Method for operating measuring device, method for training artificial intelligence, computing unit and computer program product

Abstract

The invention relates to a computer-implemented method for operating a measuring device, in particular a wall diagnostic device, comprising: receiving radar data of a radar sensor unit of the measuring device, the radar data mapping a wall to be diagnosed; carrying out wall diagnosis by carrying out an analysis of the radar data through a diagnosis module of the measuring device, and providing a diagnosis result; providing the diagnosis result and the uncertainty value to a display unit of the measuring equipment through the diagnosis module; and displaying the diagnosis result and the uncertainty value in the display unit. The invention also relates to a method, a computing unit and a computer program product for training artificial intelligence of a measuring device for wall diagnostics.

Description

Technical Field

[0001] The invention relates to a method for operating a measuring device, in particular a wall diagnostic device. Background Art

[0002] Diagnostic devices for diagnosing walls and for detecting objects formed in walls are known from the prior art. Summary of the Invention

[0003] One object of the present invention is to provide an improved method for operating a measuring device, in particular a wall diagnostic device. Another object is to provide an improved method for training an artificial intelligence of a measuring device.

[0004] This object is achieved by a computer-implemented method for operating a measuring device, in particular a wall diagnosis device, a method for training an artificial intelligence of a measuring device for wall diagnosis, a computing unit, and a computer program product. Advantageous embodiments are described below.

[0005] According to one aspect, a method for operating a measuring device, in particular a wall diagnostic device, is provided, wherein the method comprises:

[0006] receiving radar data from a radar sensor unit of a measuring device, wherein the radar data maps a wall to be diagnosed;

[0007] The diagnostic module of the measuring device performs wall diagnosis by analyzing the radar data and provides a diagnostic result, wherein the wall diagnosis includes:

[0008] performing object recognition of an object arranged in the wall by the diagnosis module, wherein the object recognition includes object detection and object classification, wherein the diagnosis result includes at least an object position and / or an object type of the object in the wall;

[0009] Obtaining an uncertainty value of the diagnosis result by the diagnosis module, wherein the uncertainty value of the diagnosis result describes a probability value of consistency between the diagnosis result and an actual state of the wall to be diagnosed, and includes at least a probability value of the object position of the object and / or a probability value of the object type;

[0010] providing the diagnosis result and the uncertainty value to a display unit of a measuring device through the diagnosis module;

[0011] The diagnosis result and the uncertainty value are displayed on the display unit.

[0012] This achieves the technical advantage of providing an improved method for operating a measuring device, particularly a wall diagnostic device. To this end, radar data from a radar sensor unit of the measuring device that maps to a wall to be diagnosed is first received. Based on the received radar data, a diagnostic module performs a wall diagnostic and provides a diagnostic result.

[0013] Wall diagnosis involves object recognition of objects located in the wall using a diagnostic module. Object recognition includes object detection, which determines the object's position, and object classification, which determines the object's type. The corresponding diagnostic results provided include at least the object's position and type of the object located in the wall.

[0014] During the wall diagnosis, an uncertainty value is also determined for the diagnosis result. The uncertainty value describes the probability of a specified object position and the probability of a specified object type. Furthermore, the diagnosis result, including the uncertainty value, is provided by the diagnosis module to a display unit of the measuring device and displayed to a user on the display unit.

[0015] This method thus enables object recognition of objects located in the wall to be diagnosed based solely on radar data from the radar sensor unit of the measuring device. Furthermore, uncertainty values ​​for object recognition, namely, uncertainty values ​​for the object's position and type, are generated and displayed on a display unit. These uncertainty values ​​can be used to indicate to the user the quality of the object recognition performed. A high probability value for the object's position or type can indicate high quality of object recognition, while a low probability value indicates low quality of object recognition. This can further improve the wall diagnosis performed by the measuring device.

[0016] According to one embodiment, the wall diagnosis further includes:

[0017] Implementing wall type classification and determining the wall type of the wall by the diagnostic module; and / or

[0018] performing object depth determination by the diagnostic module and determining the object depth of the object in the wall, wherein the object depth is defined as the distance from the object to the surface of the wall; and / or

[0019] An object extension determination (Best Immunity) is implemented by means of a diagnostic module and the object extension of the object along a predefined direction is determined, wherein the uncertainty value also includes at least a probability value of the wall type and / or a probability value of the object depth of the object and / or a probability value of the object extension.

[0020] This achieves the technical advantage of further improving wall diagnosis. To this end, during the wall diagnosis process, a diagnostic module performs a wall type classification based on radar data from a radar sensor unit and determines the wall type of the wall to be diagnosed. Alternatively or additionally, an object depth determination is performed and the object depth is determined for a diagnostic module positioned in the wall. The object depth describes the distance of the object from the surface of the wall. Alternatively or additionally, an object extension is determined based on the radar data and the object extension along a predefined direction is determined.

[0021] Furthermore, the uncertainty value includes a probability value for the wall type and / or a probability value for the object depth and / or a probability value for the object extension. By determining the wall type and / or the object depth and / or the object type, the information provided by the wall diagnosis about the wall to be inspected can be further increased. By taking into account the probability values ​​or uncertainty values, the quality of the wall diagnosis can be determined and displayed on the display unit.

[0022] According to one embodiment, in the object recognition, multiple possible object positions and / or multiple possible object types of the object are determined as independent diagnostic results, and / or wherein, in the wall type classification, multiple possible wall types of the wall are determined as independent diagnostic results, and / or wherein, in the object depth determination, multiple possible object depths are determined as independent diagnostic results, and / or wherein, in the object extension scale determination, multiple possible object extension scales of the object are determined as independent diagnostic results, wherein each of the multiple diagnostic results is provided with a probability value, wherein the multiple diagnostic results and the multiple probability values ​​are displayed in the display unit.

[0023] This achieves the technical advantage of providing further improvements to wall diagnosis. To this end, multiple alternatives for the determined diagnostic results are provided in the wall diagnosis, including corresponding uncertainty values. Thus, multiple object positions, multiple object types, multiple object depths, multiple object extensions, and / or multiple wall types can be provided as wall diagnosis results. For each of the multiple object positions, multiple object types, multiple object depths, multiple object extensions, and / or multiple wall types, a corresponding uncertainty value is provided and displayed in the form of a probability value.

[0024] Here, different probability values ​​describe the reliability or uncertainty of the corresponding object position, corresponding object type, corresponding object depth, corresponding object extension, and / or corresponding wall type relative to the actual object position, actual object type, actual object depth, actual object extension, and / or actual wall type. Thus, additional information for wall diagnosis can be provided to the user.

[0025] According to one embodiment, the method further comprises:

[0026] A selection function is provided, wherein, by implementing the selection function, at least one of the displayed diagnosis results can be selected by a user of the measuring device.

[0027] This achieves the technical advantage of further improving the wall diagnosis. To this end, the user is provided with a selection function. This allows the user to select one or more of the provided diagnostic results as the correct one. Conversely, the other diagnostic results are no longer considered in the further wall diagnosis. This allows the user to select, for example, one of the specified wall types. For example, the user may already know the wall type of the wall to be inspected. By selecting the corresponding wall type, it can be appropriately taken into account in the wall diagnosis, i.e., in object detection. This further improves the quality of the wall diagnosis.

[0028] According to one embodiment, the method further comprises:

[0029] receiving a selection instruction of the selection function, wherein the user selects one of the displayed diagnosis results through the selection instruction;

[0030] Feedback information is provided to an external server unit, wherein the feedback information includes the selection instruction and the diagnosis result selected therein together with the corresponding radar data.

[0031] This achieves the technical advantage of enabling improved wall diagnosis by providing feedback information. To this end, when a user selects a function, the corresponding selection is recorded and provided as feedback information to an external server unit. Based on the provided feedback information, improvements can be made to the diagnostic module, for example, by retraining it. The correspondingly improved, i.e., newly trained, diagnostic module can be installed in an updated form on the measuring device, thereby achieving improved wall diagnosis.

[0032] The feedback information provided to the external server unit includes the radar data used to initially perform the wall diagnosis, the provided diagnostic results, and the diagnostic results selected by the user through a selection function. Based on this information, the external server unit can assess the quality of the initially performed wall diagnosis. Based on this information, the wall diagnosis, or in other words, the diagnostic module, can be improved by retraining the external server unit.

[0033] In the sense of the present application, feedback information includes any information that is based on the user's behavior and can be interpreted as feedback on the quality of the wall diagnosis performed. The feedback information can be direct, possibly text-based feedback from the user. Alternatively or additionally, it can be

[0034] According to one embodiment, the measuring device further comprises at least one inductive sensor and / or eddy current sensor and / or capacitive sensor and / or AC sensor and / or nuclear magnetic resonance (NMR) sensor and / or ultrasonic sensor for providing additional sensor data, wherein the diagnostic module is configured to perform the wall diagnosis taking into account the additional sensor data.

[0035] This provides the technical advantage that additional information, in addition to the information from the radar data of the radar sensor unit, can be incorporated into the wall diagnosis via further sensor data from additional sensors, each of which is configured to detect a different physical measurement variable. This additional information, which is preferably complementary to the information from the radar data of the radar sensor unit, enables further refinement of the wall diagnosis or object identification.

[0036] According to one specific embodiment, the diagnostic module comprises at least one correspondingly trained artificial intelligence, which is configured to perform object recognition and / or wall classification and / or object depth determination based on the radar data and / or the additional sensor data.

[0037] This achieves the technical advantage that by configuring the diagnostic module as a correspondingly trained artificial intelligence (which is trained to perform object recognition and / or wall classification and / or object depth determination and / or object extension determination based on radar data or, if necessary, taking into account information from additional sensors), a reliable and high-performance diagnostic module can be provided. The use of artificial intelligence technology allows for precise wall diagnosis.

[0038] According to one embodiment, the object category of the object type of the object includes: metal / non-metallic objects, cables for low voltage, cables with single-phase AC signals, cables with multi-phase AC signals, wooden beams, metal beams, plastic pipes, water-filled plastic pipes, such as water supply pipes, non-water-filled plastic pipes, such as drainage pipes, and / or wherein the wall type category of the wall type of the wall includes: concrete wall, lightweight structure / dry structure wall, brick wall and / or brick blocks of the wall, floor heating, wall heating.

[0039] This achieves the technical advantage that a wide variety of objects and walls can be recognized and classified.

[0040] According to one aspect, a method for training artificial intelligence of a measurement device for wall diagnosis is provided, the method comprising:

[0041] providing a training data set for training the artificial intelligence, wherein the training data set includes radar data mapping a wall and objects constructed in the wall and feedback information provided according to a method for operating a measurement device;

[0042] The artificial intelligence is trained based on the training data set and taking into account the feedback information to perform object recognition of objects formed in the wall, wherein the object recognition includes at least object detection and object classification.

[0043] This achieves the technical advantage that an improved training of the artificial intelligence of the wall diagnosis device is possible, in particular subsequent training which simultaneously takes into account feedback from users of the wall diagnosis device.

[0044] According to one aspect, a computing unit is provided, which is configured to carry out a method for operating a measuring device and / or a method for training an artificial intelligence according to one of the above embodiments.

[0045] According to one aspect, a computer program product is provided, which includes instructions which, when the program is executed by a data processing unit, cause the data processing unit to carry out a method for operating a measuring device and / or a method for training an artificial intelligence according to one embodiment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Embodiments of the present invention will be described with reference to the following drawings. These drawings show:

[0047] Figure 1 A schematic illustration of a measuring device according to one embodiment;

[0048] Figure 2 Another schematic illustration of a measuring device according to another embodiment;

[0049] Figure 3 Another schematic illustration of a measuring device according to another embodiment;

[0050] Figure 4 Schematic illustration of a measurement of a measuring device according to one embodiment,

[0051] Figure 5 Another schematic illustration of a measuring device according to another embodiment;

[0052] Figure 6Schematic illustration of a system for operating a measuring device according to one embodiment,

[0053] Figure 7 A flow chart of a method for operating a measuring device according to one embodiment,

[0054] Figure 8 Another flow chart of a method for operating a measuring device according to another embodiment,

[0055] Figure 9 Flowchart of a method for training an artificial intelligence of a measuring device according to one embodiment,

[0056] Figure 10 Schematic illustration of a computer program product. DETAILED DESCRIPTION

[0057] Figure 1 A schematic illustration of a measuring device 100 according to one specific embodiment is shown.

[0058] The present invention relates to a measuring device, and in particular to a wall diagnostic device for inspecting a wall 105 to be processed. Wall diagnostic devices for detecting objects arranged in a wall are known in the prior art. Such devices allow a user to inspect the wall to be processed based on the objects arranged in the wall, so that planned work, such as drilling a hole in the wall, can be carried out based on this information, in a manner that avoids damage to the objects arranged in the wall.

[0059] In the embodiment shown, the measuring device 100 comprises a housing 150 having a handle 152 for a user to hold the measuring device 100 , a display unit 111 for displaying the diagnosis result 109 of the wall diagnosis, and operating elements 154 for switching the measuring device 100 to different operating modes.

[0060] According to the present invention, measuring device 100 comprises at least one radar sensor unit 101. Radar sensor unit 101 can transmit radar signals in the direction of a wall 105 to be inspected and can receive radar signals reflected by wall 105.

[0061] For example, radar sensor unit 101 may be configured as a narrowband radar detector device in the frequency range of 2.4 GHz to 2.4835 GHz or as an ultra-wideband radar detector device in the frequency range of 1.8 GHz to 5.8 GHz.

[0062] Measuring device 100 also includes a diagnostic module 107, which can be executed on computing unit 151 of measuring device 100, for performing wall diagnostics. Diagnostic module 107 is configured to perform corresponding diagnostics of the wall to be inspected based on radar data 103 from radar sensor unit 101. Radar data 103 from radar sensor unit 101 represents wall 105 to be inspected and objects 113 that may be located within wall 105.

[0063] The wall diagnosis performed by diagnostic module 107 includes at least the implementation of object recognition. Object recognition includes object detection and object classification of objects 113 located in wall 105. Object detection includes at least the determination of object position 115. The object position describes the position of the object located in wall 105 relative to the coordinate system defined by measuring device 100. Object classification of detected objects 113 includes at least the determination of object type 117 of detected objects 113.

[0064] The diagnostic results of the wall diagnosis determined in this manner, i.e., at least the determined object position 115 and / or the determined object type 117 of the object 113 arranged in the wall 105, are then displayed to the user of the measuring device 100 on the display unit 111 of the measuring device 100. The display unit 111 can be configured as a corresponding display screen, for example, and the diagnostic results 109 can be displayed visually. In addition, the display of the diagnostic results 109 can be assisted by acoustic and / or tactile signals. For example, the tactile signals can be implemented by corresponding vibration signals.

[0065] In this case, object 113 can be displayed, for example, by a corresponding symbol in a display screen. Object 113 can be displayed in a corresponding object position 115 in the display screen. Object extent 121 can be visualized by the corresponding size of the displayed symbol. The corresponding object type 117 of object 113 can be visualized by a corresponding term, a colored background of the symbol, or a specific shape of the symbol representing object 113.

[0066] Alternatively, the wall diagnosis can additionally include determining a wall type 123 in the form of a wall type classification of the wall 105 to be inspected. Wall type 123 describes the respective type of wall 105 to be inspected. For example, the wall type can be assigned to a respective wall type category, which can include: concrete wall, lightweight / drywall wall, brick wall and / or wall block, floor heating, wall heating, or wall types found in similar buildings.

[0067] According to one embodiment, diagnostic module 107 is further configured to determine object depth 119 of object 113 within wall 105 based on radar data 103. Object depth 119 is defined as the distance of an object embedded in wall 105 from the surface of wall 105. This distance can be defined, for example, relative to the object surface or relative to the object's center point. The distance to the surface of wall 105 describes the shortest distance defined by a direction perpendicular to the surface of wall 105.

[0068] According to one embodiment, diagnostic module 107 is further configured to determine object extent 121 of object 113 in at least one predefined direction based on radar data 103. Object extent 121 of object 113 describes the spatial extent of object 113 in at least one spatial direction, preferably in two spatial directions, and particularly preferably in three spatial directions. Thus, object 113 can be described as a one-dimensional, two-dimensional, or three-dimensional object 113.

[0069] In typical use, measuring device 100 is placed on the wall surface of a wall 105 to be inspected. Radar signals are transmitted toward wall 105 via radar sensor unit 101, and radar signals reflected by wall 105 or objects 113 located behind it are received. Based on radar data 103 from radar sensor unit 101, diagnostic module 107 performs the aforementioned wall diagnosis and determines corresponding diagnostic results 109.

[0070] Diagnosis result 109 may include, for example, object position 115 and / or object type 117 of object 113 arranged in wall 105. Alternatively or additionally, diagnosis result 109 may include wall type 123 of wall 105 and / or object depth 119 and / or object extent 121 of object 113.

[0071] The diagnosis result 109 formed in this way can then be displayed to the user of the measuring device 100 on the display unit 111 of the measuring device 100. The display unit 111 can be designed as a corresponding display screen, for example. The diagnosis result 109 can be displayed in a graphical form or in a textual form on the display unit 111.

[0072] According to one embodiment, measuring device 100 further includes a motion detection unit 141. Motion detection unit 141 can be used to detect the movement of measuring device 100 relative to wall 105. To this end, motion detection unit 141 can include, for example, at least one rolling element. When the rolling element is supported on the wall surface of wall 105, the movement of measuring device 100 relative to wall 105 can be detected when measuring device 100 moves along motion direction 153 due to the rolling of the rolling element. Alternatively, motion detection unit 141 can have other configurations that can detect the relative movement of measuring device 100 relative to wall 105.

[0073] By moving measuring device 100 relative to wall 105, radar data 103 of radar sensor unit 101 can be recorded for a plurality of different positionings of measuring device 100 relative to wall 105. This allows for an inspection of wall 105 over a larger spatial area than that given by the range of action of radar sensor unit 101. This allows for the detection of objects 113 having a larger spatial extent than the range of action of radar sensor unit 101.

[0074] While measuring device 100 is moving along movement device 153, radar data 103 from radar sensor unit 101 can be continuously recorded. Based on these radar data 103, wall diagnostics can be evaluated by diagnostic module 107 while measuring device 100 is moving along movement direction 153. This allows for accelerated wall diagnostics that take into account the positioning of measuring device 100 relative to wall 105.

[0075] Depending on its embodiment, diagnostic module 107 is designed as a correspondingly trained artificial intelligence 125. Artificial intelligence 125 is trained to perform the aforementioned wall diagnosis based on radar data 103 from radar sensor unit 101 and to determine at least object position 115 and object type 117 of object 113 located in wall 105. Object classification or determination of object type 117 includes assigning detected object 113 to a predefined object class.

[0076] Object categories here can include: metallic / non-metallic objects, cables for low voltage, cables with single-phase AC signals, cables with multi-phase AC signals, wooden beams, metal beams, plastic pipes, water-filled plastic pipes, such as water supply pipes, non-water-filled plastic pipes, such as drainage pipes, or other common elements built into building walls.

[0077] Furthermore, artificial intelligence 125 may be trained to determine wall type 123 of wall 105 to be inspected based at least on radar data 103 of radar sensor unit 101. Possible wall types 123 may include: concrete walls, lightweight / drywall walls, brickwork and / or brickwork walls, floor heating, wall heating, or other common wall types constructed in buildings.

[0078] According to one embodiment, in addition to radar sensor unit 101, measuring device 100 may also include other additional sensors, by means of which additional physical variables can be detected. For example, measuring device 100 may include an inductive sensor, an eddy current sensor, a capacitive sensor, an AC sensor, a nuclear magnetic resonance sensor, an ultrasonic sensor, or other sensors commonly used in wall diagnostic equipment.

[0079] Diagnostic module 107, in particular a correspondingly trained artificial intelligence 125, can be configured to perform the aforementioned wall diagnosis based on radar data 103 from radar sensor unit 101 and taking into account additional sensor information from other sensors. The additional information from these additional sensors can be used, in particular, to identify objects 113 located in wall 105. The additional sensor information can optionally enable improved detection of objects 113 and, if necessary, improved classification of objects 113.

[0080] In particular, the material of the object 113, such as a metallic material or a non-metallic material, may be refined and classified, for example, by using additional sensor information.

[0081] Figure 2 A further schematic illustration of a measuring device 100 according to a further specific embodiment is shown.

[0082] In the illustrated embodiment, in addition to diagnostic module 107, measuring device 100 also includes a preprocessing module 127. To perform wall diagnosis, measuring device 100 first receives radar data 103 from radar sensor unit 101. Received radar data 103 is preprocessed by preprocessing module 127. Preprocessing by preprocessing module 127 can, for example, convert the radar data into a corresponding data structure required for wall diagnosis by diagnostic module 107.

[0083] As described above, the diagnostic module 107 generates the aforementioned diagnostic result 109 during the wall diagnostic process. Diagnostic result 109 may include, for example, an object position 115 of an object 113 located within the wall 105 to be inspected, an object type 117, an object depth 119, an object extension 121, and / or a wall type 123 of the wall 105 to be inspected. The correspondingly generated diagnostic result 109 may then be displayed on the display unit 111 of the measuring device 100.

[0084] According to one embodiment, in addition to radar data 103 from radar sensor unit 101, additional sensor information from the aforementioned additional sensors may also be considered in the wall diagnosis of diagnosis module 107. Preprocessing module 127 may accordingly perform corresponding preprocessing of the additional sensor information.

[0085] In the illustrated embodiment, diagnosis module 107 includes a wall type classification module 129 and an object recognition module 131. Preprocessing module 127 includes a first preprocessing module 135 and a second preprocessing module 137. First preprocessing module 135 includes S-matrix reduction 155. Second preprocessing module 137 includes background correction 157, inverse fast Fourier transform 159, and focusing and migration 161. When radar data 103 is preprocessed by preprocessing module 127, radar data 103 is first preprocessed by first preprocessing module 135 and S-matrix reduction 155 included therein.

[0086] Here, first preprocessing module 135 generates input data 133 based on radar data 103. Input data 133 serves as input data for wall type classification module 129. Wall type classification module 129 classifies the wall 105 to be inspected into a certain type based on input data 133 and generates wall type information 139. Wall type information 139 contains the wall type 123 of the wall 105 to be inspected determined in the wall type classification.

[0087] Second preprocessing module 137 then performs preprocessing based on radar data 103 and wall type information 139. Background correction 157 is performed on radar data 103, taking into account wall type 123 ascertained in wall type information 139. Depending on wall type 123 of wall 105 to be inspected, different effects on radar data 103 may occur.

[0088] These effects, which are primarily due to the respective wall type 123 and may affect object recognition, can be corrected by background correction 157. After background correction, further preprocessing can be performed by performing an inverse fast Fourier transform 159 or focusing and shifting 161, and new input data 133 can be generated for the object recognition module 131. Based on the input data 133 provided by the second preprocessing module 137, the object recognition module 133 recognizes an object 113 located in the wall 105 to be inspected and determines at least the object position 115 and the object type 117 of the respective object 113. Additionally, the object recognition module 131 can determine the object depth 119 and the object extent 121.

[0089] According to one specific embodiment, the diagnostic module is further configured to determine an object depth of an object within the wall based on the radar data, wherein the object depth is defined by the distance of the object formed in the wall from the surface of the wall.

[0090] Preprocessing is optional. Depending on the algorithm used for diagnostic module 107, completely unprocessed radar echoes of different frequencies can be used as radar data 103 and as input data for diagnostic module 107. Alternatively, radar data 103 processed in multiple steps can be used. Preprocessing steps include, for example, converting the signal from the frequency domain to the time domain or range space, background removal, denoising, and signal normalization. In the case of radar data 103 in complex form, only the absolute values ​​can be processed. Alternatively or additionally, phase information can be taken into account.

[0091] Figure 3 A further schematic illustration of a measuring device 100 according to a further specific embodiment is shown.

[0092] In the embodiment shown, the diagnostic module 107 includes a plurality of processing paths 102 extending in parallel. In each processing path 102 there is a pre-processing module 127, a diagnostic module 107 (eg, including a Figure 2 The wall type classification module 129 and / or object recognition module 131) and the post-processing module 163 of the embodiment.

[0093] exist Figure 3In the example, radar data 103 is primarily presented as input data for wall diagnosis. However, in addition to the radar data shown, additional information from additional sensors can also be used as input data for wall diagnosis. Different information from different sensor types can be processed in different parallel processing paths 102, and corresponding wall diagnosis can be performed separately based on the different sensor information. After the wall diagnosis is completed, the summary module can combine the analysis results of each component into a wall diagnosis result 109.

[0094] Alternatively or additionally, different partial aspects of the wall diagnosis can also be performed by different processing paths 102 based on the same sensor information.

[0095] For example, the various processing paths 102 can process different radar data 103 that were recorded for different positions of measuring device 100 relative to wall 105 during the movement of measuring device 100 relative to wall 105. Radar data 103 that thus map different areas of wall 105 and were recorded temporally sequentially during the movement of measuring device 100 relative to wall 105 can then be processed in the various processing paths 102 by the modules shown.

[0096] The different processing paths here carry out independent wall diagnoses, which include at least determining an object position 115 and / or an object type 117 of an object 113 arranged in the wall 105 .

[0097] The summarization module 165 can summarize the partial results of independent wall diagnosis of different areas of the wall 105 provided in each processing path 102 into a coherent Diagnosis result 109. In this case, a coherent diagnosis result describes a wall diagnosis for a coherent spatial region scanned during the movement of measuring device 100 relative to wall 105 and mapped by the correspondingly recorded radar data 103. Therefore, parallel processing of radar data 103 or additional sensor information 104 of additional sensor elements in different processing paths 102 enables accelerated wall diagnosis.

[0098] Alternatively, different functions of the wall diagnosis can also be implemented in different processing paths 102. Thus, for example, wall type classification can be implemented in one processing path 102, and the wall type 123 of the wall 105 to be inspected can be determined. In another processing path 102, object recognition of objects 113 arranged in the wall can be implemented. In this case, object detection with determination of the object position 115 and object classification with determination of the object type 113 can be implemented in one processing path 102.

[0099] Alternatively, object detection and object classification can also be performed in two separate processing paths 102. In the other processing path 102, object depth determination (i.e., determination of object depth 119) and / or determination of object extension 121 can be performed separately. In the aggregation module 165, the different partial results of the wall diagnosis can be aggregated into the corresponding diagnostic result 109.

[0100] The diagnostic module 107 can be divided into different artificial intelligences 125, as already described. Figure 2 As presented in the embodiment of FIG. , the diagnosis module 107 may include, for example, a wall type classification module 129 and an object recognition module 131. The object recognition module can be further divided into an object detection module and an object classification module. The diagnosis module 107 may also include an object depth determination module and an object extension scale module, which are configured to determine an object depth 119 and an object extension scale 121, respectively.

[0101] The corresponding modules can each be designed as an independent artificial intelligence 125, such as a neural network. Alternatively, the different modules can form parts of an overall artificial neural network, which are connected to form the overall neural network according to structures known from the prior art.

[0102] Figure 4 A schematic illustration of a measurement of a measuring device 100 according to one specific embodiment is shown.

[0103] For preprocessing, radar data 103 or additional sensor information 104 from other sensors can be normalized, in particular, to ensure numerical stability in the subsequent steps performed by diagnostic module 107 during wall diagnosis. For example, amplitude and / or offset compensation can be performed for this purpose. Furthermore, radar data 103 can be filtered to reduce interfering elements, and the corresponding sensor data can be downsampled to reduce the data rate. Furthermore, radar data 103 or additional sensor information 104 can be converted to the desired frequency range or time range. Methods known from the prior art can be used for this purpose.

[0104] Furthermore, recorded radar data 103 or additional sensor information 104 can be divided into time windows or spatial windows 167. Time windows 167 can be generated by recording radar data 103 or additional sensor information or preprocessed radar data 103 at fixed time intervals. Conversely, spatial windows 167 can be generated by assigning radar data 103 or additional sensor information 104 to the position of measuring device 100 relative to wall 105 along direction of movement 153.

[0105] Figure 4Graph a) shows such a data matrix resulting from the above steps. The data matrix in window 167 shown in graph a) shows a plurality of sensor data, which may include, for example, radar data 103 or additional sensor information 104 from other sensors, plotted along a frequency channel axis 171 and along a space / time axis 169.

[0106] The width of the time window 167 can be selected in such a way that different sampling rates of the sensors can be compensated and a new window 167 can be provided frequently enough so that the diagnosis result 109 of the wall diagnosis can be displayed on the display unit 111 without a significant time delay during the measurement or shortly after the end of the measurement of the measuring device 100.

[0107] For this purpose, a data recording rate of 2 to 20 windows per second of sensor data may be advantageous. For the spatial windows, the spatial sampling rate may be selected such that the desired position accuracy is achieved. A sampling rate of 1 mm to 1 cm may be advantageous. This means that for every 1 mm to 1 cm of movement of measuring device 100 along direction of motion 153 , corresponding sensor data is recorded.

[0108] The width of spatial window 167 can be selected such that a window contains coherent information about an object 113. A width of 1 to 20 centimeters for the respective spatial window 167 can be advantageous. This results in 4 to 100 measured values ​​per window 167. This enables efficient further algorithmic processing of the correspondingly recorded radar data 103 or additional sensor information by diagnostic module 107.

[0109] Here, once one or more sampling points are available, another time window 167 or spatial window 167 may be provided.

[0110] The diagnosis module 107 can be configured such that as input data, for example also as Figure 3 The input data of each processing path 102 in the embodiment of the present invention is recorded in a matrix corresponding to the window size of the corresponding spatial or temporal window 167. Figure 2 In this embodiment, the corresponding input data may include respectively preprocessed sensor data, namely radar data 103 and additional sensor information 104 of additional sensors.

[0111] As described above, wall diagnosis can be performed by diagnostic module 107 based on a correspondingly trained artificial intelligence. Alternatively, different processing paths can also be calculated using rule-based algorithms. It is also possible to combine artificial intelligence and rule-based algorithms within a single processing path 102, either in parallel or in series.

[0112] The diagnostic results 109 of the wall diagnosis can be clearly expressed as numerical values, vectors, or matrices. Furthermore, for object detection, the probability of detection can be given, or for wall type classification or object classification, the probability of the specified object class or wall type class can be given. The same applies to position and / or depth determination, for which corresponding probability values ​​can also be given.

[0113] If, in addition to radar data 103 , additional sensor information from other sensor types is processed in a processing path 102 , this sensor information can be combined either within artificial intelligence 125 or by rule-based combination.

[0114] exist Figure 3 In the embodiment of the present invention, in the post-processing of each processing path 102, multiple algorithm results based on multiple windows 167 can be aggregated by an aggregation module 165. This aggregation can be achieved in particular by majority voting, summing or multiplication of consecutive probability values.

[0115] Furthermore, by clustering a plurality of results, for example, clustering a plurality of objects detected close to one another, it is possible to identify which objects are the same object, so that these objects are not mistakenly identified multiple times.

[0116] It is also possible to apply a weighting function 177 in a multiplicative manner when aggregating the results from a plurality of windows 167. Advantageously, the diagnostic partial results 175 corresponding to the respective data points in space can be weighted with reference to the positioning of the diagnostic partial results 175 relative to the center point of the respective window 167. This is illustrated by way of example in diagram b), in which the individual diagnostic partial results 175 are weighted with reference to the center point of the illustrated window 167 according to the illustrated weighting function 177.

[0117] According to one embodiment, the result of one processing path 102 can influence the expansion of another processing path 102 after post-processing 163. In this case, weighting parameters can be adjusted, which can be related to the corresponding result from the processing path 102 for each window.

[0118] For example, the result of the object classification, in which object types 117 of objects arranged in wall 105 are defined, can be used to increase the weight of the wall type classification in which wall type 123 of the corresponding wall 105 is determined at locations where there are no objects 113 in post-processing, because the corresponding radar data 103 at these locations are less affected by reflections from objects 113.

[0119] Figure 5 A further schematic illustration of a measuring device 100 according to a further specific embodiment is shown.

[0120] Figure 5 Graphs a) and b) of FIG. 1 show two different alternatives for the joint data processing of radar data 103 and additional sensor information 104 by diagnostic module 107 .

[0121] Diagram b) shows the joint processing of radar data 103 and additional sensor information 104 from an additional sensor by diagnostic module 107. To this end, radar data 103 and additional sensor information 104 serve as input data for diagnostic module 107, which is designed as an artificial intelligence (particularly, as an artificial neural network). Diagnostic module 107 includes a plurality of convolutional layers 108 and a plurality of dense layers 106. Radar data 103 and additional sensor information 104 are processed together as input data via convolutional layers 108 and dense layers 106. The diagnostic result 109, already mentioned, is generated as output data from diagnostic module 107 based on this data.

[0122] In contrast, in diagram b), radar data 103 and additional sensor information 104 serve as independent input data for diagnostic module 107. Diagnostic module 107 is divided into a plurality of processing paths 102. Each processing path 102 includes a plurality of convolutional layers 108 and at least one dense layer 106. In different processing paths 102, diagnostic module 107 generates a wall diagnosis based on radar data 103 or additional sensor information 104 separately from one another.

[0123] In an additional connection layer 148, the partial results of the partial diagnosis of the different processing paths 102 are combined and fed to the last dense layer 106. The output data of the diagnosis module 107 correspond to the diagnosis result 109 described above.

[0124] The correspondingly designed diagnostic module 107 is configured to carry out the above-described wall diagnosis with the above-described features based on the radar data 103 and the additional sensor information 104 .

[0125] In the embodiment shown, the diagnosis module 107 is designed as an artificial neural network, in particular as a convolutional neural network. A corresponding network architecture with convolutional layers 108, dense layers 106 and connection layers 148 is known from the prior art.

[0126] Figure 6 A schematic illustration of a system 600 for operating a measuring device 100 according to one specific embodiment is shown.

[0127] In the embodiment shown, the system for operating the measuring device 100 comprises an external server unit 114 in addition to the measuring device 100 .

[0128] According to the present invention, measuring device 100 first receives radar data 103 from radar sensor unit 101. Radar data 103 represent a wall 105 to be inspected, including an object 113 arranged therein.

[0129] Based on this, a wall diagnosis is performed by a diagnostic module 107 based on the radar data 103 and a corresponding diagnostic result 109 is generated. The wall diagnosis includes at least object recognition, including object detection with determination of the object position 115 and object classification with determination of the object type 117 of an object 113 located in the wall 105.

[0130] According to the present invention, wall diagnosis further includes determining an uncertainty value 110 for diagnosis result 109. Here, the uncertainty value describes the probability value of the consistency between the determined diagnosis result 109 and the actual state of the wall 105 to be diagnosed. According to the present invention, uncertainty value 110 includes at least a probability value for the determined object position 115 and a probability value for the determined object type 117.

[0131] According to the present invention, the determined diagnosis result 109 (including the uncertainty value 110 ) is provided to a display unit 111 via the diagnosis module 107 and is presented on the display unit 111 .

[0132] In the diagram shown, a diagnosis result 109 and a corresponding uncertainty value 110 are presented graphically in a display unit 111 .

[0133] The user can therefore interpret the wall diagnosis or the provided diagnosis result 109 accordingly, taking into account the uncertainty value 110 .

[0134] According to one embodiment, in addition to object position 115 and object type 117, wall diagnosis may also include object depth 119 or object extension 121 of object 113 disposed in wall 105. Additionally, wall diagnosis may include determining wall type 123 of wall 105. Object depth 119, object extension 121, and wall type 123 may be displayed on display unit 111 as corresponding diagnosis results 109 (including corresponding uncertainty values ​​110).

[0135] According to one embodiment, wall diagnosis includes determining a plurality of possible object positions 115, possible object types 117, possible object depths 119, possible object extensions 121, and / or possible wall types 123. Different possible object positions 115, object types 117, object depths 119, object extensions 121, and / or wall types 123 can be provided with corresponding uncertainty values ​​110 and displayed on a display unit 111.

[0136] According to one embodiment, the user is further provided with a selection function 116. The user can select different diagnostic results from the displayed diagnostic results 109 by operating the selection function 116. For example, if the user knows the actual wall type 123, the user can select the corresponding recommended wall type 123 through the selection function 116.

[0137] The corresponding selected wall type 113 is then taken into account in the remaining wall diagnosis, while the unselected recommended wall type 123 remains out of consideration. The same applies to the corresponding other displayed diagnosis results 109 .

[0138] According to one specific embodiment, feedback information 112 is provided to an external server unit 114 based on a received selection command for selecting function 116, wherein the user correspondingly selects displayed diagnosis result 109 in response to the selection command. Feedback information 112 includes the selection command and the selected diagnosis result 109 therein, together with the corresponding radar data 103.

[0139] Here, the feedback information 112 may include, for example, information about which diagnosis results 109 the user selected as correctly presenting the state of the wall 105. In addition, the feedback information 112 may include information about which diagnosis results 109 the user did not select as correctly presenting.

[0140] Based on the feedback information 112 , the external server unit 114 may improve the diagnosis module 107 in terms of the quality of the wall diagnosis by training the algorithm of the diagnosis module 107 .

[0141] According to one embodiment, feedback information 112 is provided by a plurality of different measuring devices 100. For example, a measuring device 100 can be used by a plurality of users during normal operation. During operation of the measuring device, the corresponding feedback information 109 is first stored, for example, in a corresponding storage unit of the measuring device 100 and provided to an external server unit 114.

[0142] According to one embodiment, external server unit 114 is further configured to generate a training data set 143 while taking into account the feedback information. Based on training data set 143, diagnostic module 107 can be retrained, taking into account feedback information 112 during the retraining. The corresponding newly trained diagnostic module 107 can then be installed in a new measuring device 100 and / or in an existing measuring device 100 as an update to an existing diagnostic module 107. By taking into account feedback information during the training of diagnostic module 107, the performance of diagnostic module 107 can be improved.

[0143] Figure 7 A flow chart of a method 200 for operating a measuring device 100 according to one specific embodiment is shown.

[0144] To operate measuring device 100 , in a first method step 201 , radar data 103 of radar sensor unit 101 of measuring device 100 are initially received.

[0145] In a further method step 203 , a wall diagnosis is carried out by diagnostic module 107 of measuring device 100 based on radar data 103 .

[0146] To this end, in method step 205 , object 113 arranged in wall 105 is detected by diagnostic module 107 . Object detection includes at least object detection with determination of object position 115 and object classification with determination of object type 117 of object 113 .

[0147] In a further method step 207, a corresponding uncertainty value 110 is determined for the generated diagnosis result 109. The uncertainty value describes the probability of the agreement between the diagnosis result 109 and the actual state of the wall 105 to be diagnosed.

[0148] In a further method step 209 , the diagnosis result 109 and the uncertainty value 110 are provided to the display unit 111 of the measuring device 100 . In a further method step 211 , the diagnosis result 109 and the uncertainty value 110 are displayed on the display unit 111 .

[0149] Figure 8 A further flow chart of a method 200 for operating a measuring device 100 according to another specific embodiment is shown.

[0150] Figure 8 The implementation in Figure 7 and includes all method steps described there.

[0151] In the embodiment shown, the wall diagnosis also includes carrying out a wall type classification and determining the wall type 123 in method step 213 . Furthermore, in method step 215 , an object depth determination is carried out and the object depth 119 of the object 113 is determined.

[0152] Furthermore, in method step 217 , an object extension determination is carried out and object extension 121 of object 113 is determined.

[0153] Furthermore, in method step 219 , a selection function 116 is provided.

[0154] In method step 221 , a selection command of the selection function 216 is received, wherein one of the displayed diagnosis results is selected by the selection command.

[0155] In a further method step 223 , feedback information 112 is provided to external server unit 114 . Feedback information 112 comprises a selection instruction and the selected diagnosis result 109 (including the corresponding radar data 103 ) therein.

[0156] According to one specific embodiment, the wall diagnosis can also be performed taking into account additional sensor information 104 .

[0157] Figure 9 A flow chart of a method 400 for training artificial intelligence 125 of measuring device 100 according to one specific embodiment is shown.

[0158] To train artificial intelligence 125 of measuring device 100 for wall diagnosis, a training data set 143 is first provided in method step 401 for training artificial intelligence 125. Training data set 143 includes radar data 103 that maps wall 105 to be diagnosed (including objects 113 built into wall 105). Training data set 123 also includes feedback information 112 provided according to method 200 for operating measuring device 100 described above. Feedback information 112 is incorporated into training data set 143 as additional information to the provided radar data 103.

[0159] Radar data 103 may be based on measurements performed solely for creating the training dataset. Alternatively or additionally, radar data 103 may be recorded by multiple users during operation of multiple measuring devices 100 and provided to external server unit 116 for creating training dataset 143 .

[0160] In a further method step 403 , artificial intelligence training is performed, taking into account the feedback information, based on the previously generated, provided training data set 143 . In this case, artificial intelligence 125 is trained to perform object recognition of objects 113 formed in wall 105 .

[0161] Figure 10 A schematic illustration of a computer program product 500 is shown, which includes instructions which, when executed by a data processing unit, cause the data processing unit to carry out the method 200 for operating the measuring device 100 and / or the method 400 for training the artificial intelligence 125 .

[0162] In the embodiment shown, the computer program product 500 is stored on a storage medium 501. In this case, the storage medium 501 may be any storage medium known from the prior art.

Claims

1. A method (200) for operating a measuring device (100), in particular a wall diagnostic device, comprising: Receiving (201) radar data (103) from a radar sensor unit (101) of the measuring device (100), wherein the radar data (103) maps a wall (105) to be diagnosed; The diagnostic module (107) of the measuring device (100) performs (203) wall diagnosis by analyzing the radar data (103) and provides a diagnostic result (109), wherein the wall diagnosis includes: performing (205) object recognition of an object (113) arranged in the wall (105) by the diagnosis module (107), wherein the object recognition comprises object detection and object classification, wherein the diagnosis result (109) comprises at least an object position (115) and / or an object type (117) of the object (113) in the wall (105); Obtaining (207) an uncertainty value (110) of the diagnosis result (109) by the diagnosis module (107), wherein the uncertainty value (110) of the diagnosis result (109) describes a probability value (110) of the consistency between the diagnosis result (109) and the actual state of the wall (105) to be diagnosed, and at least includes a probability value of the object position (115) of the object (113) and / or a probability value of the object type (117); providing (209) the diagnosis result (109) and the uncertainty value (110) to a display unit (111) of the measuring device (100) via the diagnosis module (107); The diagnosis result (109) and the uncertainty value (110) are displayed (211) in the display unit (111).

2. The method (200) according to claim 1, wherein: The wall diagnosis further includes: Performing (213) wall type classification and determining the wall type (123) of the wall (105) by the diagnostic module (107); and / or performing (215) object depth determination by the diagnostic module (107) and determining an object depth (119) of the object (113) in the wall (105), wherein the object depth (119) is defined as the distance from the object (113) to the surface of the wall (105); and / or The object extension dimension determination is implemented (217) by the diagnostic module (107) and the object extension dimension (121) of the object (113) along a predefined direction is determined, wherein the uncertainty value (110) also includes at least a probability value of the wall type (123) and / or a probability value of the object depth (119) of the object (113) and / or a probability value of the object extension dimension (121).

3. The method (200) according to claim 1 or 2, wherein: In the object recognition, multiple possible object positions (115) and / or multiple possible object types (117) of the object (113) are determined as independent diagnosis results (109), and / or, in the wall type classification, multiple possible wall types (123) of the wall (105) are determined as independent diagnosis results (109), and / or, in the object depth determination, multiple possible object depths (119) are determined as independent diagnosis results (109), and / or, in the object extension scale determination, multiple possible object extension scales (121) of the object (113) are determined as independent diagnosis results (109), wherein each of the multiple diagnosis results (109) is provided with a probability value, and wherein the multiple diagnosis results (109) and the multiple probability values ​​are displayed on the display unit (111).

4. The method (200) of claim 3, further comprising: A selection function (116) is provided (219), wherein at least one of the displayed diagnosis results (109) can be selected by a user of the measuring device (100) implementing the selection function (116).

5. The method (200) of claim 4, further comprising: receiving (221) a selection instruction of the selection function (116), wherein one of the displayed diagnosis results (109) is selected by the selection instruction made by the user; Feedback information (112) is provided (223) to an external server unit (114), wherein the feedback information (112) includes the selection instruction and the diagnosis result (109) selected therein together with the corresponding radar data (103).

6. The method (200) according to any one of the preceding claims, wherein: The measuring device (100) further comprises at least one inductive sensor and / or eddy current sensor and / or capacitive sensor and / or AC sensor and / or nuclear magnetic resonance sensor and / or ultrasonic sensor for providing additional sensor data, wherein the diagnostic module (107) is configured to perform the wall diagnosis while taking into account the additional sensor data.

7. The method (200) according to any one of the preceding claims, wherein: The diagnostic module (107) comprises at least one correspondingly trained artificial intelligence (125) which is configured to perform object recognition and / or wall classification and / or object depth determination based on the radar data (103) and / or the additional sensor data.

8. The method (200) according to any one of the preceding claims, wherein: The object categories of the object type (117) of the object (113) include: metal / non-metal objects, cables for low voltage, cables with single-phase AC signals, cables with multi-phase AC signals, wooden beams, metal beams, plastic pipes, water-filled plastic pipes, such as water supply pipes, non-water-filled plastic pipes, such as drainage pipes, and / or wherein the wall type categories of the wall type (123) of the wall (105) include: concrete wall, lightweight structure / dry structure wall, brick wall and / or brick blocks of wall, floor heating, wall heating.

9. A method (400) for training an artificial intelligence (125) of a measuring device (100) for wall diagnosis, the method comprising: providing (401) a training data set (143) for training the artificial intelligence (125), wherein the training data set (143) includes radar data (103) mapping a wall (105) and objects (113) constructed in the wall (105) and feedback information (112) provided by the method (200) for operating a measuring device (100) according to claim 5; The artificial intelligence (125) is trained (403) based on the training data set (143) and taking into account the feedback information for performing object recognition of objects (113) constructed in the wall (105), wherein the object recognition includes at least object detection and object classification.

10. A computing unit (151) configured to implement a method (200) for operating a measuring device (100) according to any one of claims 1 to 8 and / or a method (400) for training an artificial intelligence (125) of a measuring device (100) for wall diagnosis according to claim 9.

11. A computer program product (500), comprising instructions which, when executed by a data processing unit, cause the data processing unit to execute the method (200) for operating a measuring device (100) according to any one of claims 1 to 8 and / or the method (400) for training an artificial intelligence (125) of a measuring device (100) for wall diagnosis according to claim 9.

Images