Vehicle active noise cancellation system diagnostic tool

Abstract

Embodiments of a diagnostic tool for an active noise cancellation system are presented. The diagnostic tool may generate an output from the active noise cancellation system and capture an input to the active noise cancellation system resulting from the output of the active noise cancellation system. The diagnostic tool may generate a visual output to assist in evaluating the active noise cancellation system.

Description

Technical Field

[0001] The present disclosure relates to a diagnostic tool for evaluating a vehicle active noise cancellation system. Background Art

[0002] An active noise cancellation system for a vehicle may generate noise that cancels noise within a passenger compartment of the vehicle. The active noise cancellation system may sense noise within the passenger compartment of the vehicle via one or more microphones. The active noise cancellation system may respond to the sensed noise by generating noise via one or more speakers that cancels the noise sensed via the one or more microphones. The result is a significant reduction in noise heard by vehicle occupants within the passenger compartment of the vehicle. More complex active noise cancellation systems may include multiple microphones, multiple speakers, and multiple accelerometers to reduce the perceived noise within the passenger compartment of the vehicle.

[0003] Active noise cancellation systems may include multiple components that, when operated as desired, can provide a significant level of noise reduction for a vehicle passenger compartment. However, during vehicle manufacturing, one or more of the active noise cancellation system components may be degraded. For example, the polarity of a wire connected to a speaker may be reversed, or a vehicle component may block a microphone or speaker. If one of the active noise cancellation system components experiences degradation, it may be difficult to determine which component is degraded when the vehicle is manufactured or when the vehicle is in use, because the active noise cancellation system may still remain partially active, and because some forms of component degradation may not be obvious to some vehicle occupants. In addition, due to differences between the active noise cancellation system components and the internal construction of the vehicle, the response of the active noise cancellation system of one vehicle may be slightly different from the response of the active noise cancellation system of the second vehicle. In addition, it may be desirable that the active noise cancellation system diagnostics performed at the end of the vehicle assembly line be reliable (e.g., fewer false positive or false negative indications of active noise cancellation system degradation) and performed quickly, because inaccurate diagnostic evaluations and diagnostics that require a lot of time to perform may create excessive problems, which may include, but are not limited to, generating a large number of rework prompts and slowing down the vehicle assembly line. Summary of the invention

[0004] The inventors herein have recognized that it may be difficult to apply time-based signal processing techniques to provide reliable operational assessments of multiple-input multiple-output systems. In particular, manufacturing tolerances between vehicles and slight variations in cabin acoustics may make it difficult to establish metrics (e.g., performance thresholds) that can be applied to a large number of units under test. Therefore, it may be desirable to provide a method of diagnosing an active noise cancellation system to more reliably indicate the presence or absence of active noise cancellation system degradation.

[0005] By applying a cross-correlation between a reference impulse response and an impulse response of a particular active noise cancellation system, a threshold value can be generated that makes the active noise cancellation more reliable. Alternatively or additionally, a correlation coefficient can be generated based on the reference impulse response and the impulse response of a particular vehicle to determine whether the active noise cancellation system of the particular vehicle is operating as expected. In addition, the cross-correlation technique can be quickly performed on a system including multiple input terminals and output terminals. The present disclosure provides a diagnostic tool for an active noise cancellation system, including: a controller, the controller including executable instructions stored in a non-transitory memory, the executable instructions causing the controller to initiate an impulse response diagnostic for the active noise cancellation system, wherein the impulse response diagnostic includes performing a cross-correlation between a reference impulse response and an impulse response of the active noise cancellation system.

[0006] In a first example, the diagnostic tool includes wherein the impulse response diagnostic includes supplying a broadband signal to one or more speakers of the active noise cancellation system and generating sound through the one or more speakers in response to the broadband signal. In a second example that may include the first example, the sound is sensed by one or more microphones of the active noise cancellation system. In a third example that may include one or both of the first and second examples, the diagnostic tool also includes generating a correlation coefficient based on a reference impulse response and an impulse response of the active noise cancellation system. In a fourth example that may include one or more of the first to third examples, the reference impulse response is based on a plurality of impulse responses determined from an output device and an input device. In a fifth example that may include one or more of the first to fourth examples, the output device is a speaker, and wherein the input device is a microphone or an accelerometer. In a sixth example that may include one or more of the first to fifth examples, the diagnostic tool also includes generating an indication of degradation of the active noise cancellation system based on the value of the correlation coefficient. In a seventh example that may include one or more of the first to sixth examples, the degradation indication is an indication that the speaker wiring has reversed electrical polarity.

[0007] Active noise cancellation systems can be evaluated by driving the vehicle and observing whether the passenger compartment is quiet. However, such evaluations can be subjective and can be time consuming. Additionally, such evaluations can make it difficult to determine whether the noise in the passenger compartment of one vehicle is less or more than the noise in the passenger compartment of a similar vehicle. The present disclosure overcomes the limitations of methods that can be more subjective and evaluates the operation of the system under test relative to one or more reference vehicles so that a group of vehicles can perform similarly.

[0008] As an example, the present disclosure provides an active noise cancellation system diagnostic method, including: supplying a broadband signal to a speaker of the active noise cancellation system; sensing an audible output of the speaker in response to the broadband signal through a microphone; and performing a cross-correlation analysis on an impulse response generated from the signal output of the microphone sensing the audible output of the speaker. In this way, the crossover frequency of the boost circuit can be adjusted during a boost circuit condition of zero frequency change, so that the boost circuit response can be improved.

[0009] In a first example, the impulse response is generated by dividing the signal output from the microphone by the broadband signal, and wherein the active noise cancellation system is performed at the end of a vehicle assembly line after the speaker and the microphone are installed in the vehicle. In a second example, which may include the first example, the method further includes: wherein performing the cross-correlation analysis includes performing cross-correlation based on a reference impulse response and the impulse response generated from the signal output from the microphone. In a third example, which may include one or both of the first and second examples, the method further includes generating a correlation coefficient based on a reference impulse response and an impulse response generated from the signal output from the microphone. In a fourth example, which may include one or more of the first to third examples, the method includes wherein the reference impulse response is an average impulse response of multiple active noise cancellation systems of multiple vehicles. In a fifth example, which may include one or more of the first to fourth examples, the method further includes displaying the correlation coefficient through a human-machine interface. In a sixth example, which may include one or more of the first to fifth examples, the method further includes displaying the correlation coefficient and multiple other correlation coefficients from speakers other than the speaker in a polygonal shape through a human-machine interface.

[0010] Components of an active noise cancellation system may be tested prior to assembly; however, testing the active noise cancellation system in a vehicle may provide actual proof of operation of the entire system and its individual components. Even so, evaluating an active noise cancellation system at the end of an assembly line may be time-constrained, and the testing environment may affect the evaluation. The present disclosure provides a system that can fully evaluate an active noise cancellation system, including all of its sensors and actuators, in a short time frame. In addition, the method can utilize a single threshold, thereby reducing or avoiding time-consuming interpolation and searching for generating a degradation / no degradation classification. Therefore, the system and method of the present disclosure can meet the standards for end-of-line testing.

[0011] In one or more examples, the present description provides an end-of-assembly-line diagnostic tool for an active noise cancellation system, comprising: a controller including executable instructions stored in a non-transitory memory of a vehicle being assembled on the assembly line, the executable instructions causing the controller to initiate an impulse response diagnostic for the active noise cancellation system when the vehicle is at the end of the assembly line, wherein the impulse response diagnostic includes generating correlation coefficients for a plurality of microphones and generating output on a display (e.g., a vehicle display or a remote display) in response to the generated correlation coefficients.

[0012] In a first example, the end-of-line diagnostic tool includes wherein the correlation coefficients of the plurality of microphones include the correlation coefficients of each microphone-speaker pair. In a second example, which may include the first example, the end-of-line diagnostic tool includes wherein the impulse response diagnostic includes performing a cross-correlation analysis, the cross-correlation analysis including performing a cross-correlation between an average impulse response of a plurality of active noise cancellation systems of a plurality of vehicles and an impulse response of the active noise cancellation system. In a third example, which may include one or both of the first and second examples, the end-of-line diagnostic tool also includes determining a minimum cross-correlation between the impulse responses of a plurality of active noise cancellation systems of a plurality of vehicles, and adding a margin to the minimum cross-correlation between the impulse responses of the plurality of active noise cancellation systems. In a fourth example, which may include one or more of the first to third examples, the end-of-line diagnostic tool also includes: in response to a cross-correlation of an average impulse response of the plurality of active noise cancellation systems and the impulse response of the active noise cancellation system being less than the margin plus the minimum cross-correlation between the impulse responses of the plurality of active noise cancellation systems, indicating degradation of the active noise cancellation system.

[0013] The present disclosure also provides an active noise cancellation system diagnostic method, including: supplying a broadband signal to a speaker of the active noise cancellation system; sensing the vibration output of the speaker in response to the broadband signal through an accelerometer; and performing cross-correlation analysis on an impulse response generated from a signal output from the accelerometer that senses the vibration output of the speaker. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present disclosure may be better understood by reading the following description of non-limiting embodiments with reference to the accompanying drawings, in which:

[0015] Figure 1 showing a plan view of a vehicle at the end of an assembly line including an active noise cancellation system;

[0016] Figure 2 Example impulse response plots for different microphone-loudspeaker pairs are shown;

[0017] Figure 3A graph showing an example wideband signal for evaluating the impulse response of an active noise cancellation system;

[0018] Figure 4 An example frequency domain plot of an impulse response is shown;

[0019] Figure 5 A graphical representation of a first frequency domain technique for evaluating an active noise cancellation system is shown;

[0020] Figure 6 shows a graphical representation of a second frequency domain technique for evaluating active noise cancellation systems;

[0021] Figure 7 is a graph illustrating different example signal types that may be evaluated by applying cross-correlation;

[0022] Figure 8-Figure 11 shows an example of how the correlation analysis can be visualized for a diagnostic tool operator;

[0023] Fig.12 A diagram illustrating how a vehicle accelerometer may be evaluated is shown;

[0024] Figure 13-Figure 16 Features of a vehicle active noise cancellation system diagnostic tool are shown; and

[0025] Fig.17 A method for evaluating operation of an active noise cancellation system of a vehicle is shown. DETAILED DESCRIPTION

[0026] The present disclosure provides systems and methods for solving the above-mentioned problems that may arise when assembling an active noise cancellation system for a vehicle. The methods and systems described herein may be implemented in a handheld tool (e.g., a notebook computer, a notepad, a pendant, etc.) or in an active noise cancellation system. In one example, the method may stimulate the output of a speaker in a passenger compartment of a vehicle, and a microphone in the passenger compartment may respond to the output of the speaker. Each speaker of the active noise cancellation system may be stimulated with a broadband impulse signal to determine how the active noise cancellation system responds to the stimulus. The response of the active noise cancellation system may be evaluated relative to evaluations of similar systems to determine whether the active noise cancellation system currently being evaluated meets the operating specifications.

[0027] Figure 1A vehicle 100 is shown at the end of a vehicle assembly line 101. The vehicle 101 includes an active noise cancellation system 15, which includes a speaker 110, a microphone 108, an accelerometer 106, and an embedded controller 102. The embedded controller 102 includes a controller 102a (e.g., a central processing unit (CPU), a digital signal processor (DSP), etc.), a memory 102b (e.g., a read-only memory, a random access memory, etc.), and an input / output terminal 102c (e.g., a digital input terminal, a digital output terminal, an analog input terminal, an analog output terminal). The embedded controller 102 can display information (e.g., data, messages, operating parameters, etc.) and receive user input through a display 130. The diagnostic tool described herein can be part of the embedded controller 102, or it can be a remote device 104 that is not part of the vehicle 100. The remote device 104 may include a controller 104a (e.g., a central processing unit (CPU), a memory 104b (e.g., a read-only memory, a random access memory, etc.), and input / output terminals 104c (e.g., a digital input terminal, a digital output terminal, an analog input terminal, an analog output terminal). The remote device 104 may display information (e.g., data, messages, programs, applications, and vehicle operating parameters) via a display 131. The remote device 104 may communicate with the embedded controller 102 via a network link 152 to initiate diagnostic evaluations, receive data from sensors and actuators coupled to the embedded controller 102, upload and download applications / programs, etc.

[0028] Reference now Figure 2 , impulse response graphs of different microphone-loudspeaker pairs for multiple vehicles are shown. Each of the four graphs includes a vertical axis representing the loudspeaker output and a horizontal axis representing time. Each graph contains the trajectories of six different vehicles of the same type (e.g., vehicle 1 (201); vehicle 2 (202); vehicle 3 (203); vehicle 4 (204); vehicle 5 (205); vehicle 6 (206)). The impulse responses of the six vehicles are represented by different line types.

[0029] Figure 2 The upper left graph of is a graph of the impulse response of a first speaker-microphone pair (eg, speaker #1 and microphone #1). Figure 2 The lower left graph of is a graph of the impulse response of a second speaker-microphone pair (eg, speaker #2 and microphone #1). Figure 2 The upper right graph of is a graph of the impulse response of a third speaker-microphone pair (eg, speaker 1 and microphone #2). Figure 2 The lower right graph of is the impulse response of the fourth speaker-microphone pair (eg, speaker #2 and microphone #2).

[0030] It can be observed that no two impulse responses are identical. In particular, the traces include amplitude variations and some phase variations. Such variations may make it difficult to apply time domain analysis with thresholds (e.g., amplitude thresholds) to determine degraded or non-degraded operation of a particular speaker-microphone pair. An impulse response plot may be generated by dividing the output of a particular microphone by the input of a particular speaker over a defined time frame during which the speaker input and microphone output are sampled, measured, and stored to memory. The speaker input may be, for example, Figure 3 The broadband impulse is shown.

[0031] Reference now Figure 3 , a diagram of a broadband signal for evaluating an impulse response of an active noise cancellation system of a vehicle is shown. In this example, the broadband signal 302 is a sine wave of varying frequency. The sine wave starts at a first predetermined frequency and ends at a second predetermined frequency, the second predetermined frequency being greater than the first predetermined frequency. The signal 302 may be applied to a speaker of an active noise cancellation system to stimulate the active noise cancellation system. The microphone of the active noise cancellation system may respond to the sound produced by the speaker driven by the broadband impulse signal.

[0032] Reference now Figure 4 , frequency domain impulse response graphs of different microphone-loudspeaker pairs of multiple vehicles are shown. Each of the four graphs includes a vertical axis representing the amplitude of the loudspeaker output and a horizontal axis representing the frequency. Each graph contains the trajectories of six different vehicles of the same type (e.g., vehicle 1 (401); vehicle 2 (402); vehicle 3 (403); vehicle 4 (404); vehicle 5 (405); vehicle 6 (406)). The frequency domain impulse responses of the six vehicles are indicated by different line types.

[0033] Figure 4 The upper left graph of is a graph of the frequency domain impulse response of a first speaker-microphone pair (eg, speaker #1 and microphone #1). Figure 4 The lower left graph of is a graph of the frequency domain impulse response of a second speaker-microphone pair (eg, speaker #2 and microphone #1). Figure 4 The upper right graph of is a graph of the frequency domain impulse response of a third speaker-microphone pair (eg, speaker 1 and microphone #2). Figure 4 The lower right graph of is the frequency domain impulse response of the fourth speaker-microphone pair (eg, speaker #2 and microphone #2).

[0034] A frequency domain impulse response plot may be generated by providing a broadband impulse signal to a speaker and applying a Fourier transform (e.g., a fast Fourier transform) to the result of dividing the output of a particular microphone by the input of a particular speaker within a defined time frame during which the speaker input and microphone output are sampled, measured, and stored to a memory. In one example, a discrete Fourier transform (DFT) may be expressed as:

[0035]

[0036] Where F is a series of complex numbers as a function of the imaginary number j, ω is the angular velocity, f is the function or vector to be transformed into the frequency domain, e is the Euler number, k is the number of samples, and T is the sampling period. The DFT can be modified to generate a fast Fourier transform, which reduces the computation time of the active noise cancellation system signal.

[0037] Reference now Figure 5 , shows a first method of frequency domain impulse response analysis. The figure shows how a threshold value for evaluating the operation of an active noise cancellation system may be determined. Figure 5 Plots of different microphone-speaker pairs for a plurality of vehicles are included. Each of the four plots includes a vertical axis representing the amplitude of the speaker output and a horizontal axis representing the frequency. Each plot contains traces for six different vehicles of the same type (e.g., vehicle 1 (501); vehicle 2 (502); vehicle 3 (503); vehicle 4 (504); vehicle 5 (505); vehicle 6 (506)), minimum (508) and maximum (507) amplitude traces for the six vehicles, and maximum (509) and minimum (510) thresholds for determining a degraded / non-degraded classification of the speaker-microphone pair. The frequency domain impulse responses of the six vehicles are indicated with different line types, and the six vehicles serve as reference vehicles. Note that there may be more than six reference vehicles or less than six reference vehicles.

[0038] Figure 5 The upper left graph of is a graph of the frequency domain impulse response of a first speaker-microphone pair (eg, speaker #1 and microphone #1). Figure 5 The lower left graph of is a graph of the frequency domain impulse response of a second speaker-microphone pair (eg, speaker #2 and microphone #1). Figure 5 The upper right graph of is a graph of the frequency domain impulse response of a third speaker-microphone pair (eg, speaker 1 and microphone #2). Figure 5The lower right plot is the frequency domain impulse response of a fourth speaker-microphone pair (e.g., speaker #2 and microphone #2). Each of the plots shows the frequency domain impulse response of six reference vehicles (e.g., vehicles that serve as a basis for determining whether the current vehicle being evaluated for active noise cancellation system operation is operating within acceptable parameters or specifications). The six reference vehicle frequency domain impulse responses provide an indication of the distribution of impulse responses for a group of vehicles.

[0039] A frequency domain impulse response plot may be generated by providing a broadband impulse signal to a speaker and applying a Fourier transform (e.g., a Fast Fourier Transform) to the result of dividing the output of a particular microphone by the input of a particular speaker within a defined time frame during which the speaker input and microphone output are sampled, measured, and stored to memory.

[0040] A minimum amplitude curve or value may be generated by selecting the minimum amplitude value for each frequency from each of the speaker-microphone amplitude-frequency pairs. For example, if there are three reference vehicles, and vehicle #1 has an amplitude of -10 dB at 100 Hz, vehicle #2 has an amplitude of -11 dB at 100 Hz, and vehicle #3 has an amplitude of -9 dB, then the value of the minimum amplitude is -11 dB at 100 Hz, because -11 is the minimum value of the three vehicles at 100 Hz. If vehicle #1 has an amplitude of 0 dB at 150 Hz, vehicle #2 has an amplitude of -1 dB at 150 Hz, and vehicle #3 has an amplitude of -3 dB, then the value of the minimum amplitude is -3 dB at 150 Hz, because -3 is the minimum value of the three vehicles at 150 Hz. The values ​​of the minimum amplitude curve at all frequencies may be determined in a similar manner.

[0041] A maximum amplitude curve or value may be generated by selecting the maximum amplitude value for each frequency from each of the speaker-microphone amplitude-frequency pairs. For example, if there are three reference vehicles, and vehicle #1 has an amplitude of -10 dB at 100 Hz, vehicle #2 has an amplitude of -11 dB at 100 Hz, and vehicle #3 has an amplitude of -9 dB, then the maximum amplitude curve is -9 dB at 100 Hz, since -9 is the maximum value of the three vehicles at 100 Hz. If vehicle #1 has an amplitude of 0 dB at 150 Hz, vehicle #2 has an amplitude of -1 dB at 150 Hz, and vehicle #3 has an amplitude of -3 dB, then the value of the maximum amplitude is 0 dB at 150 Hz, since 0 is the maximum value of the three vehicles at 150 Hz. The values ​​of the maximum amplitude curve at all frequencies may be determined in a similar manner.

[0042] The threshold curve for determining whether the impulse response of the current vehicle under test is correct can be determined by adding a positive margin value (e.g., 0.5 dB) to the value of the maximum curve and a negative margin value (e.g., -0.5 dB) to the value of the minimum curve. Figure 5 As shown, the threshold maximum curve for a particular speaker-microphone pair is a margin larger than the maximum curve for the speaker-microphone pair. The threshold minimum curve for a particular speaker-microphone pair is a margin smaller than the minimum curve for the speaker-microphone pair.

[0043] The impulse response in the frequency domain of the current active noise cancellation system speaker-microphone pair may be compared to a threshold minimum curve and a threshold maximum curve generated from a reference vehicle for the same speaker-microphone pair. If all values ​​in the impulse response in the frequency domain of the current active noise cancellation system speaker-microphone pair (e.g., speaker #1, microphone #2) are between the threshold maximum curve and the threshold minimum curve, it may be determined that the speaker-microphone pair of the current system is not degraded. If one or more values ​​in the impulse response in the frequency domain of the current active noise cancellation system speaker-microphone pair (e.g., speaker #1, microphone #2) are outside the threshold maximum curve and the threshold minimum curve, it may be determined that the speaker-microphone pair of the current system is degraded.

[0044] Reference now Figure 6 , shows a second method of frequency domain impulse response analysis. This figure shows how a threshold value for evaluating the operation of an active noise cancellation system may be determined. Figure 6 Plots of different microphone-speaker pairs for a plurality of vehicles are included. Each of the four plots includes a vertical axis representing the amplitude of the speaker output and a horizontal axis representing the frequency. Each plot contains traces for six different vehicles of the same type (e.g., vehicle 1 (601); vehicle 2 (602); vehicle 3 (603); vehicle 4 (604); vehicle 5 (605); vehicle 6 (606)), a mean curve (607) amplitude trace for the six vehicles, and maximum (608) and minimum (609) thresholds for determining a degraded / non-degraded classification for the speaker-microphone pair. The frequency domain impulse responses for the six vehicles are indicated with different line types, and the six vehicles serve as reference vehicles. Note that there may be more than six reference vehicles or less than six reference vehicles.

[0045] Figure 6 The upper left graph of is a graph of the frequency domain impulse response of a first speaker-microphone pair (eg, speaker #1 and microphone #1). Figure 6 The lower left graph of is a graph of the frequency domain impulse response of a second speaker-microphone pair (eg, speaker #2 and microphone #1). Figure 6The upper right graph of is a graph of the frequency domain impulse response of a third speaker-microphone pair (eg, speaker 1 and microphone #2). Figure 6 The lower right graph of is the frequency domain impulse response of the fourth speaker-microphone pair (e.g., speaker #2 and microphone #2). Each of the graphs shows the frequency domain impulse response of six reference vehicles. The six reference vehicle frequency domain impulse responses provide an indication of the impulse response distribution of a group of vehicles.

[0046] A frequency domain impulse response plot may be generated by providing a broadband impulse signal to a speaker and applying a Fourier transform (e.g., a Fast Fourier Transform) to the result of dividing the output of a particular microphone by the input of a particular speaker within a defined time frame during which the speaker input and microphone output are sampled, measured, and stored to memory.

[0047] A mean amplitude curve or value may be generated by determining the mean amplitude value for each frequency from each of the speaker-microphone amplitude-frequency pairs. For example, if there are three reference vehicles, and vehicle #1 has an amplitude of -10 dB at 100 Hz, vehicle #2 has an amplitude of -11 dB at 100 Hz, and vehicle #3 has an amplitude of -9 dB, then the value of the mean amplitude is -10 dB at 100 Hz, because -10 is the mean of the three vehicles at 100 Hz. If vehicle #1 has an amplitude of 0 dB at 150 Hz, vehicle #2 has an amplitude of -1 dB at 150 Hz, and vehicle #3 has an amplitude of -3 dB, then the value of the mean amplitude is -1.3 dB at 150 Hz, because -1.3 dB is the minimum value of the three vehicles at 150 Hz. The values ​​of the minimum amplitude curve at all frequencies may be determined in a similar manner.

[0048] The threshold curve for determining whether the impulse response of the current vehicle under test is correct can be determined by adding a positive margin value (e.g., 0.5 dB) to the value of the mean curve to generate a maximum threshold curve and adding a negative margin value (e.g., -0.5 dB) to the value of the mean curve to generate a minimum threshold curve. Figure 6 As shown, the threshold maximum curve for a particular speaker-microphone pair is a margin larger than the mean curve for the speaker-microphone pair. The threshold minimum curve for a particular speaker-microphone pair is a margin smaller than the mean curve for the speaker-microphone pair.

[0049] The impulse response in the frequency domain of the current active noise cancellation system speaker-microphone pair may be compared to a threshold minimum curve and a threshold maximum curve generated from a reference vehicle for the same speaker-microphone pair. If all values ​​in the impulse response in the frequency domain of the current active noise cancellation system speaker-microphone pair (e.g., speaker #1, microphone #2) are between the threshold maximum curve and the threshold minimum curve, it may be determined that the speaker-microphone pair of the current system is not degraded. If one or more values ​​in the impulse response in the frequency domain of the current active noise cancellation system speaker-microphone pair (e.g., speaker #1, microphone #2) are outside the threshold maximum curve and the threshold minimum curve, it may be determined that the speaker-microphone pair of the current system is degraded.

[0050] Now go to Figure 7 , a diagram showing different signals that can be evaluated by cross-correlation. Figure 7 Active noise cancellation system conditions that may be indicated by cross-correlation and / or correlation coefficients are also described. Graph 700 shows three curves. Curve 702 is a reference curve. Curve 704 is the inverse of the reference curve. Curve 706 is a delayed version of reference curve 702.

[0051] The discrete cross-correlation can be expressed as:

[0052]

[0053] Where k is the number of samples, x(k) is the reference speaker-microphone pair impulse response data vector, y(k) is the speaker-microphone pair impulse response data vector currently being evaluated, and N is the total number of data points in the two data vectors. A cross-correlation may be performed to determine the similarity between the reference impulse response of a speaker-microphone pair or speaker-accelerometer pair and the impulse response of the speaker-microphone pair or speaker-accelerometer pair of the active noise cancellation system currently under test. Alternatively or additionally, a correlation coefficient may be determined to indicate the similarity between the reference impulse response of the speaker-microphone pair and the speaker-microphone pair of the active noise cancellation system currently under test. The correlation coefficient may be expressed as:

[0054]

[0055] where x is the data vector, y is the data vector, and c xyis the correlation coefficient. If the impulse response of the currently evaluated speaker-microphone pair exactly matches the reference impulse response, the correlation coefficient value is 1. If the impulse response of the currently evaluated speaker-microphone pair is the inverse of the reference impulse response, the correlation coefficient value is -1. If the impulse response of the currently evaluated speaker-microphone pair is delayed from the reference impulse response, the correlation coefficient value is less than 1. A correlation coefficient of -1 may indicate that the electrical connection polarity of the speaker or microphone is incorrect.

[0056] Reference now Figure 8 , shows a visual aid for evaluating active noise cancellation systems via correlation coefficients. Figure 8 A first graph 802 is included, which shows correlation coefficient values ​​for one microphone (e.g., microphone 2) and seven speakers (S1-S7). The speaker indications S1-S7 are arranged in a polygon and are equally spaced around an origin 804. Black line 806 represents the maximum correlation coefficient of 1 at each speaker indication S1-S7. Line 808 represents an example correlation threshold value (e.g., 0.8 or 80%) for microphone #2 that is not degraded. In this example, the correlation coefficients of each of the speaker pairs S1-S4 and microphone #2 are acceptable, but the correlation coefficients of each of the speaker pairs S5-S6 are below the threshold. Therefore, the speaker-microphone pairs S5 / M2, S6 / M2, and S7 / M2 can be considered degraded.

[0057] Figure 8 Included with Figure 8 The second graph 820 is similar to the first graph, and the second graph 820 shows the correlation coefficient values ​​for one minus loudspeaker S1-S7 and microphone #2. This may be illustrative to show the error with the maximum correlation.

[0058] Go to Fig. 9 , shows a graph of correlation coefficient values ​​for a plurality of speaker-microphone pairs. In this example, forty correlation coefficient graphs for four different vehicles are shown. The correlation coefficient values ​​for each vehicle indicated by circles 901-904 are shown. The graphs are labeled to show which of the speaker-microphone pairs the correlation coefficient corresponds to. For example, Fig. 9 The top left graph shows the correlation coefficient values ​​for speaker #1 and microphone #1. The graph is labeled "Spk1Mic1" as an indication of the speaker-microphone pair to which the correlation coefficient corresponds.

[0059] Fig. 9 The steps of evaluating the impulse response of the speaker-microphone pair based on the correlation coefficient are also listed. Specifically, the impulse response curves of a specific speaker-microphone pair of multiple vehicles (e.g., Figure 2) and averaged at each sampling time to generate a speaker-microphone data vector that is the average of all impulse response data vectors at each identical time for each reference vehicle. For example, if there are only two impulse response data vectors for the speaker #1 / microphone #2 pair, and one of the data vectors has a value of 0.9 at 0.5 seconds and the other has a value of 0.8 at 0.5 seconds, then the value of the average impulse response vector at 0.5 seconds is 0.85. If one impulse response data vector has a value of 0.8 at 0.75 seconds and the other has a value of 0.7 at 0.75 seconds, then the value of the average impulse response data vector at 0.75 seconds is 0.75. The average values ​​at other times used to generate the average impulse response data vector can be determined in a similar manner.

[0060] In a second step, a correlation coefficient is determined based on the average impulse response data vector of the speaker-microphone pair (e.g., speaker #1 / microphone #2) of the plurality of vehicles and the impulse response of the speaker-microphone pair (e.g., speaker #1 / microphone #2) of the vehicle currently being tested or evaluated. This correlation coefficient may be compared to the minimum correlation coefficient of a reference vehicle, minus an offset, that was used to generate the average impulse response vector of the reference vehicle in the above step. For example, if the correlation coefficient of the speaker-microphone pair of speaker #1 / microphone #2 of the vehicle currently being tested or evaluated is 0.7, the correlation coefficient of the speaker-microphone pair of speaker #1 / microphone #2 of the first reference vehicle is 0.82, the correlation coefficient of the speaker-microphone pair of speaker #1 / microphone #2 of the second reference vehicle is 0.85, and the correlation coefficient of the speaker-microphone pair of speaker #1 / microphone #2 of the third reference vehicle is 0.9, and the offset is 0.02, it can be determined that the speaker-microphone pair speaker #1 / microphone #2 is degraded because 0.82 (the minimum correlation coefficient of the three reference vehicles) minus 0.02 (the offset / margin value) is 0.8, which is greater than 0.7 (the correlation coefficient of the vehicle currently being tested).

[0061] Reference now Fig.10 , a diagram showing a visual correlation analysis. Fig.10 There are sixteen graphs, one for each microphone, and the microphone is identified at the top of each graph. Figure 8 Each graph shows the correlation coefficient values ​​for a speaker-microphone pair, and the correlation coefficient values ​​are tied together by a line to show the relationship between the speaker input and the microphone output. Fig.10 It is shown that the correlation coefficient of microphone #1 is below the threshold. Therefore, it can be determined that microphone #1 is degraded.

[0062] Reference now Fig.11 , a diagram showing a visual correlation analysis. Fig.11 11 plots are included and show a minus correlation coefficient analysis. The microphone plot indicates which microphone the plot is associated with (e.g., Mic 1 = microphone #1). The speaker plot indicates which speaker the plot is associated with (e.g., Spk 1 = speaker #1). Speakers are indicated with SX labels, where X corresponds to an integer. Microphones are indicated with MX labels, where X corresponds to an integer. Fig.11 The diagram indicates that the wiring polarity of speaker No. 7 is reversed, and the reception of microphone No. 5 is blocked.

[0063] Reference now Fig.12 , shows a diagram illustrating how a vehicle accelerometer can be evaluated by a frequency domain impulse response analysis procedure. Fig.12 Twelve graphs are included. Each of the twelve graphs includes a vertical axis representing accelerometer amplitude and a vertical axis representing frequency in Hertz. The accelerometer data includes accelerometer data collected from five different vehicles operating on the same road section, and the data is shown as five different lines (1201)-(1205). In addition, each graph includes a maximum line (1206) and a maximum threshold line (1207). These graphs can be generated by calculating the frequency spectrum of the accelerometer output according to multiple driving conditions (for example, driving on different roads). The maximum curve is determined by determining the maximum amplitude value at each frequency from the multiple driving conditions and connecting the maximum amplitude values ​​to generate a maximum curve. The maximum curve can be converted to a maximum threshold curve by adding a margin or offset value to each value of the maximum curve. The output of the vehicle accelerometer can be converted to the frequency domain and compared with the maximum threshold curve. If the accelerometer frequency domain output exceeds the maximum threshold curve at a certain point or position, it can be determined that the accelerometer is degraded.

[0064] Figure 13-Figure 16 The features of the vehicle active noise cancellation system diagnostic tool are shown. Specifically, Fig.13 Configuration data, channel numbers, channel names, impulse response tolerances, and accelerometer amplitudes can be loaded into the tool. The tool can display the boundary ranges for the analysis of each of the various system components. The frequency domain responses of the various system actuators and sensors can be plotted, as shown in Figure 1. Fig.14 As shown. Fig.15 As shown, different markers are provided to divide the impulse response data. In addition, the frequency domain impulse response data out of range can be highlighted to increase visibility. Fig.16 Shows how a degraded sensor may be indicated.

[0065] Reference now Fig.17 , a method for evaluating the operation of an active noise cancellation system of a vehicle is shown. Fig.17The method may be included as executable instructions stored in the controller memory. Figure 1 in the system. Fig.17 The method may be executed by a controller of a remote device or an in-vehicle device. Fig.17 The method can diagnose the active noise cancellation system and generate the curves or traces shown in the figures included herein.

[0066] At 1702 , an embedded controller of an active noise cancellation system is installed into a vehicle on an assembly line. The installation includes securing the embedded controller to the vehicle. The method 1700 proceeds to 1704 .

[0067] At 1704 , a speaker of an active noise cancellation system is installed into a vehicle on an assembly line. The installation includes securing the speaker to the vehicle and electrically connecting the speaker to an embedded controller. The method 1700 proceeds to 1706 .

[0068] At 1706 , a microphone of an active noise cancellation system is installed into a vehicle on an assembly line. The installation includes securing the microphone to the vehicle and electrically connecting the microphone to an embedded controller. The method 1700 proceeds to 1708 .

[0069] At 1708 , an accelerometer of an active noise cancellation system is installed into a vehicle on an assembly line. The installation includes securing the accelerometer to the vehicle and electrically connecting the accelerometer to the embedded controller. The method 1700 proceeds to 1710 .

[0070] At 1710, a broadband impulse signal is transmitted to each speaker of the active noise cancellation system. Each speaker is provided with a broadband impulse signal at a different time than other speakers in the system so that the output of one speaker does not interfere with the output of a different speaker. The method 1700 also captures the output from each microphone and accelerometer during the time window when each speaker is driven by the broadband impulse signal. The microphone output can be stored as data in the controller memory. The method 1700 proceeds to 1712.

[0071] At 1712, method 1700 calculates an impulse response for each speaker-microphone combination and speaker-accelerometer combination. The impulse response may be calculated by dividing a data vector from a microphone by data representing an input to a speaker, where an output of the speaker is sensed by the microphone. In other words, for a speaker-microphone pair, data representing the output of the microphone is divided by data representing the input to the speaker. Thus, for a speaker-microphone pair speaker #1 / microphone #2, data representing the output of microphone #1 is divided by data representing the input to speaker #1. Method 1700 proceeds to 1714.

[0072] At 1714, method 1700 calculates the frequency spectrum of the impulse response determined at step 1712. The frequency response can be calculated by Fourier transform as discussed above. Alternatively, method 1700 can calculate the correlation coefficient between the reference impulse response and the impulse response determined at step 1712. In other examples, method 1700 can calculate the cross-correlation between the reference impulse response and the impulse response determined at step 1712. Method 1700 proceeds to 1716.

[0073] At 1716, for frequency domain analysis, the method 170 determines whether the amplitude in the frequency domain of the impulse response determined at step 1712 is outside a threshold range between two boundaries. Alternatively, the method 170 determines whether the amplitude in the frequency domain of the impulse response determined at step 1712 exceeds a threshold boundary. The amplitude threshold may be determined by one of the methods described herein.

[0074] For correlation coefficient analysis, method 1700 may determine whether the correlation coefficient value generated at step 1714 is less than a threshold correlation coefficient value.

[0075] If the amplitude is out of bounds or exceeds the boundaries of the frequency domain analysis for one of the speaker-microphone pairs, or if the correlation coefficient is less than the correlation coefficient generated at step 1714, the answer is yes and method 1700 proceeds to 1718. Otherwise, the answer is no and method 1700 proceeds to 1720.

[0076] If method 1700 performs a cross-correlation between the reference impulse response and the impulse response of the current unit under test, method 1700 may proceed to 1718 if the result of the cross-correlation is less than a threshold boundary.

[0077] At 1718, method 1700 displays the frequency domain analysis or cross-correlation analysis data and provides an indication of sensor or actuator degradation. Method 1700 proceeds to exit.

[0078] At 1720, method 1700 displays the frequency domain analysis or cross-correlation analysis data and indicates no sensor or actuator degradation. Method 1700 proceeds to exit.

[0079] The description of the embodiments has been presented for purposes of illustration and description. Suitable modifications and variations of the embodiments may be performed in light of the above description or may be acquired by practicing the methods. For example, unless otherwise indicated, one or more of the methods described may be performed by suitable apparatus and / or combinations of apparatus (such as Figure 3The described systems are exemplary in nature and may include additional elements and / or omit elements. The subject matter of the present disclosure includes all novel and non-obvious combinations and sub-combinations of the various systems and configurations disclosed and other features, functions, and / or properties.

[0080] As used in this application, an element or step recited in the singular and preceded by the word "one" or "a kind of" should be understood as not excluding a plurality of said elements or steps, unless such exclusion is provided. In addition, reference to "one embodiment" or "an example" of the present disclosure is not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the described features. The terms "first," "second," and "third," etc. are used only as labels and are not intended to impose numerical requirements or specific positional order on their objects. The following claims specifically point out subject matter from the above disclosure that is considered novel and non-obvious.

Claims

1. A diagnostic tool for an active noise cancellation system, comprising: A controller including executable instructions stored in a non-transitory memory that causes the controller to initiate an impulse response diagnostic for the active noise cancellation system, wherein the impulse response diagnostic includes performing a cross-correlation between a reference impulse response and an impulse response of the active noise cancellation system.

2. The diagnostic tool of claim 1, wherein the impulse response diagnostic comprises supplying a broadband signal to one or more speakers of the active noise cancellation system and generating sound through the one or more speakers in response to the broadband signal.

3. The diagnostic tool of claim 2, wherein the sound is sensed by one or more microphones or one or more accelerometers of the active noise cancellation system. 4 . The diagnostic tool of claim 3 , further comprising generating a correlation coefficient based on the reference impulse response and the impulse response of the active noise cancellation system.

5. The diagnostic tool of claim 4, wherein the reference impulse response is based on a plurality of impulse responses determined from an output device and an input device.

6. The diagnostic tool of claim 5, wherein the output device is a speaker, and wherein the input device is a microphone or an accelerometer.

7. The diagnostic tool of claim 4, further comprising generating an indication of degradation of the active noise cancellation system based on the value of the correlation coefficient.

8. The diagnostic tool of claim 7, wherein the indication of degradation is an indication that the speaker wiring has reversed electrical polarity.

9. A method for diagnosing an active noise cancellation system, comprising: supplying a broadband signal to a speaker of an active noise cancellation system; sensing, via a microphone, an audible output of the speaker in response to the broadband signal; as well as A cross-correlation analysis is performed on an impulse response generated from a signal output from the microphone sensing the audible output of the speaker.

10. The method of claim 9, wherein the impulse response is generated by dividing the signal output from the microphone by the wideband signal, and wherein the active noise cancellation system is implemented at the end of a vehicle assembly line after the speaker and the microphone are installed in a vehicle.

11. The method of claim 10, wherein performing the cross-correlation analysis comprises performing a cross-correlation based on a reference impulse response and the impulse response generated from the signal output from the microphone.

12. The method of claim 11, further comprising generating a correlation coefficient based on the reference impulse response and the impulse response generated from the signal output from the microphone.

13. The method of claim 12, wherein the reference impulse response is an average impulse response of a plurality of active noise cancellation systems of a plurality of vehicles. The method according to claim 11 , further comprising displaying the correlation coefficient through a human-computer interface. 15 . The method of claim 14 , further comprising displaying the correlation coefficient and a plurality of other correlation coefficients from speakers other than the speaker in a polygonal shape through the human-machine interface.

Images