Accelerating insights, improving efficiency, and enabling predictive maintenance in test and measurement systems through artificial intelligence assistants

Abstract

A test and measurement instrument comprises one or more ports for connecting to a device under test (DUT), a user interface with one or more controls, a display, a memory, one or more processors for receiving test signals from the DUT via the one or more ports as a test of the DUT, using the test signals to generate test data, displaying test data on the display, displaying a control button on the user interface indicating that an artificial intelligence assistant (AI assistant) is available, receiving input via the control button to launch the AI assistant, providing areas on the user interface to allow the user to interact with the AI assistant, and applying a machine learning model represented by the AI assistant to provide additional information to the user,relating to one or more of the tests and the DUT.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This disclosure claims priority under 35 USC § 119 to Indian Provisional Patent Application No. 202421002876, entitled "ACCELERATION INSIGHTS, ENHANCING EFFICIENCY, AND ENABLING PREDICTIVE MAINTENANCE IN TEST AND MEASUREMENT SYSTEMS USING ARTIFICIAL INTELLIGENCE ASSISTANT," filed on January 15, 2024, the disclosure of which is incorporated herein by reference in its entirety. FIELD OF TECHNOLOGY

[0002] This disclosure relates to artificial intelligence (AI), specifically an AI assistant for use in test and measurement systems to accelerate insights, increase efficiency, and enable predictive maintenance. BACKGROUND

[0003] AI and machine learning (ML) have proven to be powerful tools for analyzing data and extracting insights from test data. AI-driven models can autonomously interpret complex data patterns, enabling more efficient and accurate analysis of test results. Machine learning, with its ability to adapt and improve performance over time, plays a critical role in predictive maintenance, which detects potential equipment failures before they occur.

[0004] Currently, ML and AI can make these predictions and perform some of these types of data analyses, but they can require long testing times. Furthermore, current approaches rely on pre-trained models. Using pre-trained models requires collecting large amounts of training data, training the model, and then validating it. This process takes longer than desired and disrupts the user's workflow to enable the testing cycle. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows an embodiment of an architecture for an AI assistant Fig. 2 shows an embodiment of training settings of the AI assistant in a measurement setting. Fig. Figure 3 shows an embodiment of a user interface showing an AI assistant indicating required changes to a running test. Fig. Figure 4 shows an embodiment of an AI assistant in support mode that helps the user change a specific setting to obtain an accurate measurement. Fig. Figure 5 shows an embodiment of a user interface indicating a setting change that eliminates a measurement warning and enables an accurate measurement Fig. Figure 6 shows an embodiment of a user interface resulting from input on the assistant icon for the natural language “AI Assistant Interface.” Fig. Figure 7 shows an embodiment of a user interface resulting from input from the user interface for chatting with an “AI assistant interface.” Fig. Figure 8 shows an embodiment of a chat panel with a suggested action button. Fig. Figure 9 shows an embodiment of an AI assistant-suggested phase diagram with the DQ0 component added for advanced analysis. Fig. Figure 10 shows an embodiment of a proposed control loop response (Bode plot) with extrapolated data for frequencies above 100 kHz. Fig. 11 shows an embodiment of a flowchart for an AI assistant. Fig. Figure 12 shows an embodiment of a training and validation efficiency representation. DETAILED DESCRIPTION

[0005] The embodiments presented here comprise an AI-powered test and measurement system that extends the capabilities of test and measurement systems, provides unparalleled insights, and improves testing and cost efficiency. The embodiments presented here can autonomously interpret complex data patterns and enable more accurate analysis of test results. The solutions presented here improve performance over time and play a critical role in predictive maintenance by identifying potential equipment failures before they occur. The devices use machine learning models that develop and train in real time while the user continues to use the test and measurement instrument during development and validation.

[0006] The embodiments described herein include an "AI assistant," which refers to a machine learning model. In the discussion herein, these terms are used interchangeably, so that references to the AI assistant refer to the machine learning model's user interface. Unlike previous machine learning models, the models presented here do not change the user's workflow. As the user performs tests on the devices under test (DUT), the embodiments collect data from the user's use of the test and measurement instrument to first define and build the model and then train the model. The embodiments provide a model that can be deployed across multiple test and measurement endpoints, enabling consistent results across multiple devices.Version control allows other endpoints to update the machine learning model without overwriting previous versions. Similarly, embodiments extend the model to be deployed across multiple measurements, leveraging the same data collection and gaining insights. External users of the system, which includes all endpoints and the model repository, can use a subscription service to provide these external users with access to machine learning models tailored to their needs.

[0007] Fig. Figure 1 shows the architecture of one embodiment of a system that creates and deploys one embodiment of an AI assistant. The instrument 10, referred to herein as an oscilloscope or "osci," can be any test and measurement instrument. No limitation to an oscilloscope should be implied. In a test environment, the oscilloscope 10 is connected to the DUT 12. The oscilloscope may include one or more processors (e.g., 20), a user interface (U / I) 22 with one or more controls, a display 24, and memory 26. The controls may include, for example, buttons, knobs, and sliders and may be located on the instrument body, as touch buttons on the display as part of the user interface, or on both. The display may comprise at least a portion of the user interface.The reference to a storage may include both local storage and a connection to external storage, including one or more databases such as or cloud storage 16. The cloud storage may be connected to many different elements of the architecture, but most likely at least to the database 14 and the neural network 18.

[0008] When the user operates the test and measurement instrument 10, the instrument receives test signals from the DUT 12. In some cases, the processor(s) 20 cause signals to be applied to the DUT 12 to initiate the test. The DUT 12 generates test signals in response to the signal applied to it. In other cases, the instrument receives test signals from the DUT without requiring any signals to be applied to the DUT to initiate the test. The test signals received by the instrument can be analog or digital signals received via connector 28. In a common example, the DUT 12 generates analog signals, which the instrument 10 converts into test data, such as a waveform. The instrument then displays the test data on the device display.

[0009] For example, during the deployment of the test and measurement instrument as described above, the instrument would export the test data and associated metadata to 30. The metadata may include user inputs, oscilloscope and probe parameters, state-of-the-art information such as standards and versions, voltage levels, currents, temperatures, and other operating parameters. The oscilloscope also imports the data into database 14 at 32. The raw waveforms and user interactions are automatically labeled at 34. Automatic labeling classifies and removes unwanted data that is not useful for further analysis suggestions. The automatically labeled data is also sent to database 14. The system can employ efficient storage management by removing the unwanted data and compressing, serializing, and storing it in real time in the cloud, on the instrument, or in one or more databases.The trained weights are portable and can include version control within the organization.

[0010] As the oscilloscope firmware receives sampled analog data, minimal labeling of the data characteristics is performed at the firmware level. These labels may include the sampling rate, the time / frequency domain of the data, the technology being tested, the data based on the probe used, etc. All data accepted by the oscilloscope is classified based on the usage pattern metadata and the user's model suggestion. The labeling system according to some embodiments of the disclosure may include a monitoring block that performs weak labeling of data with probabilistic values.

[0011] A classifier can be housed in the oscilloscope for each technology analysis. For example, the sequential ensemble technique can be used to combine each technology classifier to achieve a final classification. With waveform data associated with a likely technology, suggested measurements, charts, and predictive analysis for waveform data are provided.

[0012] Back to Fig. 1: This information is also used to determine whether there is enough data to select a suitable model (36). The architecture may offer the possibility of generating a model using a plug-in or a predefined program (38). Returning to 36: If there is not enough data available, the process waits until more data is available, which is indicated by the return path to the oscilloscope. If a suitable model is available, the neural network 18 performs 40 calculations to select the appropriate model for the neural network 18.

[0013] The pseudocode below shows one implementation of extracting features from the data, collecting data samples, and selecting the model. It also demonstrates validation and deciding whether the model is ready for the user or requires more data.

[0014] If the corresponding model (42) is available, it processes the data collected at the instrument to provide predictive data (44) and generates graphical and scalar results (46). Once the model has enough data to train itself and operate, an AI assistant interface 50 becomes available on the instrument's interface. When the user interacts with the AI assistant, the AI assistant can use a large language model (LLM) (52) to perform an action (56) and update the knowledge base (54).

[0015] The system architecture provides the flexibility to store system elements on the instrument, in the cloud, or on another computer connected to the instrument. Those system elements that can be stored in the cloud carry a cloud assignment 16.

[0016] Fig. Figure 2 shows one embodiment of a measurement setting for training the AI assistant. The first box 60 shows the settings screen with the training process turned off. Selected options in subsequent screens are surrounded by a darker line, as is the case with OFF in box 60. The second box 62 shows that the setting is turned ON and the user has selected the AUTO setting. This would cause the system to train the AI assistant in a pre-trained, automatic manner. Box 64 shows one embodiment of some labels for CUSTOM training. The user has set the percentage of data distribution between the training set and the cross-validation set to 80 percent for training. The user can set which parts of the dataset are used for training and validation.The user has also selected a set of labels for the data for the automatic labeling process described above. In the lower part of the window, where AI Assistant training is enabled, the user has the option to import the model from the model repository. This might be the case, for example, if the model already exists and is undergoing retraining or fine-tuning. The user can also export the model once retraining and / or fine-tuning is complete.

[0017] In the following figures, the availability of the AI Assistant is shown as an AI Assistant icon, which is Fig. 3 is labeled 70 both on the screen and in the exploded view. In some embodiments, user interactions with the AI assistant may have various options. For example, three known mouse inputs may be used: a double-click, a single-click, or a right-click. Each of these inputs would result in a different response from the AI assistant. If the user selects one of the above inputs, such as a double-click, the AI assistant may automatically perform the action indicated by the position of the AI assistant icon. With another option, such as a single-click, the AI assistant may enter an "assist" mode in which the AI assistant guides the user through the steps needed to perform the specified action. With another option, e.g.With a right-click, the AI assistant can open a chatbot and allow the user to ask questions. This will most likely involve the aforementioned LLM so that the AI assistant can "understand" the questions and provide answers.

[0018] As a first example of interaction with the AI Assistant, the AI Assistant indicates that the current setting requires changes to capture the waveform and perform a measurement. The problem can be seen in the Output window 76. In this discussion, the output takes the form of a waveform, except that no waveform appears. The appearance of the AI Assistant icon 70 in the "Meas 1" button 72 indicates that the AI Assistant has a solution for the empty input. Similarly, the AI Assistant icon appears on the "Ch 1" button 74. The user can click either icon in any of the three input modes described above. It should be noted that a typical instrument display will show many different elements than those in Fig. 3-10. These are not shown in the figures to simplify the illustration, but this does not mean that these elements are not present in the following figures.

[0019] Fig. Figure 4 shows how the AI Wizard enters a "assistive" mode. In this mode, the AI Wizard can guide the user through the process of correcting the measurement settings. The AI Wizard opens the Settings window 77, which displays the current setting previously selected by the user. The AI Wizard indicates the settings that need to be changed. In the Settings window, the AI Wizard indicates that the user should select the SETUP button 79, highlighted by the dashed lines. The AI Wizard also indicates that the user should change the setting from Vds to Id at 75. Fig. Figure 4 also shows the icons in the bottom bar indicating Ch 1, 74, Ref 2 73, and Ref 3 80. Note that these windows are not shown in later figures.

[0020] Fig. 5 shows the change in setting at 75. The partially hidden output window 76 shows that waveforms now appear. Fig. Figure 6 shows another example of interactions with the AI assistant. In this example, the user has chosen to chat with the AI assistant, which is shown in Figure 82.

[0021] Fig. 7 shows window 84, where the user can select whether to chat with the AI assistant. Fig. 8 shows the chat window 86 as part of the display. The chat window offers the AI assistant the opportunity to make suggestions and give the user the opportunity to perform an action to implement the suggestion. In the Fig. In the example shown in Figure 8, the AI Assistant recommended that the user add a DQ0 representation to support the analysis. The AI Assistant also provided a button 88 to make this addition.

[0022] Fig. 9 shows the resulting DQ0 representation and the phase diagram 90. Note that in Fig. 9 the middle window of the display in Fig. 8 was removed to make more room for the phase diagram and chat window. The display would still show the center window.

[0023] As already mentioned, the AI assistant is able to predict results of measurements that the instrument has not actually performed or calculated. Fig. Figure 10 shows an example of this. As can be seen in window 88 of the Harmonics / FRA results, the instrument has calculated index, frequency, gain, linear gain, and phase for several indices 32-43. As the AI Wizard icon next to indices 44-54 indicates, the AI Wizard predicted these measurements rather than the instrument actually having to perform these calculations. The AI Wizard extrapolated this data from the previous measurements. This saves the user considerable time. Reliability testing involves acquiring very long events, sometimes more than a million events, which take tens of thousands of hours of test time and are calculated over a range of operating parameters. The approach described in the working examples helps with prediction with smaller data sets, as shown above, and is not limited to one measurement but can be extended to multiple measurements.

[0024] Fig. Figures 3-10 show examples of some of the AI Assistant's features. These examples are not intended to demonstrate all of the AI Assistant's capabilities, but are intended merely as examples.

[0025] Fig. Figure 11 shows a flowchart of an embodiment of the AI assistant. Similar to the system architecture of Fig. 1, the process begins with the acquisition of test data. In the example of Fig. 11, the test data takes the form of waveforms 100. The waveforms 100 and the trainings 102 are used to perform measurements 1 to n, 104 to 106. The waveforms undergo data preprocessing and cleaning 110. The results 107 and the other information constitute the metadata 108. As already mentioned, the data undergoes weak labeling and feature extraction 114 as part of the auto-labeling process at 112.

[0026] The labeling system includes a monitoring block that performs weak labeling with probabilistic values. The tool may contain a classifier for each technology analysis. For example, the sequential ensemble technique can combine each technology classifier to achieve the final classification. After the test data has been classified as a probable technology, the AI assistant can make suggestions for measurements, charts, and predictive analyses for the test data.

[0027] Returning to the flowchart, at 116, it is determined whether there are enough samples or not. If not, the process continues and waits at 118. If there are enough samples at 116, the process builds the model according to the original process or retrains / fine-tunes the model at 120. The model is then validated at 122. If the predicted values match the actual values (124), the AI assistant can display the results and provide further information (126). If the result at 124 does not match, the model can be fine-tuned at 128 until the result at 124 matches.

[0028] The term "test data" used here refers to the data from the test, such as the data represented by the waveform. The term "operational data" used here refers to the test data after preprocessing / cleaning, automatic labeling, and feature extraction, as this is the data the model operates on. The data associated with the test, including the test settings, measurements, results, and possibly others, comprises the metadata.

[0029] The operational data and metadata are stored in the database 14. The user can select the type of data storage between the database 14 and the cloud 16 at 130. The data store provides the user interface 50. Fig. 1, in this case as a chatbot that uses the LLM at 52 to store information in the knowledge base 54 and take actions at 56. This flowchart represents one embodiment of a process flow as an example. A limitation to a particular implementation or order of these processes is not intended and should not be implied.

[0030] As mentioned above, the user can customize the portions of training data collected during operation between training and validation. The test and measurement tool can generate graphs such as an efficiency chart, a loss chart, and histograms of the machine learning model's performance based on the analysis. Fig. Figure 12 shows an efficiency diagram between the training accuracy 132 and the validation accuracy 134.

[0031] In this way, the AI assistant expands the capabilities of test and measurement instruments. Benefits include reduced test times and predictive interpretation of measurement results by developers and validation engineers. The user can easily gain additional insights, for example, into DUT parameters such as DUT durability and time-to-failure prediction. This helps validation engineers with predictive maintenance. As mentioned earlier, the system here does not change the customer's existing workflow; rather, the proposed algorithm learns on the fly as the user continues to use the oscilloscope during its design and validation. This model provides additional insights, such as failure points and the decision for selecting the optimal filters from a large set.

[0032] Aspects of the disclosure may operate on specially designed hardware, firmware, digital signal processors, or on a specially programmed general-purpose computer having a processor that operates according to programmed instructions. As used herein, the terms "controller" or "processor" are intended to encompass microprocessors, microcomputers, application-specific integrated circuits (ASICs), and special-purpose hardware controllers. One or more aspects of the disclosure may be embodied in computer-usable data and computer-executable instructions, such as one or more program modules executed by one or more computers (including supervisory modules) or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc.that perform specific tasks or implement specific abstract data types when executed by a processor in a computer or other instrument. The computer-executable instructions may be stored on a non-transitory, computer-readable medium such as a hard disk, an optical disk, a removable storage medium, solid-state memory, random access memory (RAM), etc. As will be apparent to one skilled in the art, the functionality of the program modules may be arbitrarily combined or distributed in various aspects. Furthermore, the functionality may be embodied in whole or in part in firmware or hardware equivalents such as integrated circuits, FPGAs, and the like.Certain data structures may be used to more effectively implement one or more aspects of the disclosure, and such data structures are contemplated as part of the computer-executable instructions and computer-usable data described herein.

[0033] The disclosed aspects may, in some cases, be implemented in hardware, firmware, software, or a combination thereof. The disclosed aspects may also be implemented in the form of instructions stored on one or more non-transferable computer-readable media that can be read and executed by one or more processors. Such instructions may be referred to as a computer program product. Computer-readable media, as described herein, is any media accessible by a computer. Computer-readable media may include, for example, but is not limited to, computer storage media and communications media.

[0034] Computer storage media is any media that can be used to store computer-readable information. Examples of computer storage media include RAM, ROM, EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory or other storage technologies, CD-ROM (Compact Disc Read-Only Memory), DVD (Digital Video Disc) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, and any other volatile or non-volatile, removable or non-removable media employed in any technology. Computer storage media excludes signals as such and transient forms of signal transmission.

[0035] Communication media refers to any medium that can be used to transmit computer-readable information. Examples of communication media include coaxial cable, fiber optic cable, air, or any other medium suitable for transmitting electrical, optical, radio frequency (RF), infrared, acoustic, or other signals. EXAMPLES

[0036] Examples of the technologies disclosed herein are listed below. An embodiment of the technologies may include one or more, and any combination, of the examples described below.

[0037] Example 1 is a test and measurement instrument comprising: one or more ports for connecting to a device under test (DUT); a user interface having one or more controls; a display; memory; and one or more processors configured to execute code that causes the one or more processors to: receive test signals from the DUT via the one or more ports as a test of the DUT; use the test signals to generate test data; display test data on the display; display a control button on the user interface indicating that an artificial intelligence (AI) assistant is available; receive input via the control button from a user to launch the AI assistant; provide areas on the user interface to enable the user to interact with the AI assistant;and after receiving inputs through the areas on the user interface, applying a machine learning model represented by the AI assistant to provide the user with additional information related to one or more of the tests and the DUT;

[0038] Example 2 is the test and measurement instrument of Example 1, wherein the code that causes the one or more processors to provide additional information includes code that causes the one or more processors to provide recommendations for measurement settings, generate additional results to the results selected by the user, recommend additional instruments, and predict time to failure for the DUT or components on the DUT.

[0039] Example 3 is the test and measurement instrument of either Example 1 or 2, wherein the code that causes the one or more processors to provide regions on the user interface includes code that causes the one or more processors to display control buttons of the AI assistant at relevant locations on the display.

[0040] Example 4 is the test and measurement instrument of Example 3, wherein the one or more processors are further configured to execute code to receive input via one of the control buttons and respond to the input, wherein the code to cause the one or more processors to respond to the input comprises code to cause the one or more processors to: automatically perform a recommended action represented by the control button when the input comprises a first input; display steps that enable the user to perform the recommended action when the input comprises a second input; and display an interactive window that enables the user to interact with the AI assistant when the input comprises a third input.

[0041] Example 5 is the test and measurement instrument of any of Examples 1 to 4, wherein the one or more processors are further configured to execute code that causes the one or more processors to generate and train the machine learning model.

[0042] Example 6 is the test and measurement instrument of Example 5, wherein the one or more processors are configured to generate and train the machine learning model while the user uses the test and measurement instrument without interrupting the user's workflow.

[0043] Example 7 is the test and measurement tool of any of Examples 1 to 6, wherein the code that causes the one or more processors to generate the machine learning model comprises code that causes the one or more processors to preprocess the data before the one or more processors collect data, including test data and associated metadata; extracting main feature data from the data; analyzing the linearity of the main feature data to select an activation function; using the activation function to activate portions of a neural network to create the machine learning model; using a portion of the main feature data to train the machine learning model; and using another portion of the main feature data to validate the machine learning model.

[0044] Example 8 is the test and measurement instrument of 7, wherein the one or more processors are further configured to execute code that causes the one or more processors to preprocess the data prior to extracting key feature data.

[0045] Example 9 is the test and measurement instrument of any of Examples 1 to 8, wherein the one or more processors are further configured to execute code that causes the one or more processors to manage storage of the data by automatically labeling the data to classify the data, removing unwanted data, and compressing and serializing the data, and then backing up the data in real time.

[0046] Example 10 is the test and measurement instrument of any of Examples 1 to 9, wherein the one or more processors are further configured to execute code to cause the one or more processors to tune the machine learning model during use of the test and measurement instrument to keep the machine learning model up to date.

[0047] Example 11 is a method for deploying an artificial intelligence assistant (AI assistant) with a test and measurement instrument, comprising: receiving test signals from the DUT via a port as a test of the DUT; using the test signals to generate test data; displaying the test data on a display; displaying a control button on the user interface indicating that an artificial intelligence assistant (AI assistant) is available; receiving input via the control button from a user to launch the AI assistant; providing areas on the user interface to enable the user to interact with the AI assistant;and after receiving inputs through the areas on the user interface, using a machine learning model connected to the AI assistant to provide the user with additional information related to one or more of the tests and the DUT;

[0048] Example 12 is the method of Example 11, wherein providing additional information includes providing recommendations for measurement settings, generating additional results to the results selected by the user, recommending additional instruments, and predicting lifetime for the DUT or components on the DUT.

[0049] Example 13 is the method of examples 11 or 12, wherein providing areas on the user interface comprises displaying control buttons at relevant locations on the display of the test data.

[0050] Example 14 is the method of Example 13, further comprising receiving input via one of the control buttons and responding to the input by: performing a recommended action represented by the control button if the input includes a first input; displaying steps that enable the user to perform the recommended action if the input includes a second input; and displaying an interactive window that enables the user to interact with the AI assistant if the input includes a third input.

[0051] Example 15 is the method of any one of Examples 11 to 14, further comprising generating the machine learning model.

[0052] Example 16 is the method of Example 15, wherein generating the machine learning model comprises generating the machine learning model when no machine learning model exists.

[0053] Example 17 is the method of Example 15, wherein generating the machine learning model comprises training the machine learning model in real time.

[0054] Example 18 is the method of Example 15, wherein generating the machine learning model occurs while the user is using the test and measurement instrument without interrupting the user.

[0055] Example 19 is the method of Example 15, wherein generating the machine learning model comprises: collecting the test and associated metadata as data; extracting main feature data from the data; analyzing the linearity of the main feature data and selecting an activation function; using the activation function to activate portions of a neural network to create the machine learning model; using an adjustable portion of the main feature data to train the machine learning model; and using another adjustable portion of the main feature data to validate the machine learning model.

[0056] Example 20 is the method of Example 17, further comprising preprocessing the data before extracting key features from the data.

[0057] Example 21 is the method of any one of Examples 11 to 20, further comprising managing the storage of the data by automatically tagging the data to classify the data, removing unwanted data, and compressing and serializing the data, and then backing up the data in real time.

[0058] Example 22 is the method of any one of Examples 1 to 21, further comprising tuning the machine learning model during use of the test and measurement tool to keep the model up to date.

[0059] Example 23 is the method of Example 15, further comprising sharing the machine learning model among multiple test and measurement endpoints after the machine learning model is generated.

[0060] Example 24 is the method of Example 23, wherein sharing the machine learning model among multiple endpoints further comprises providing version control when other endpoints update the machine learning model.

[0061] Example 25 is the method of any one of Examples 1 to 24, further comprising using a subscription service to provide external users with access to customize and optimize the machine learning model for specific applications and requirements.

[0062] Furthermore, this written description refers to specific features. It is understood that the disclosure in this specification encompasses all possible combinations of these particular features. Where a particular feature is disclosed in connection with a particular aspect or example, that feature may, where possible, also be used in connection with other aspects and examples.

[0063] Although this application refers to a method comprising two or more defined steps or operations, the defined steps or operations may be performed in any order or simultaneously, unless the context precludes such possibilities.

[0064] All features disclosed in the description, including the claims, the abstract, and the drawings, and all steps in any disclosed method or process may be combined in any combination, except for combinations in which at least some of those features and / or steps are mutually exclusive. Any feature disclosed in the description, including the claims, the abstract, and the drawings, may be replaced by alternative features serving the same, equivalent, or similar purpose, unless expressly stated otherwise.

[0065] Although specific examples of the invention have been shown and described for purposes of illustration, various modifications may be made without departing from the spirit and scope of the invention. Accordingly, the invention should not be limited except as by the appended claims. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] IN 202421002876

[0001]

Claims

[1] A test and measuring instrument comprising: one or more terminals for connecting to a device under test (DUT); a user interface with one or more controls; a display; a memory; and one or more processors configured to execute code that causes the one or more processors to: Receiving test signals from the DUT via the one or more ports as a test of the DUT; Using the test signals to generate test data Display test data on the display; Displaying a control button on the user interface indicating that an artificial intelligence assistant (AI Assistant) is available; Receiving control key input from a user to launch the AI assistant; Providing areas on the user interface to enable the user to interact with the AI assistant; and When receiving inputs through the areas on the user interface, apply a machine learning model represented by the AI assistant to provide the user with additional information related to one or more tests and the DUT. [2] The test and measurement instrument of claim 1, wherein the code that causes the one or more processors to provide additional information comprises code that causes the one or more processors to provide recommendations for measurement settings, generate additional results to the user-selected results, recommend additional instruments, and predict the time to failure of the DUT or components on the DUT. [3] The test and measurement instrument of claim 1 or 2, wherein the code that causes the one or more processors to provide regions on the user interface comprises code that causes the one or more processors to display control buttons of the AI assistant at relevant locations on the display. [4] The test and measurement instrument of claim 3, wherein the one or more processors are further configured to execute code to receive an input via one of the control buttons and to respond to the input, wherein the code that causes the one or more processors to respond to the input comprises code that causes the one or more processors to: Automatically perform a recommended action represented by the control key when the input includes a first input; Displaying steps that enable the user to perform the recommended action when the input includes a second input; and Display an interactive window so the user can interact with the AI assistant when the input includes a third input. [5] The test and measurement instrument of any one of claims 1 to 4, wherein the one or more processors are further configured to execute code that causes the one or more processors to generate and train the machine learning model. [6] The test and measurement instrument of claim 5, wherein the one or more processors are configured to generate and train the machine learning model while the user uses the test and measurement instrument without interrupting the user's workflow. [7] The test and measurement instrument of any one of claims 1 to 6, wherein the code that causes the one or more processors to generate the machine learning model comprises code that causes the one or more processors to preprocess the data before the one or more processors perform: Collecting data consisting of the test data and associated metadata; Extracting key features from the data; Analyze the linearity of the main feature data to select an activation function; Using the activation function to activate parts of a neural network to build the machine learning model; Using a portion of the main feature data to train the machine learning model; and Using a different part of the main feature data to validate the machine learning model. [8] The test and measurement instrument of claim 7, wherein the one or more processors are further configured to execute code that causes the one or more processors to pre-process the data prior to extracting key feature data. [9] The test and measurement instrument of any one of claims 1 to 8, wherein the one or more processors are further configured to execute code that causes the one or more processors to manage storage of the data by automatically labeling the data to classify the data, removing unwanted data, and compressing and serializing the data and then backing up the data in real time. [10] The test and measurement instrument of any one of claims 1 to 9, wherein the one or more processors are further configured to execute code to cause the one or more processors to tune the machine learning model during use of the test and measurement instrument to keep the machine learning model up to date. [11] A method of using an artificial intelligence (AI) assistant with a test and measurement instrument, comprising: Receiving test signals from the DUT via a port as a test of the DUT; Using the test signals to generate test data; Displaying the test data on a display; Displaying a control button on the user interface indicating that an artificial intelligence assistant (AI Assistant) is available; Receiving control key input from a user to launch the AI assistant; Providing areas on the user interface to enable the user to interact with the AI assistant; and When receiving inputs through the areas on the user interface, use a machine learning model connected to the AI assistant to provide the user with additional information related to the test or DUT. [12] The method of claim 11, wherein providing additional information comprises providing recommendations for measurement settings, generating additional results to the results selected by the user, recommending additional instruments, and predicting lifetime expectancy for the DUT or components on the DUT. [13] The method of claim 11 or 12, wherein providing areas on the user interface comprises displaying control buttons at relevant locations on the display of the test data. [14] The method of claim 13, further comprising receiving an input via one of the control buttons and responding to the input, comprising: Performing a recommended action represented by the control key when the input includes a first input; Displaying steps that enable the user to perform the recommended action when the input includes a second input; and Display an interactive window so the user can interact with the AI assistant when the input includes a third input. [15] The method of any one of claims 11 to 14 further comprising generating the machine learning model. [16] The method of claim 15, wherein generating the machine learning model comprises generating the machine learning model if no machine learning model exists. [17] The method of claim 15 or 16, wherein generating the machine learning model comprises training the machine learning model in real time. [18] The method of any one of claims 15 to 17, wherein generating the machine learning model occurs while the user is using the test and measurement instrument without disturbing the user. [19] The method of any one of claims 15 to 18, wherein generating the machine learning model comprises: Collecting the test and associated metadata as data; Extracting key feature data from the data; Analyze the linearity of the main feature data and select an activation function; Using the activation function to activate part of a neural network to build the machine learning model; Using an adjustable portion of the main feature data to train the machine learning model; and Using a different adjustable part of the main feature data to validate the machine learning model. [20] The method of any one of claims 17 to 19 further comprising pre-processing the data prior to extracting key feature data from the data. [21] The method of any one of claims 11 to 20 further comprising managing the storage of the data by automatically tagging the data to classify the data, removing unwanted data, compressing and serializing the data, and then backing up the data in real time. [22] The method of any one of claims 11 to 21 further comprising tuning the machine learning model during use of the test and measurement instrument to keep the model up to date. [23] The method of any one of claims 15 to 22 further comprising sharing the machine learning model between multiple test and measurement endpoints after the machine learning model has been generated. [24] The method of claim 23, wherein sharing the machine learning model among multiple endpoints further comprises providing version control when other endpoints update the machine learning model. [25] The method according to any one of claims 11 to 24 further comprising using a subscription service to provide external users with access to adapt and optimize the machine learning model for specific applications and requirements.

Images