Automatic test method, device and system for airborne system
Through the automated testing method of airborne systems, combined with voice and video recognition technology, the uncertainty and non-repeatability problems caused by traditional audio testing relying on manual labor are solved, and fast and accurate detection and evaluation of audio and video system failures are achieved.
Patent Information
- Application Number
- CN202411192185.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-26
AI Technical Summary
Audio testing of traditional airborne systems relies on manual evaluation, with uncertainty and lack of unified standards, and the inability to conduct quantitative evaluation, resulting in low efficiency and unrepeatable test results.
Using the automated test method of airborne system, combined with voice and video recognition technology, audio and video data is collected through the pickup and video collector, parameters such as delay, frequency, length of alarm sounds are identified and parsed, equipment failure is judged, and voice exciter is used. Test with video exciter.
It realizes automatic audio testing of airborne systems, quickly and accurately discovers audio and video system problems, provides qualitative and quantitative evaluation results, and improves testing efficiency and reliability.
Smart Images

Figure CN119031122A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of testing technology, and in particular relates to an airborne system automatic testing method, device and system. Background Art
[0002] As the information management and communication center of the helicopter, the helicopter avionics system is responsible for the communication and interconnection between various devices. Voice, as an important part of the interaction of airborne systems, plays a vital role in the communication between the aircraft and the outside world. With the continuous development of artificial technology, intelligent voice interaction technology is being applied and implemented, and with the continuous development of intelligent airborne systems, more and more audio human-computer interaction is being applied to assisted flight.
[0003] Traditional airborne system warning and voice tests are evaluated and judged by testers who listen directly to the audio, which is highly uncertain, and different testers have different sensory sensitivities to sound. In addition, there is currently no unified test standard for audio testing, or the test standard is not objective enough. Moreover, there is no quantitative description of the frequency, volume, timbre, content, etc. of the audio, so quantitative evaluation is impossible. For the same test item, the test record results are not repeatable, and the test results mainly depend on the richness of the tester's experience.
[0004] Therefore, there is an urgent need to provide an automated audio testing method for airborne systems based on speech and video recognition, and then to qualitatively and quantitatively evaluate the test results. Summary of the invention
[0005] In order to solve the technical problem that the airborne audio system mainly relies on manual testing, which is inefficient and has poor test results, the present invention provides an airborne system automated testing method, device and system, which can quickly and accurately find problems in the audio and video system during the avionics joint test, provide sound recognition and analysis tools, can receive equipment prompt sounds such as warning sounds, and analyze the delay, frequency, length, and number of repetitions of the prompt sounds; and provide a video recognition function, combined with the complete logic of the equipment warning sound, to determine whether it meets the equipment function requirements according to the test requirements. The technical solution is as follows:
[0006] In a first aspect, an airborne system automated test system is provided, comprising: an airborne system automated test device, a sound pickup, a video collector, an audio device under test, and a video device under test;
[0007] The airborne system automated test device is used to generate fault information of the airborne system and send the fault information to the video device under test and the audio device under test;
[0008] The video device under test is used to display the fault information on the screen; the audio device under test is used to play the fault information in voice;
[0009] The video collector is used to collect the video data displayed by the video device under test and send it to the airborne system automatic test device;
[0010] The pickup is used to collect audio data output by the audio device under test and send it to the airborne system automated test device;
[0011] The airborne system automated test device is used to identify and compare video data and audio data to determine whether the tested video equipment and tested audio equipment are faulty.
[0012] Among them, the airborne system automated testing device is used to determine that the audio device under test has no fault when the fault reflected by the audio data is consistent with the actual fault and the preset audio control parameters are consistent with the parameters of the audio data; otherwise, it is determined that the audio device under test has a fault.
[0013] Among them, the airborne system automatic testing device is used to determine that the video device under test has no fault when the fault reflected by the video data is consistent with the actual fault and the preset video control parameters are consistent with the parameters of the video data; otherwise, it is determined that the video device under test has a fault.
[0014] The audio control parameters include: audio delay, frequency, length, and number of repetitions.
[0015] Among them, the video control parameters include: duration, dynamic display frequency, and number of repetitions.
[0016] In a second aspect, an airborne system automated testing device is provided, comprising: a test case writing and execution module, a voice actuator, a video actuator, a voice recognition module, a video recognition module, and a recognition result comparison module.
[0017] The test case writing and execution module is used to generate fault information of the airborne system and send the fault information to the voice actuator and the video actuator; the voice actuator sends the fault information to the audio device under test according to the preset communication protocol requirements, and the video actuator sends the fault information to the video device under test according to the preset communication protocol requirements;
[0018] The video recognition module is used to receive the video data collected by the video collector and send it to the recognition result comparison module;
[0019] The speech recognition module is used to receive the audio data collected by the microphone and send it to the recognition result comparison module;
[0020] The recognition result comparison module is used to recognize and compare the video data and the audio data to determine whether the tested video device and the tested audio device are faulty.
[0021] Furthermore, the device further comprises: a speech analysis module, the speech analysis module being used to determine parameters of the audio data collected by the microphone;
[0022] The recognition result comparison module is used to determine that the audio device under test has no fault when the fault reflected by the audio data is consistent with the actual fault and the preset audio control parameters are consistent with the parameters of the audio data; otherwise, it is determined that the audio device under test has a fault.
[0023] Among them, the recognition result comparison module is used to determine that the tested video device has no fault when the fault reflected by the video data is consistent with the actual fault and the preset video control parameters are consistent with the parameters of the video data; otherwise, it is determined that the tested video device has a fault.
[0024] Furthermore, the device also includes: a test result output module,
[0025] The test result output module is used to output the judgment result of the recognition result comparison module.
[0026] In a third aspect, an airborne system automated testing method is provided, which is used in any system described in the first aspect, and the method comprises:
[0027] The airborne system automatic test device generates fault information of the airborne system and sends the fault information to the video device under test and the audio device under test;
[0028] The video device under test displays the fault information on the screen; the audio device under test plays the fault information in voice;
[0029] The video collector collects the video data displayed by the video device under test and sends it to the airborne system automatic test device;
[0030] The pickup collects the audio data output by the audio device under test and sends it to the airborne system automated test device;
[0031] The airborne system automated test device identifies and compares the video data and audio data to determine whether the tested video device and the tested audio device are faulty.
[0032] The airborne system automatic test device in the present invention generates fault information of the airborne system, and sends the fault information to the video device under test and the audio device under test; the video device under test displays the fault information in a screen; the audio device under test plays the fault information in a voice; the video collector collects the video data displayed by the video device under test, and sends it to the airborne system automatic test device; the microphone collects the audio data output by the audio device under test, and sends it to the airborne system automatic test device; the airborne system automatic test device identifies and compares the video data and the audio data, and determines whether the video device under test and the audio device under test have faults; the problems existing in the audio and video system during the avionics joint test can be quickly and accurately discovered, and the airborne system automatic test can be realized. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 A schematic diagram of an airborne system automated testing system provided by an embodiment of the present invention;
[0034] Figure 2 A speech recognition flow chart provided by an embodiment of the present invention;
[0035] Figure 3 A flow chart of speech content recognition provided by an embodiment of the present invention;
[0036] Figure 4 A video recognition flow chart provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0037] The present application is further described in detail below through specific implementation methods and drawings.
[0038] An embodiment of the present invention provides an airborne system automated testing system, such as Figure 1 As shown, the system includes: an airborne system automatic test device, a sound pickup, a video collector, an audio device under test, and a video device under test;
[0039] The airborne system automated test device is used to generate fault information of the airborne system and send the fault information to the video device under test and the audio device under test;
[0040] The video device under test is used to display the fault information on the screen; the audio device under test is used to play the fault information in voice;
[0041] The video collector is used to collect the video data displayed by the video device under test and send it to the airborne system automatic test device;
[0042] The pickup is used to collect audio data output by the audio device under test and send it to the airborne system automated test device;
[0043] The airborne system automated test device is used to identify and compare video data and audio data to determine whether the tested video equipment and tested audio equipment are faulty.
[0044] Among them, the airborne system automated testing device is used to determine that the audio device under test has no fault when the fault reflected by the audio data is consistent with the actual fault and the preset audio control parameters are consistent with the parameters of the audio data; otherwise, it is determined that the audio device under test has a fault.
[0045] Among them, the airborne system automatic testing device is used to determine that the video device under test has no fault when the fault reflected by the video data is consistent with the actual fault and the preset video control parameters are consistent with the parameters of the video data; otherwise, it is determined that the video device under test has a fault.
[0046] Optionally, the audio control parameters include: audio delay, frequency, length, and number of repetitions.
[0047] Optionally, the video control parameters include: duration, dynamic display frequency, and number of repetitions.
[0048] An embodiment of the present invention provides an automatic test device for an airborne system, such as Figure 1 As shown, the device includes: a test case writing and execution module, a voice actuator, a video actuator, a voice recognition module, a video recognition module, and a recognition result comparison module.
[0049] The test case writing and execution module is used to generate fault information of the airborne system and send the fault information to the voice actuator and the video actuator; the voice actuator sends the fault information to the audio device under test according to the preset communication protocol requirements, and the video actuator sends the fault information to the video device under test according to the preset communication protocol requirements;
[0050] The video recognition module is used to receive the video data collected by the video collector and send it to the recognition result comparison module;
[0051] The speech recognition module is used to receive the audio data collected by the microphone and send it to the recognition result comparison module;
[0052] The recognition result comparison module is used to recognize and compare the video data and the audio data to determine whether the tested video device and the tested audio device are faulty.
[0053] Further, in another embodiment, the airborne system automated test device further comprises: a speech analysis module, the speech analysis module being used to determine parameters of the audio data collected by the microphone;
[0054] The recognition result comparison module is used to determine that the audio device under test has no fault when the fault reflected by the audio data is consistent with the actual fault and the preset audio control parameters are consistent with the parameters of the audio data; otherwise, it is determined that the audio device under test has a fault.
[0055] In the embodiment of the present invention, the parameters of the audio data collected by the microphone may be determined by the speech analysis module, the recognition result comparison module, or the speech recognition module, and this embodiment does not limit this.
[0056] The recognition result comparison module is also used to determine that the video device under test has no fault when the fault reflected by the video data is consistent with the actual fault and the preset video control parameters are consistent with the parameters of the video data; otherwise, it is determined that the video device under test has a fault.
[0057] Furthermore, in yet another embodiment, the airborne system automated testing device further includes: a test result output module, configured to output the determination result of the recognition result comparison module.
[0058] Specifically, see Figure 1 An airborne system automated test device provided by an embodiment of the present invention includes: a test case writing and execution module, a voice actuator, a video actuator, a voice recognition module, a video recognition module, and a recognition result comparison module.
[0059] 1. The test case writing and execution module provides testers with the function of writing and executing test scripts. The test process is represented by test cases for test tasks, and the relevant audio and video parameters are sent to the voice actuator and video actuator during the execution of the test cases.
[0060] 2. The voice exciter and the video exciter respectively receive the parameters issued during the execution of the test case, and perform data excitation on the audio device and the video device under test, thereby generating audio and video signals;
[0061] 3. The sound pickup obtains audio signals by collecting data from the audio device under test, and the video collector obtains video signals by collecting data from the video device under test;
[0062] 4. The speech recognition module recognizes the audio data collected by the microphone and obtains the speech content (i.e. the fault reflected by the audio data), which is a text sequence, such as Figure 2 As shown, the specific steps include:
[0063] Step 1: First, perform a pre-processing operation of silence removal at the beginning and end of a speech signal collected by the microphone to reduce the interference to subsequent speech recognition. Then, segmented filtering is used for processing. Finally, acoustic feature extraction is performed through Fourier transform to obtain an acoustic feature vector X, which is expressed as:
[0064] X=[x1,x2,x3,…], where xi represents a feature vector of an audio frame (Frame).
[0065] Step 2: Construct the candidate text sequence W corresponding to the acoustic feature vector X. W is expressed as:
[0066] W = [w1, w2, w3, ...], where wi represents the text corresponding to the feature vector of the audio frame.
[0067] Step 3: The test case writing and execution module continuously stimulates the audio device under test to generate audio signals. The speech recognition module obtains acoustic feature vectors through preprocessing. After the above repeated operations, a speech database is constructed. The speech database is composed of basic components of speech and text sequences.
[0068] Step 4: Construct an audio recognition model through the attention mechanism unit, CTC unit, and hidden Markov model DNN-HMM unit, such as Figure 3 As shown in the figure, the attention mechanism unit is used to complete the mapping of the entire speech to the entire sentence text sequence, the CTC unit is used to complete the mapping of the pronunciation range of each word or each phoneme to the corresponding word and phonetic symbol, and the DNN-HMM unit is used to expand the speech modeling scale from a single phoneme length to a three-phoneme pronunciation range based on the context range.
[0069] Step 5: Train and verify the audio recognition model constructed above, where the training input is the acoustic feature vector X in the speech database, and the output is the text sequence W corresponding to the acoustic feature vector X.
[0070] Step 6: Write a new test case through the test case writing and execution module, and execute the test case. Repeat the operation in step 1, input the obtained acoustic feature vector x into the speech recognition model to obtain the text sequence w.
[0071] 5. The video recognition module recognizes and processes the video data collected by the video collector, and obtains Chinese characters, English characters and icon contents of the UI interface, such as Figure 4 As shown, the specific steps include:
[0072] Step 1: The test case writing and execution module continuously stimulates the video device under test to generate video signals. The video recognition module saves the collected videos, builds a video database, and selects characters and icons in the video frames to obtain labels.
[0073] Step 2: Use deep learning methods to build a target detection model, and send the constructed video database and labels into the target detection model for training and verification. The training input is each frame of the video database, and the output is the position and text corresponding to the characters and icons.
[0074] Step 3: Write a new test case through the test case writing and execution module, and execute the test case. Repeat the operation in step 1, input the obtained video frame m into the video recognition model to obtain the text t.
[0075] Step 4: Process the videos acquired within a period of time to obtain the duration, frequency, etc. of the icon performing related actions (flashing, etc.).
[0076] 6. The recognition result comparison module can determine whether the content of the alarm or prompt tone of the tested audio device is consistent with the benchmark content in the test case, whether the content displayed on the UI interface of the tested video device is consistent with the benchmark content in the test case, and whether the content of the alarm or prompt tone of the tested audio device is consistent with the content displayed on the UI interface of the tested video device.
[0077] 7. The speech analysis module analyzes the delay, frequency, length, and number of repetitions of the audio data collected by the microphone.
[0078] 8. The test result output module comprehensively analyzes the identification and comparison results and displays the test results in the form of charts.
[0079] An embodiment of the present invention also provides an audio recognition method, which adopts a deep learning algorithm to effectively recognize speech content by integrating an attention mechanism, a CTC model and a DNN-HMM model; a video recognition model is provided, which recognizes text, icons and video element actions (such as flashing) by adopting a deep learning method; the application scenarios involved can be extended to other fields, thereby realizing the recognition, alarm and automated testing of other parameters.
[0080] An embodiment of the present invention further provides an airborne system automated testing method, which is used in the system described in the embodiment of the present invention. The method comprises the following steps:
[0081] Step 1: The automated test device generates fault information of the airborne system and sends the fault information to the video device under test and the audio device under test;
[0082] Step 2: The video device under test displays the fault information on the screen; the audio device under test plays the fault information in voice;
[0083] Step 3: The video collector collects the video data displayed by the video device under test and sends it to the automated test device;
[0084] Step 4: The microphone collects the audio data output by the audio device under test and sends it to the automated test device;
[0085] Step 5: The automated testing device identifies and compares the video data and the audio data to determine whether the tested video device and the tested audio device are faulty.
[0086] The present invention can quickly and accurately discover problems existing in the audio and video system during the avionics joint test, provide sound recognition and analysis tools, can receive equipment prompt sounds such as alarm sounds, and analyze the delay, frequency, length, and number of repetitions of the prompt sounds; and provide a video recognition function, combined with the complete logic of the equipment alarm sound, to determine whether it meets the equipment function requirements according to the test requirements.
[0087] The above only expresses the implementation methods of the present application, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present application, which all belong to the protection scope of the present application. In addition, the parts that are not detailed in this application are all conventional technologies.
Claims
1. An airborne system automated test system, It is characterized in that include: Airborne system automated test equipment, sound pickup, video acquisition device, audio equipment under test and video equipment under test; The airborne system automated testing device is used to generate fault information of the airborne system and send the fault information to the video device under test and the audio device under test; The video device under test is used to display the fault information on a screen; the audio device under test is used to play the fault information in voice; The video collector is used to collect video data displayed by the video device under test and send it to the airborne system automatic test device; The pickup is used to collect audio data output by the audio device under test and send it to the airborne system automatic test device; The airborne system automatic test device is used to identify and compare the received video data and audio data, and determine whether the tested video device and the tested audio device are faulty.
2. The system according to claim 1, It is characterized in that The airborne system automated testing device is used to determine that the audio device under test has not failed when the failure reflected by the audio data is consistent with the actual failure and the parameters of the audio data are consistent with the preset audio control parameters; otherwise, it is determined that the audio device under test has failed.
3. The system according to claim 1, It is characterized in that The airborne system automated testing device is used to determine that the video device under test has no fault when the fault reflected by the video data is consistent with the actual fault and the parameters of the video data are consistent with the preset video control parameters; otherwise, it is determined that the video device under test has a fault.
4. The system according to claim 2, It is characterized in that The audio control parameters include: audio delay, frequency, length, and number of repetitions.
5. The system according to claim 3, It is characterized in that The video control parameters include: duration, dynamic display frequency, and number of repetitions.
6. An automated test device for airborne systems, It is characterized in that include: Test case writing and execution module, voice actuator, video actuator, voice recognition module, video recognition module, recognition result comparison module, The test case writing and execution module is used to generate fault information of the airborne system and send the fault information to the voice actuator and the video actuator; the voice actuator is used to send the fault information to the audio device under test according to the preset communication protocol requirements; the video actuator is used to send the fault information to the video device under test according to the preset communication protocol requirements; The video recognition module is used to receive the video data collected by the video collector and send it to the recognition result comparison module; The speech recognition module is used to receive the audio data collected by the microphone and send it to the recognition result comparison module; The recognition result comparison module is used to recognize and compare the video data and the audio data to determine whether the tested video device and the tested audio device are faulty.
7. The device according to claim 6, It is characterized in that The device further comprises: a speech analysis module, the speech analysis module being used to determine parameters of the audio data collected by the microphone; The recognition result comparison module is specifically used to determine that the audio device under test has no fault when the fault reflected by the audio data is consistent with the actual fault and the parameters of the audio data are consistent with the preset audio control parameters; otherwise, determine that the audio device under test has a fault.
8. The device according to claim 6, It is characterized in that The recognition result comparison module is specifically used to determine that the video device under test has no fault when the fault reflected by the video data is consistent with the actual fault and the parameters of the video data are consistent with the preset video control parameters; otherwise, it is determined that the video device under test has a fault.
9. The device according to claim 6, It is characterized in that The device also includes: a test result output module, The test result output module is used to output the determination result of the recognition result comparison module.
10. An automated testing method for an airborne system, It is characterized in that For use in the system according to any one of claims 1 to 5, the method comprises: The airborne system automatic test device generates fault information of the airborne system and sends the fault information to the video device under test and the audio device under test; The video device under test displays the fault information on a screen; the audio device under test plays the fault information in voice; The video collector collects the video data displayed by the video device under test and sends it to the airborne system automatic test device; The pickup collects the audio data output by the audio device under test and sends it to the airborne system automatic test device; The airborne system automated testing device identifies and compares the video data and the audio data to determine whether the tested video device and the tested audio device are faulty.
Citation Information
Cited By
Automatic troubleshooting system for fighter avionics joint test based on machine learning
CN119781361A
An automatic troubleshooting system for fighter aircraft avionics joint testing based on machine learning
CN119781361B
