time 125 describes the transition from operating state B (in which the conical screw head is screwed into the workpiece) to operating state C (in which the screw head is completely screwed into the workpiece). Event time 125 describes the point at which the screw head ends flush with the workpiece surface. This point is also referred to as the flush point. Furthermore, the point in time marked by event time 126 (at which the conical screw head comes into contact with the workpiece surface) is also referred to as the pre-flush point.

[0147] In the exemplary embodiment shown, a flat profile of the motor current I is shown in operating state C. This can be achieved by fracture of the screw head or damage to the workpiece material when the screw is tightened further beyond the flush point.

[0148] Graph b) shows the rotation angle α rot The time trend of the rotation angle sensor is shown as an example. In the example shown, the sensor has a clear range of 360°. This is not necessary, and a smaller clear range is also sufficient. Usually, the clear range of the sensor can be coupled with the rotation speed of the motor 101. With the help of so-called phase unwrapping, the absolute rotation angle information of the entire tightening process from the start (by user manipulation of the trigger switch 109) to the desired closure can be obtained from the rotation angle sensor signal. For this purpose, it is possible to use the knowledge about the continuous rotation of the motor 101 to identify that the transformation from +180° to, for example, -179° actually corresponds to a rotation angle of +181°.

[0149] In the graph b), the rotation time t is also marked. rot At the rotation time point t rot The motor 101 performs one complete rotation. rot Or rotation angle α rot The knowledge of can be used below to predict the event times 125, 126. When the event time 126 is reached (when the conical screw head bears against the surface of the workpiece), the event time 125 (when the screw head ends flush with the surface of the workpiece) can be predicted knowing the rotation angle. Graph b) also shows the rotation angle difference Δ rot , that is, the rotation angle α corresponding to event time points 125 and 126 rot The difference between.

[0150] exist Figure 3 The situation shown in the figure merely illustrates a possible application scenario of the method according to the present invention. The present invention also covers other application scenarios in which the handheld power tool 100 is not configured as a screwdriver, but rather as a drill, chisel, or jigsaw, for example. More, fewer, or different operating states (A, B, C) than those described in this example are also conceivable.

[0151] For example, the handheld power tool 100 can be configured as a screwdriver. The possible operating states A, B, and C can describe the screwing depth of the screw to be screwed in, tightening in a specific tightening mode (impact mode, screwing in / out), operating efficiency, a limited torque, the selection of the screw used, or backlash at high torques.

[0152] Furthermore, the handheld power tool 100 can be configured as a drilling tool. The operating states A, B, and C can describe the drilling mode (hammer drilling, etc.), the orientation of the handheld power tool 100 relative to the surface of the workpiece being machined, the type of insert tool (metal drill, wood drill, stone drill), the drilling depth, the orientation of the handheld power tool 100 relative to the intended drilling location, the drilling of a hole for a cable or pipe, the punching out of a drill hole, the generation of drill chips, or the backlash in the event of a sudden jam of the handheld power tool 100.

[0153] Furthermore, the handheld power tool 100 can be designed as a hammer drill or a hammer drill. The operating states A, B, C can describe the operating mode, the progress of the work, the risk of damage, vibrations and noise, or the lack of pressing force.

[0154] Furthermore, the handheld power tool 100 can be designed as a chisel. The possible operating states A, B, C can describe an incorrect inserted tool, an active grip of the handheld power tool 100, or the progress of the work.

[0155] Furthermore, independently of the design of the handheld power tool 100, the possible operating states A, B, C can include aspects of work safety / comfort, such as vibration / noise generation, vibration monitoring, torque control with respect to the maximum torque, simple control of the handheld power tool 100, correct clamping of the tool in the handheld power tool 100 or recognition of situations (such as falling from a ladder, drilling into a cable or pipe).

[0156] By correspondingly training the artificial intelligence of the state determination module 107 , the listed operating states A, B, C or similar operating states that may occur during operation of the handheld power tool 100 can be ascertained based on the measured values ​​121 of the operating parameters 119 during operation of the handheld power tool 100 .

[0157] Operating parameters 119 may include, for example, motor current I or motor position angle, motor rotational speed, voltage of a power supply, movements and / or vibrations within the handheld power tool (100), or similar measurable parameters of the handheld power tool 100.

[0158] Figure 4 A schematic diagram of a handheld power tool 100 according to one embodiment is shown, wherein the handheld power tool 100 is shown in different operating states A, B, C.

[0159] Figure 4 Graphically describes the Figure 3The operating states A, B, and C of the handheld power tool 100 are described. The handheld power tool 100 is in turn configured as a screwdriver. The operating states A, B, and C describe the tightening process of screwing a screw 169 into a workpiece 137. The screw 169 can be, for example, a self-tapping screw, and the workpiece 137 is made of wood. For this purpose, the handheld power tool 100 also has a screwdriver bit 168.

[0160] FIG. a ) shows operating state A, in which screw 169 is screwed into workpiece 137 .

[0161] FIG. b ) depicts operating state B, which is characterized by the conical screw head 170 abutting against the surface 167 of the workpiece 137 and being screwed into the workpiece 137 in a subsequent process.

[0162] FIG. c ) depicts operating state C, which is characterized in that the screw 169 terminates flush with the surface 167 of the workpiece 137 and thus reaches a flush point.

[0163] As mentioned above, Figure 3 Graph a) shows the temporal progression of the motor current of the motor 101 of the handheld power tool 100 in different operating states A, B, C while the screw 169 is being screwed into the workpiece 137 .

[0164] In operating state A, in which the screw 169 is screwed uniformly into the workpiece 137 , the motor current I shows a largely horizontal curve, which increases slightly with increasing screw-in depth.

[0165] Operating state B is characterized by the conical screw head 170 abutting the surface 167 of the workpiece 137 and screwing the screw head 170 into the workpiece 137 , with the motor current I showing a sharp increase due to the increased torque required to screw the conical screw head 170 into the workpiece 137 .

[0166] In contrast, operating state C is characterized by the screw head 170 terminating at surface 167. Figure 3 The graph shows a flat trend of the motor current I, which may be caused by damage to the screw or damage to the workpiece 137.

[0167] exist Figure 4 In the embodiment shown, the handheld power tool 100 further includes an external sensor 171. In the embodiment shown, the external sensor 171 is designed as a camera sensor and is capable of recording camera data, with the aid of which the tightening process of screwing the screw 169 into the workpiece 137 can be reflected. As will be described in more detail with reference to the following figures, the corresponding camera data can be used to train the state determination module 107 or the artificial intelligence 149 of the state determination module 107.

[0168] Figure 5A schematic diagram of a drive control 197 of a handheld power tool 100 according to one specific embodiment is shown.

[0169] Figure 5 The adjustment chain of the handheld power tool 100 is shown.

[0170] According to the invention, the control chain comprises an inner control loop 193 and an outer control loop 191. Inner control loop 193 serves to control the drive of handheld power tool 100 based on user input 173 of a user.

[0171] In order to regulate the drive of the handheld power tool 100 solely via the inner control loop 193 based on the user input 173, the user first inputs 173. This can be accomplished, for example, by actuating the trigger switch 109. Thus, the user inputs can define first target values ​​for control parameters of the handheld power tool 100. Control parameters can include, for example, motor speed, motor power, and torque defined by the motor speed. Alternatively, control parameters can include the direction of rotation (for example, when screwing in or out a screw) or an operating mode (for example, impact mode or hammering mode). User input 173 describes the value of the control parameter.

[0172] By means of sensor measurement 175 of operating parameter 119 , an actual value of a control parameter is recorded, which describes the actual state of an actuator 195 of a drive control 197 of the handheld power tool 100 .

[0173] Sensor measurements 175 of operating parameters 119 may include measurements of motor current, motor speed, motor power, vibrations of the motor or handheld power tool 100 , or other effective operating parameters, with the aid of which operating states A, B, C can be determined.

[0174] The first target value of the user input 173 and the actual value of the sensor measurement 175 of the control variable are transmitted to the inner control loop 193 with the implementation of digital signal preprocessing 177 .

[0175] Corresponding control signals are output via the inner control circuit 193 to the actuator device 195 for controlling the handheld power tool 100 .

[0176] The outer control circuit 191 is now used to take into account the operating states A, B, C assumed by the handheld power tool 100 during operation when controlling the handheld power tool 100 .

[0177] To this end, the first target value of the control parameter input by the user 173 and, in particular, the actual value of the sensor measurement 175 of the control parameter undergoes model inference 183 after digital signal preprocessing 177. The aforementioned specific embodiment of the state determination module 107 is employed in the model inference 183. This is because the state determination module 107 is configured to identify the operating state A, B, or C in which the handheld power tool 100 is located based on the sensor measurement 175 of the control parameter or corresponding sensor data. Alternatively or additionally, the state determination module 107 can be configured to predict event times 125 and 126 at which a transition between the different operating states A, B, or C of the handheld power tool 100 occurs based on the sensor data of the sensor measurement 175 of the operating parameter 119.

[0178] State determination module 107 implemented in model reasoning 183 may be designed as a correspondingly trained artificial intelligence that is trained to determine operating states A, B, C or predict event times 125 , 126 based on measured values ​​of operating parameters 119 .

[0179] The information determined by the state determination module 107 in the form of model reasoning 183 about the current operating state A, B, C or the predicted event time 125 , 126 is provided to the outer control loop 191 after further processing 187 .

[0180] The outer control loop 191 is then configured to define corresponding second target values ​​for the control parameters based on the information from the model reasoning 183 about the current operating state A, B, C or the predicted event time 125, 126. The second target values ​​for the control parameters defined by the outer control loop 191 are coordinated with the respective current operating state A, B, C or the respective predicted event time 125, 126. Taking into account the second target values ​​for the control parameters generated by the outer control loop 191, the control of the handheld power tool 100 can be optimally adapted to the respective current operating state A, B, C or the respective predicted event time 125, 126.

[0181] The second target value for the control parameter generated by the outer control loop 191 is then supplied to the inner control loop 193 .

[0182] According to the present invention, the inner control loop 193 is now configured to calculate an output target value while taking into account a first target value for a control parameter provided in a user input 173 by a user and a second target value for a control parameter provided by an outer control loop 191 while taking into account the current operating state A, B, C or the predicted event time point 125, 126, and to control the actuator 195 of the handheld machine tool 100 based on the output target value.

[0183] In this case, the output target value can be calculated by inner control loop 193, for example, as the product of a first target value of user input 173 and a second target value of outer control loop 191. By multiplying the first and second target values, the first target value of user input 173 can be sensitized, i.e., adapted to the respective operating state A, B, C or expected event time 125, 126, by the second target value of outer control loop 191 (which is determined based on the current operating state A, B, C or the predicted event time 125, 126). Alternatively, the output target value can be defined as the minimum or maximum of the first and second target values. Thus, the output target can be defined as the value of the first and second target values ​​that best matches the respectively determined operating state A, B, C.

[0184] Alternatively, if the second target value of outer control loop 191 is less than a predetermined threshold value, the output target value can be defined by the first target value of user input 173; if the second target value is greater than or equal to the predetermined threshold value, the output target value can be defined as a predetermined target value. The predetermined target value can be given as a constant value of a control parameter, which, for example, has already been adapted to the respective current operating state A, B, C or expected event time 125, 126 during the presetting of handheld power tool 100. The predetermined threshold value can be adapted empirically, for example, through appropriate measurements, to the respective operating state A, B, C and any second target value.

[0185] Alternatively, if the second target value of outer control loop 191 is less than a predetermined threshold, the output target value can be defined as the product of the first target value of user input 173 and the first predetermined target value. If the second target value of outer control loop 191 is greater than or equal to the predetermined threshold, the output target value can be defined as the product of the first target value of user input 173 and the second predetermined target value. Thus, based on the second target value determined by outer control loop 191, the first target value of user input 173 can be sensitized or adapted to the respective current operating state or expected event time in the form of first and second predetermined target values. These first and second predetermined target values ​​can each be defined as constant values ​​of a control parameter and adapted to the respective current operating state or expected event time. The first and second predetermined target values ​​and the predetermined threshold can, in turn, be determined empirically for possible operating states A, B, and C.

[0186] Therefore, the hand-held machine tool 100 can be controlled in the form of an output target value determined by the internal control loop 193 taking into account the user input 173 manipulated by the user of the hand-held machine tool 100 and the corresponding first target value of the control parameter, as well as taking into account the second target value of the control parameter determined regarding the current operating state A, B, C or the expected event time point 125, 126.

[0187] exist Figure 3 and Figure 4 In an embodiment, the hand-held power tool 100 is constructed as a screwdriver, wherein the tightening of a self-tapping screw 169 into a wood workpiece 137 is described, and wherein different operating states A, B, and C respectively describe the screw 169 being screwed into the workpiece 137 in operating state A, the conical screw head 170 being screwed into the workpiece 137 in operating state B, and the screw head 170 ending flush with the surface 167 of the workpiece 137 in operating state C. The control parameter can be, for example, the rotational speed of the motor 101 of the hand-held power tool 100.

[0188] User input 173 may include, for example, a signal of trigger switch 109 actuated by a user, with the aid of which a respective first target value for the motor speed is defined.

[0189] Operating parameter 119 determined in sensor measurement 175 may be given, for example, by motor current I of motor 101 of handheld power tool 100 .

[0190] The state determination module 107 applied to the measured value of the motor current I according to the model reasoning 183 can be correspondingly configured to be used for determining the state of the motor current I based on the motor current I (according to Figure 3 The graph A) of FIG. 1 shows different operating states A, B, C. Furthermore, the state determination module 107 can be configured to predict corresponding event times 125 , 126 based on the measured values ​​121 .

[0191] In the following, according to Figure 3 and Figure 4 The above process is described in an embodiment in which the handheld power tool 100 is configured as a screwdriver and the screw 169 is screwed into the workpiece 137 by the screwdriver and in particular the second target value is determined by the outer control loop 191 or the output target value is determined by the inner control loop 193.

[0192] Based on the user input 173 for actuating the trigger switch 109, a correspondingly higher value for the motor speed can be defined as the first target value. Conversely, based on the sensor measurement 175 of the motor current, the state determination module 107 determines that the handheld power tool 100 is already at the curve Figure 3In operating state C, the flush point has been reached in the graph, and screw head 170 has come to rest with surface 167 of workpiece 137. On this basis, outer control loop 191 calculates a significantly lower value for the motor speed as a second target value to prevent damage to the screw or workpiece 137 that could occur in the event of a high motor speed input by user 173. Thus, by determining the output target value by inner control loop 193, a significantly excessive target value for the motor speed input by user 173 can be reduced by the significantly lower second target value for the motor speed calculated by outer control loop 191, so that handheld power tool 100 can be controlled according to current operating state C and, if necessary, shut down.

[0193] For this purpose, the second target value, like the first and second predetermined target values, can assume the value 0. The output target value can also be reduced to the value 0, whereby the handheld power tool 100 can be stopped.

[0194] In the illustrated embodiment, information from model reasoning 183 may also be displayed in status display 189 .

[0195] Furthermore, the model inference can undergo a reset process, in which the model execution is reset to initial values. This can be performed, for example, during individual tightening processes. For example, the state determination module 107 can be reset for each new tightening process (in which a single screw is screwed into or out of the workpiece 137). Alternatively or additionally, the reset can also be performed when the handheld power tool 100 is turned on or off.

[0196] For this purpose, a reset preprocessing 181 is first performed based on the sensor measurements 175 and based on this a reset decision 185 is made. The reset decision 185 can also be brought about by taking into account the results of the model inference 183 .

[0197] The information for model reasoning 183 and subsequent processing 187 may be provided in digital or quasi-analog form.

[0198] Figure 6 A flow chart of a method 200 for controlling a handheld power tool 100 according to one embodiment is shown.

[0199] To control the handheld power tool 100, in a first method step 201, sensor data of at least one operating parameter 119 of the handheld power tool 100 are first received. The operating parameters 119 may include, for example, motor current, motor position angle, motor speed, supply voltage, measured values ​​of movement or vibration of the handheld power tool 100, or similar parameters, from which operating states A, B, and C of the handheld power tool 100 can be determined.

[0200] In a further method step 203, on this basis, input values ​​for control parameters of the handheld power tool 100 are received based on user input 173 by a user of the handheld power tool 100. The control parameters may include, for example, motor speed, motor current, motor power, actuation of the trigger switch 109, input regarding an operating mode (e.g., switching an impact mode or a chiseling mode on or off), the direction of rotation of the motor 101 (e.g., for screwing in or out a screw 169), or similar control parameters.

[0201] In a further method step 205 , a first target value for the control parameter is ascertained based on the sensor data of the operating parameter 119 and the input value of the user input 173 regarding the control parameter.

[0202] In a further method step 207, the state determination module 107 is executed and applied to the sensor data of the operating parameters 119. The execution of the state determination module 107 determines the operating states A, B, and C of the handheld power tool 100. The determination of the operating states may include predicted event times 125 and 126 at which a transition between two operating states A, B, and C occurs. The operating states A, B, and C may include the load range in which the handheld power tool 100 is operated, the vibration or movement intensity of the handheld power tool 100 and / or the workpiece 167 being processed, the temperature within the handheld power tool 100 and / or on the workpiece 167, the different operating modes in which the handheld power tool 100 is operated, the progress of the work of the handheld power tool 100, the material of the workpiece 137 being processed, or, for example, the already described form-fit connection between the handheld power tool 100 and the workpiece 137 (where the screw head 170 of the screw 169 to be screwed in terminates flush with the surface 167 of the workpiece 137). Alternatively, other and additional operating states of the handheld power tool 100 are conceivable, which can be detected by analyzing measurable operating parameters 119 of the handheld power tool 100 .

[0203] In a further method step 209, the handheld power tool 100 is controlled based on the first target value of the user input 173 and taking into account the determined operating state A, B, C of the handheld power tool 100. As described above, for this purpose, the first target value of the control parameter of the user input 173 can be adapted to the current operating state A, B, C in order to achieve a control adapted to the respective current operating state A, B, C.

[0204] Controlling 209 may include, for example, adjusting the motor speed, the motor torque or other control parameters.

[0205] Figure 7 A further flow chart of a method 200 for controlling a handheld power tool 100 according to another specific embodiment is shown.

[0206] Figure 7 The implementation method is based on Figure 6 In accordance with an embodiment of the present invention, method 100 includes all method steps shown there.

[0207] In the specific embodiment shown, method step 211 is first carried out in order to control 209 the handheld power tool 100 .

[0208] In method step 211, a second target value of the control parameter is determined based on the determined operating states A, B, C of the handheld power tool 100 or the predicted event times 125, 126. Figure 5 As described, the second target value can be determined via outer control loop 191. Determining the second target value can include calculating the second target value, wherein the corresponding second target value is calculated during execution and upon recognition of the respective current operating state A, B, or C or prediction of the expected event time 125 or 126. Alternatively, the determination can include reading the corresponding second target value from a lookup table, in which suitable second target values ​​for different possible operating states A, B, or C or predicted event times 125 or 126 are pre-stored. Alternatively, state determination module 107 can also be configured to determine a correspondingly adapted second target value for the control parameter based on the determined operating states A, B, or C.

[0209] In another method step 213, an output target value of the control parameter is determined based on the first target value and the second target value. Figure 5 As explained, the output target value is determined via inner control loop 193. This determination can also include calculating the corresponding output target value based on the first and second target values. Alternatively, the corresponding output target value can be read from a pre-stored lookup table for this purpose. Alternatively, state determination module 107 can also be configured to determine the output target value of the control parameter based on the first and second target values.

[0210] In a further method step 213 , the output target value is output to the actuator 195 of the drive control 197 of the handheld power tool 100 in order to control the handheld power tool 100 .

[0211] Figure 8 A further flow chart of a method 200 for controlling a handheld power tool 100 according to another specific embodiment is shown.

[0212] The embodiment shown is based on Figure 7 and includes all method steps described there.

[0213] In the specific embodiment shown, ascertaining 207 the operating states A, B, C also includes predicting 225 the event times 125 , 126 at which the transition between the different operating states A, B, C occurs.

[0214] Furthermore, the determination 213 of the output target value includes a method step 217. In a method step 225, the output target value is defined as the minimum or maximum of the first and second target values.

[0215] Furthermore, method step 213 includes method step 219. In method step 219, if the second target value is less than the predetermined threshold, the output target value is defined as the first target value. If the second target value is greater than or equal to the predetermined threshold, the output target value is defined as the predetermined target value.

[0216] In method step 221 , the output target value is defined as the product of the first and second target values.

[0217] In method step 223, if the second target value is less than the predetermined threshold, the output target value is defined as the product of the first target value and the first predetermined target value. If the second target value is greater than or equal to the predetermined threshold, the output target value is defined as the product of the first target value and the second predetermined target value.

[0218] The predetermined target value can be configured as a constant value of the control parameter and pre-stored. Corresponding threshold values ​​can also be pre-stored. The defined target value can also optimize the operation of the handheld power tool 100 for the respective current operating state A, B, C or expected event time 125, 126, just like the predetermined threshold value.

[0219] As about Figure 5 As already explained, in the embodiment shown, the output target value is determined 213 via inner control loop 193. The predetermined target value, like the second target value determined via outer control loop 191, can assume the value 0. This allows the handheld power tool 100 to be shut down depending on the selection of the control parameters.

[0220] By taking into account predetermined or second target values ​​according to described method steps 215 to 221 , the first target value of the control parameter based on user input 173 by the user can be adapted to the respective current operating state A, B, C or expected event time 125 , 126 .

[0221] Figure 9 A schematic diagram of an artificial intelligence 149 configured for use in controlling a handheld power tool 100 is shown.

[0222] State determination module 107 may include a correspondingly trained artificial intelligence 149, which is trained to determine the current operating state A, B, C or the predicted event time 125, 126 based on the measured values ​​of operating parameter 119. State determination module 107 may also be configured to determine a second target value or output a target value.

[0223] Figure 9 An embodiment of such an artificial intelligence 149 is shown, which can be used to determine operating states A, B, C or to predict event times 125 , 126 .

[0224] In the embodiment shown, artificial intelligence 149 is designed as an artificial neural network, in particular a long short-term memory (LSTM) network.

[0225] In the embodiment shown, the artificial neural network comprises an input layer 153 for receiving input data 151. The input data may comprise sensor data of the operating parameters 119 in a correspondingly pre-processed form.

[0226] Furthermore, the artificial neural network comprises two dense layers 155 and two pooling layers 157, which are arranged one after the other in an alternating manner. Furthermore, the artificial neural network comprises two long short-term memory layers 159, between which a dropout layer 161 is arranged. Finally, the artificial neural network further comprises two dense layers and an output layer 163.

[0227] In data processing, downsampling 164 is first performed through the input layer 153, the first two dense layers 155, and two pooling layers 157. The next two long short-term memory layers 159 and the dropout layer 161 located in between them lead to feature extraction 165. The last two dense layers 155 and the output layer 163 enable prediction 166.

[0228] Deviating from the embodiment shown, the artificial intelligence 149 used can also be constructed with another model architecture that can perform regression or classification based on the time series 123 of the measured values ​​121 of the operating parameter 119. However, a prerequisite for the model architecture of the artificial intelligence 149 used is that the corresponding model can be provided in a format that can be executed on the microcontroller of the handheld power tool 100.

[0229] For the implementation described here, a Tensorflow / Keras model with the following mapping architecture can be used: After training, the model can first be converted to the Tensorflow-Lite format and then converted to C code for a microcontroller using the TVM converter.

[0230] The three input channels may be, for example, the motor current I of the motor 101 , the trigger voltage of the trigger switch 109 , and the motor speed per second.

[0231] The architecture used can be constructed as follows:

[0232] Layer (Type) Output format Number of parameters denseG0_8 (dense layer) (None, None, 8) 32 pool0_4 (average pooling layer) (None, None, 8) 0 denseG1_16 (dense layer) (None, None, 16) 114 pool1_4 (average pooling layer) (None, None, 16) 0 lstm0_32 (LSTM layer) (None, None, 32) 2272 dropout0_0.25 (dropout layer) (None, None, 32) 0 lstm1_8 (LSTM layer) (None, None, 8) 1312 denseG2_4 (dense layer) (None, None, 4) 36 dense0_1 (dense layer) (None, None, 1) 5

[0233] Absolute number of parameters used: 7801

[0234] Trainable parameters: 7801

[0235] Non-trainable parameters: 0

[0236] The first dense layer 155 may be configured with a 6x8 core and 8 biases. The second dense layer 155 may be configured with an 8x16 core and 16 biases. The first LSTM layer 159 may be configured with a 16x128 core, a 32x128 recurrent core, and a 128 bias. The second LSTM layer 159 may be configured with a 32x32 core, an 8x32 recurrent core, and a 32 bias. The third dense layer 155 may be configured with an 8x4 core and 4 biases. The fourth dense layer 155 may be configured with a 4x1 core and 1 bias.

[0237] The dense layer 155 and the LSTM layer 159 may be constructed with a TanH activation function.

[0238] Figure 10 A flow chart of a method 300 for generating a training data set for training artificial intelligence 149 is shown according to one embodiment.

[0239] To train artificial intelligence 149, according to the present invention, a training data set is first generated. To this end, method 300 according to the present invention first provides, in method step 301, a plurality of measured values ​​121 of operating parameters 119 of handheld power tool 100. Each measured value 121 has a time stamp defining the time at which the measured value 121 was recorded. The measured values ​​121 can be recorded during operation of handheld power tool 100. Alternatively, the measured values ​​121 can be recorded during operation of multiple different handheld power tools 100 of the same type. The measured values ​​121 of operating parameters 123 can be recorded by multiple different handheld power tools 100 and transmitted, for example, to a server architecture configured to generate the training data set. The server architecture can process and archive the measured data accordingly in order to generate the corresponding training data set based thereon.

[0240] For this purpose, in a further method step 303 , event times 125 are identified within the plurality of measurement data 121 at which operating state transitions occur between two operating states A, B, C. The identification of event times 125 is performed based on the time stamps of the individual measurement values ​​121 .

[0241] In a further method step 305, the measured values ​​121 are provided with label values ​​suitable for assigning the respective measured values ​​to operating states A, B, and C or for identifying these operating states with respective event times 125 and 126. The provision of label values ​​can be performed, for example, by an algorithm, suitably configured for this purpose and known from the prior art, for labeling training data for artificial intelligence 149. Alternatively, the labeling of measured values ​​121 can be performed manually by appropriately trained personnel. Alternatively, measured values ​​automatically labeled by the algorithm can be monitored by appropriately trained personnel to ensure that the labels are correct.

[0242] In a further method step 307 , the tagged measured values ​​121 of the operating parameter 119 are arranged in a time series 123 . This is done based on the time stamps. The time series 123 describes the temporal arrangement of the tagged measured values ​​121 according to the time stamps.

[0243] After method step 307 and before a further method step 309, a plurality of existing labels can be merged. This can be done, in particular, based on statistical criteria. For example, by calculating the median of event time points from a plurality of manual and automatic sources, a more robust label can be created for the actual event time point, thereby achieving a higher accuracy of the time point label.

[0244] In a further method step 309 , a time series 123 of labeled measured values ​​121 of the operating parameter 119 is provided as a corresponding training data set.

[0245] Figure 11 A further flow chart of a method 300 for generating a training data set for training artificial intelligence 149 according to another specific embodiment is shown.

[0246] The embodiment shown is based on Figure 10 and includes all method steps described there.

[0247] In the embodiment shown, first in method step 211 , the rotation angle α of the motor 101 of the handheld power tool 100 is ascertained for each measured value 121 of the operating parameter 119 . rot To this end, the rotation angle α of the motor 101 is measured by a corresponding rotation angle sensor 129 rot and is assigned to the measured value 121 based on its time stamp. The rotation angle α of the motor 101 rotHere, the rotation angle of motor 101 at the time when the corresponding measured value 121 of operating parameter 119 was recorded is described.

[0248] In a further method step 313 , a rotation time t is calculated for each measured value 121 of the operating parameter 119 , taking into account the time interval required for carrying out one complete rotation of the motor 101 and taking into account the time stamp of the corresponding measured value 121 . rot Rotation time point t rot Here the rotation angle α is described from the corresponding measured value 121 rot The point in time at which one full rotation of the motor 101 begins.

[0249] In the embodiment shown, the identification 303 of the event times 125, 126 also includes a method step 315. In this method step, sensor data from the external sensor 171 are received. The sensor data from the external sensor 171 reflect the operating state A, B, C of the handheld power tool 100. The external sensor 171 can be designed as a camera sensor, for example. The camera data of the camera sensor can be used to reflect the operating state A, B, C of the handheld power tool 100 at the time when the measured value 121 of the operating parameter 119 was recorded. Figure 4 As shown in FIG, the progress of the process of driving a screw 169 into a workpiece 137 can be reflected or tracked by means of a camera sensor 171 , which is provided on the handheld power tool 100 , for example.

[0250] In a further method step 317, the event times 125, 126 are determined within the time series of the sensor data of the external sensor 171. For this purpose, the sensor data of the external sensor 171 are arranged into a time series according to their time stamps. By analyzing the image data of the camera sensor, it is possible to identify, for example, the exact time at which the screw head 170 ends flush with the surface 167 of the workpiece 137, as in Figure 4 As indicated in the operating state C of the camera sensor 171. Therefore, the exact time of the event time 125 at which the event occurred can be determined within the time series of the image data by means of the corresponding time stamp of the recorded image data of the camera sensor 171. Figure 4 The transition between the illustrated operating states B and C thus results in a flush termination between the screw head 170 of the screw 169 and the surface 167 of the workpiece 137 .

[0251] In a further method step 319 , time series 123 of measured values ​​121 of operating parameter 119 is synchronized with the time series of sensor data of external sensor 171 by means of corresponding time stamps of measured values ​​121 or sensor data of external sensor 171 .

[0252] In a further method step 321, event times 125, 126 are identified in time series 123 of measured values ​​121 of operating parameter 119 based on the event times 125, 126 identified in the time series of sensor data from external sensor 171. By synchronizing time series 123 of measured values ​​121 of operating parameter 119 (e.g., given by motor current I of motor 101) with the time series of sensor data from external sensor 171, each measured value 121 of operating parameter 119 can be assigned image data or sensor data from external camera sensor 171 that were recorded simultaneously.

[0253] By checking or analyzing the image data of the external camera sensor 171 (for example by correspondingly trained personnel), it is possible to precisely describe the various operating states A, B, C or the various event times 125, 126 (for example at Figure 4 By screwing the screw 169 into the workpiece 137, screwing the screw head 170 into the workpiece 137 and terminating the conical screw head 170 flush with the surface 167 of the workpiece 137, each measured value 121 can be assigned to an operating state of the handheld machine tool 100 due to the synchronization of two time series (i.e., the time series 123 of the measured values ​​121 of the operating parameter 119 and the time series of the sensor value of the external sensor 171).

[0254] By comparing the image data of the external camera sensor 171 (wherein correspondingly trained personnel can easily identify the different operating states of the handheld power tool 100) and by synchronizing the two time series, the measured values ​​121 of the operating parameters 119 can be precisely assigned to the different operating states A, B, C.

[0255] Thus, by taking into account the sensor data of external sensor 171, measured values ​​121 of operating parameter 119 can be precisely labeled and assigned to clearly defined operating states via corresponding labels. This can also be done in particular for operating parameters 119 whose profile does not clearly disclose the individual different operating states A, B, C or the current event times 125, 126.

[0256] Therefore, by considering Figure 4 The image data of the external camera sensor 171 in the embodiment can be accurately based on Figure 3 The measured values ​​121 of the time series 123 of the course of the motor current I in the graph A of FIG. 1 determine the different operating states A, B, C and the current event times 125, 126, while Figure 3 The corresponding curve of the motor current I shown in the diagram A does not explicitly indicate different operating states A, B, C or event times 125 , 126 .

[0257] Instead of the above-described method steps, further steps can also be performed to generate a training data set. For example, the recorded measured values ​​121 of the operating parameters 119 or the sensor data of the external sensor 171 can be converted into standard units. In addition, the rotation angle α can be preprocessed. rot Especially when the rotation angle α rot When the measured values ​​are recorded by the Hall sensor, the possibility of overflow of the Hall sensor data with respect to the unit 16 of the data type used can be taken into account.

[0258] In particular, the rotation angle α of the records, which may overflow due to continuous operation and data type, can be set rot Correction to a continuous value range This can advantageously be achieved by shifting the angle values ​​to a starting value of zero, whereby the rotation angle reflects the angle change relative to the starting point (eg of the tightening process) at any point in time.

[0259] Furthermore, the time series 123 of the measured values ​​121 of the operating parameter 119 can be shortened to the time range of the actual operation of the handheld power tool 100, just like the time series of the sensor data of the external sensor 171. Figure 4 In the embodiment shown, the time series can be limited to the actual tightening of screw 169 into workpiece 137. Measurement values ​​recorded before the start and after the end of the tightening process can be removed from the time series. Alternatively, all data before or after a specific time point before or after the labeled event time points 125 and 126 can be removed to shorten time series 123 and reduce the amount of data.

[0260] Furthermore, the sampling rate of the recording of measured values ​​or sensor data can be reduced. Furthermore, filtering of the recorded measured values ​​or sensor data can be initiated.

[0261] Additionally, feature extraction can be performed to obtain derived sensor signals, such as the vector length of acceleration, the vector length of rotation rate from a gyroscope, and the number of motor revolutions per second.

[0262] Furthermore, the training data can be divided into training data, validation data, or test data, as is common in machine learning. For example, the division can be as follows: 70% training data, 20% test data, and 10% validation data.

[0263] In the process of dividing the data into training, validation, and test data, you can perform what is called stratified sampling.

[0264] However, to demonstrate the generalization of the function within a typical operating range, different sampling methods can be advantageously used, in particular by using only certain operating ranges for training and other operating ranges for validation or testing. In this case, the desired generalization is demonstrated by the accuracy of the function on the test dataset being only slightly lower than in the stratified sampling case. Suitable operating parameters include, in particular, screw length, rotational speed, screw diameter, wood type and hardness, and other application parameters.

[0265] Furthermore, normalization parameters can be calculated. For example, a min-max scaling method (parameters per signal: minimum, maximum) or a standardization method (parameters per signal: mean and standard deviation) can be used. The parameters can be calculated solely based on the training data set. It is important to ensure that the normalization parameters are not distorted by the selected filling method.

[0266] Because according to the operation performed by the handheld power tool 100 (for example, Figure 3 and Figure 4 (e.g., the tightening scenarios described in the embodiment of FIG. 2 ), each time series 123 may have different lengths. Since tightening scenarios of varying lengths are performed, the training data can be trimmed to a uniform length. This is necessary because the input data for artificial intelligence must have a uniform data format. Further considerations can be made regarding whether the padded data is zero in all signals or whether it constitutes a copy of the first (start padding) or last (end padding) real data.

[0267] In About Figure 9 In the described neural network architecture, due to the network architecture used (in which effective downsampling is performed within the network through pooling layers), padding is additionally used to ensure that all time series data are divisible by the total pooling factor Fp, which is generated by the product of the individual pooling factors in the network.

[0268] This ensures that for each group of Fp samples there is an unambiguous assignment of output value labels. In the suitable network architecture shown, Fp=16 results from the pooling layer.

[0269] However, the test dataset may be used without padding, where the results cannot usually be computed for the last sample.

[0270] Alternatively, an enhancement technique may be performed to thereby generate new data that in principle corresponds to the real data. For example, a new time series of data of the operating parameters 119 may thereby be generated.

[0271] Finally, normalization can be performed based on the previously calculated normalization parameters (normalization or min-max scaling).

[0272] In addition to being implemented as a camera sensor, the external sensor 271 can also be designed as a microphone, for example. A plurality of external sensors can also be used.

[0273] Data acquisition can be performed on various devices, such as a microcontroller and a Raspberry Pi. Data on operating parameters 119 on the handheld power tool 100 and data from external sensors can be recorded and transmitted to an external server architecture. Due to differences in sampling rates between the various sensors 171 and the sensors used to acquire operating parameters 119, preprocessing of the acquired data may be necessary. This can include, in particular, checking metadata.

[0274] The labeling of the measured values ​​121 can be implemented as event labeling, wherein each measured value 121 is assigned a corresponding time point. Alternatively or additionally, state labeling can be implemented, wherein each measured value 121 is assigned a corresponding time range.

[0275] Furthermore, measured values ​​121 of operating parameters 119 and / or sensor data of external sensors 171 may be subjected to an integrity check, in which the respective data are checked for error-freeness.

[0276] Figure 12 A flow chart of a method 400 for training artificial intelligence 149 is shown.

[0277] To train artificial intelligence 149 according to method 400 of the present invention for training artificial intelligence 149 to ascertain operating states A, B, C of handheld power tool 100 , a training data set generated according to method 300 of one of the above-described embodiments is first provided in method step 401 .

[0278] In method step 403 , artificial intelligence 149 is trained based on the training data set for determining operating states A, B, C of handheld power tool 100 and / or predicting event times 125 , 126 . The training can be performed using training methods known from the prior art.

[0279] Experiments show that about Figure 9For the training of the model architecture of the described artificial intelligence, a batch size of 16 is advantageous. The Adam algorithm can lead to satisfactory results as an optimization algorithm. In particular, for the application scenario processed here, a relatively small learning rate of 0.0001 is sufficient. Therefore, a relatively high number of training rounds is required. Usually, there is no need to predefine an upper limit for the number of rounds, and the training is only ended (so-called early stopping) when it is determined with the help of a validation dataset that no improvement of the model can be achieved for a longer time. In a training setting that is advantageous for the application scenario processed here, this "longer time" is defined as 300 rounds.

[0280] However, during development, it was consistently observed that for the chosen learning rate, training was interrupted relatively early (after fewer than 2000 epochs), and the corresponding models performed significantly worse than models trained for more epochs. Therefore, early stopping was somewhat extended compared to the minimum number of epochs that could be defined. Therefore, a minimum of 1500 or 2000 epochs was typically set. The mean squared error (MSE) was used as the loss function.

[0281] It is often advantageous to begin training with a portion of the time series data before using the entire time series. In particular, you can use a region around a critical point (e.g., to be detected) to train the model's approximate behavior. Then, in a further step (e.g., after 10%-50% of the total training rounds), the entire time series is used to begin training the model's behavior in regions further in time.

[0282] This procedure is particularly advantageous because it reduces the probability of non-optimal training results. For typical applications of screwdrivers and drills, a range of 0.1 s to 10 s before and after the event time can be used, with 0.5 s proving to be particularly suitable for the application being discussed here.

[0283] To evaluate the model's quality after training, an automated theoretical evaluation is first performed on a test dataset. The test dataset undergoes the same preprocessing steps as the inference process on the microcontroller. As described above, these preprocessing steps include conversion to SI units, calculation of revolutions per second based on the Hall sensor signals, and normalization based on the normalization parameters determined on the training dataset.

[0284] During the evaluation process, these data, corresponding to actual application data, are input into the model to analyze its behavior. Because the activation function of the output neuron is defined as the hyperbolic tangent (tanh), the output value is within the range of [-1, 1]. The time point in the predicted time series when the threshold is first exceeded is considered the trigger time point. This threshold can be 0, but can also be close to -1 or 1 to adjust the trigger sensitivity.

[0285] The threshold value can be set before delivery (advantage: consistent behavior) or can be adapted during use, for example by adjusting a controller. This has the advantage of being able to select a more appropriate behavior for different situations. Typical fixed threshold values ​​are between 0 and -1, enabling earlier triggering and compensating for delays caused by process and machine inertia. A value of -0.85 has proven to be particularly suitable for achieving increased sensitivity while maintaining an acceptable probability of false triggering.

[0286] For about Figure 3 and Figure 4 The described embodiment primarily considers two application-related criteria, in which each material-screw combination is considered separately. This involves:

[0287] - The mean absolute deviation of the triggering time from the expected (= labeled) triggering time, measured as the motor speed. Furthermore, the relative deviation from the expected triggering time is also considered in the form of a frequency distribution, because this shows whether the model tends to trigger too early or too late, or whether the modal value is zero (as expected).

[0288] -The number or share of tightenings that the model has not triggered at any point in time.

[0289] The MSE of the loss plays a minor role because for real applications, accurate prediction of the entire time series is less important than the trigger time point. However, it has been shown that the MSE is generally correlated with application-relevant metrics.

[0290] For a handheld power tool 100 designed as a screwdriver, a plurality of combinations of the screw type of the screw 169 and the material type of the corresponding workpiece 137 can be considered for the training of the artificial intelligence 149 of the state determination module 107 .

[0291] Different parameters that influence the quality of the training of the artificial intelligence 149 and its subsequent performance in determining the operating state can be taken into account. These parameters include, in particular, the material (e.g., wood species), the length / diameter / pitch / thread type of the screw, the speed during tightening, and user influences (e.g., tilt and pressure consumed). Since it is practically impossible to record data for every possible combination, let alone a sufficient amount of data, it is advantageous to use a combination of different common screw types with material choices of different hardnesses. Information about the screw-material combinations used can be stored in the form of metadata and used in training. Generally, it should be noted that a sufficient number of recordings are performed for each combination (at least 25 times, preferably more). It is also advantageous that no combination is tightened significantly more times than another combination, so that a balanced data base is ultimately obtained for the AI ​​training.

[0292] Examples of possible screw-wood combinations for the application case “screwing in wood” are provided below.

[0293] Combinations of different application-related wood species (e.g., particleboard, pine, beech, spruce) with screw types are used. These typically have diameters of 2 mm to 6 mm, lengths of 20 mm to 100 mm, a head shape of Hexalobular or Pozidriv, and either a self-tapping thread or not. A record need not be kept for every grade and combination of material properties, but a sufficiently broad coverage should be provided. Combinations of 100 or more have proven suitable. In particular, undesirable associations in the record that do not correspond to the actual application (e.g., between wood types and screw types) should be avoided.

[0294] Alternatively, the state determination module 107 may include artificial intelligence 149 of different configurations or multiple artificial intelligences 149. For example, the state determination module 107 may include artificial neural networks of different configurations, such as a temporal convolutional network or a convolutional network. Alternatively, the state determination module 107 may include a model such as a decision tree or an ensemble model.

[0295] The state determination module 107 can be designed, in particular, as a classifier and configured to classify the individual measured values ​​121 of the operating parameter 119 in the time series 123, assign the individual measured values ​​121 of the operating parameter 119 to the corresponding operating states A, B, and C, and thereby determine or predict event times 125 and 126. The measured values ​​121 of the operating parameter 119 can be processed by the state determination module 107 during operation of the handheld power tool 100 according to a sliding window and assigned to the corresponding operating states A, B, and C according to the classification. Thus, the time series 123 of the recorded measured values ​​121 is analyzed by the state determination module 107 according to the sliding window, and thereby the operating states A, B, and C are determined or the event times 125 and 126 are predicted.

[0296] Figure 13 A schematic diagram of a computer program product 500 is shown, which includes instructions that, when the program is executed by a data processing unit, cause the data processing unit to execute a method 200 for controlling a handheld machine tool 100 according to one of the above-mentioned embodiments and / or a method 300 for generating a training data set according to one of the above-mentioned embodiments and / or a method 400 for training an artificial intelligence 149 and / or the artificial intelligence 149.

[0297] In the embodiment shown, computer program product 500 is stored on a storage medium 501. Storage medium 501 may be any storage medium known from the prior art.

Claims

1. A method (200) for controlling a handheld power tool (100), comprising: Receiving (201) sensor data of at least one operating parameter (119) of the handheld power tool (100); receiving (203) an input value based on a user input of a user of the handheld power tool (100), the input value being for a control parameter of the handheld power tool (100); Determining (205) a first target value of the control parameter based on the sensor data and the input value; executing (207) a state determination module (107) on the sensor data and determining an operating state (A, B, C) of the handheld power tool (100); and The handheld power tool (100) is controlled (209) based on the first target value of the control parameter and the determined operating state (A, B, C).

2. The method (200) according to claim 1, wherein: The control (209) of the handheld power tool (100) includes: determining (211) a second target value for the control parameter based on the determined operating state (A, B, C) of the handheld power tool (100); deriving (213) an output target value of the control parameter based on the first target value and the second target value; and The output target value is output (215) to an actuating device of the handheld power tool (100) for controlling the handheld power tool (100).

3. The method (200) according to claim 2, wherein: The obtaining of the output target value (213) includes: defining (217) the output target value as the minimum or maximum of the first target value and the second target value; and / or The output target value is defined (219) as the product of the first target value and the second target value.

4. The method (200) according to claim 2 or 3, wherein: The obtaining of the output target value (213) includes: If the second target value is less than a predetermined threshold, defining (221) the output target value as the first target value; and If the second target value is greater than or equal to the predetermined threshold, the output target value is defined as a predetermined target value.

5. The method (200) according to any one of the preceding claims 2 to 4, wherein: The obtaining of the output target value (213) includes: If the second target value is less than a predetermined threshold, defining (223) the output target value as the product of the first target value and a first predetermined target value; and If the second target value is greater than or equal to the predetermined threshold, the output target value is defined as the product of the first target value and a second predetermined target value.

6. The method (200) according to any one of the preceding claims, wherein: The determination (207) of the operating state (A, B, C) includes: An event time (125, 126) is predicted (225), wherein at the event time (125, 126) the handheld power tool (100) transitions from one operating state (A, B, C) to another operating state (A, B, C).

7. The method (200) according to any one of the preceding claims, wherein: The control parameters include one or more of the following: motor speed, motor current, and motor power of the motor of the handheld power tool (100).

8. The method (200) according to any one of the preceding claims, wherein: The operating parameters (119) include one or more of the following: motor current, motor position angle, motor rotation speed, and voltage of a voltage source of the handheld power tool (100).

9. The method (200) according to any one of the preceding claims, wherein: The operating states (A, B, C) include one or more of the following: the load range in which the handheld machine tool (100) operates, the vibration intensity in the handheld machine tool (100) and / or on the workpiece (167) being processed and / or in the user of the handheld machine tool (100), the temperature in the handheld machine tool (100) and / or on the workpiece (167), the operating mode in which the handheld machine tool (100) operates, the working progress of the handheld machine tool (100), the material of the workpiece (167), and the existence of a positive connection between the handheld machine tool (100) and the workpiece (167).

10. The method (200) according to any one of the preceding claims, wherein The state determination module (107) includes a trained artificial intelligence (149) which is trained to determine the operating state (A, B, C) of the handheld power tool (100) based on sensor data of the operating parameters and / or to predict the event time (125, 126).

11. A computing unit (105) configured to carry out the method (200) for controlling a handheld power tool (100) according to any one of the preceding claims 1 to 10.

12. A computer program product (500) comprising instructions which, when the program is executed by a data processing unit, cause the data processing unit to carry out the method (200) for controlling a handheld power tool (100) according to any one of the preceding claims 1 to 10.

13. A handheld power tool (100) comprising a computing unit (107) according to claim 11 and at least one sensor for determining sensor data of at least one operating parameter of the handheld power tool (100).

Images