A multi-sensor based active noise reduction method for automobiles

Abstract

The present invention relates to the technical field of automobile noise reduction, and in particular to a method for active automobile noise reduction based on multiple sensors, comprising: selecting deployment points inside an automobile, deploying audio sensors based on the selected deployment points, and using the audio sensors to sense noise audio; acquiring the noise audio sensed by the audio sensors, marking the location information of the source audio sensors for each noise audio, transmitting the marked noise audio to Audacity for format conversion, and outputting a corresponding audio spectrum; the present invention adaptively deploys multiple groups of audio sensors to collect noise audio inside the automobile, further analyzes the complexity of the noise audio, makes a decision and selects the most complex noise audio, generates noise reduction audio corresponding to the decided and selected noise audio, and then plays it through a vehicle-mounted speaker, so as to achieve the goal of active noise reduction in the vehicle, provide a more comfortable noise reduction experience for automobile driving users, and thereby realize a more intelligent and effective noise reduction service.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile noise reduction, and in particular to an automobile active noise reduction method based on multiple sensors. Background Art

[0002] Active noise reduction in cars is an advanced technology. It emits sound waves with opposite frequency and amplitude to the noise through the car audio system, thereby effectively canceling out the noise inside the car, improving driving comfort, reducing external noise interference, and allowing the driver to enjoy a quiet environment while driving and focus on driving or relaxing.

[0003] The invention patent with application number 201510845558.4 discloses an active noise reduction method for automobiles, which is characterized by comprising: detecting the seat position of a person in the car by means of a pressure sensor arranged in a car seat; collecting the sound source in the car by means of microphones arranged at different positions in the car; estimating the central area of ​​the seat position of the person in the car; separating the noise at the position of the person in the car according to the collected sound source; calculating the center position of the field sound formed by the noise at the position of each person in the car; and analyzing the spectrum diagram of the noise; turning on the car audio system, adjusting the center position of the sound field of the car audio system to the center position of the field sound formed by the noise; generating a suppression wave with the same frequency and amplitude as the noise and a phase difference of 180 degrees to reduce the noise.

[0004] The application aims to solve the problem that "in the prior art, noise suppression in a car cabin is mainly achieved by improving the sealing of the cabin, the shock absorption system of the chassis, the sound insulation of the engine compartment and the selection of silent tires. The above technologies are all passive noise reduction technologies, which can indeed reduce the decibels of noise in the cabin, but the degree of noise reduction is limited, especially when the car is driving at high speed, the noise in the cabin is still very strong."

[0005] However, in the actual application of active noise reduction in automobiles, the noise audio of its noise reduction target often has the characteristics of multi-source, which results in the noise reduction audio emitted by the car speakers often failing to offset the key noise audio, resulting in poor active noise reduction effect of the automobile.

[0006] To this end, we propose a multi-sensor based active noise reduction method for automobiles. Summary of the invention

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a multi-sensor based active noise reduction method for an automobile, which solves the technical problems raised in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A multi-sensor based automobile active noise reduction method, comprising:

[0010] Deployment points are selected inside the car, audio sensors are deployed based on the selected deployment points, and the audio sensors are used to sense noise audio; the noise audio sensed by the audio sensor is obtained, the location information of the source audio sensor of each noise audio is marked, and the marked noise audio is transmitted to Audacity for format conversion to output the corresponding audio spectrum; each group of audio spectrum output by Audacity is obtained, and each group of audio spectrum is used as an analysis target to analyze the complexity of each group of audio spectrum; the analysis results of the complexity of the audio spectrum are obtained, and a group of audio spectrum with the highest complexity is selected as the processing target, and a speaker is installed inside the car, and the speaker is connected to the vehicle center control panel, and the vehicle center control panel is connected to the computer device driving Audacity to run; the group of audio spectrum with the highest complexity is used as the editing target, and a group of audio spectrum with the opposite waveform to the audio spectrum with the highest complexity is edited through Audacity, and the edited audio spectrum is converted into noise reduction audio through Audacity again, and the noise reduction audio is transmitted to the speaker through the vehicle center control panel; after receiving the noise reduction audio transmitted by the vehicle center control panel, the speaker plays the noise reduction audio in real time.

[0011] Furthermore, the deployment point is located in the stage of selecting the interior of the car, and the specification parameters of the car structure and interior accessories are uploaded synchronously, a three-dimensional model of the car is constructed based on the specification parameters of the car structure and interior accessories, the interior space structure model of the car is picked up in the three-dimensional model of the car, the inner corner is captured on the interior space structure model of the car, and the inner corner is recorded as a pre-selected deployment point, and the number of deployment point selections is further customized by the user end, and the audio sensor deployment point is selected from the pre-selected deployment points based on the number of deployment point selections;

[0012] Each deployment point has a set of audio sensors deployed. When selecting a deployment point from the pre-selected deployment points, the following conditions apply:

[0013] Identify the distance between each pre-selected deployment point and the center point of the vehicle's internal space structure model, and based on the number of deployment point selections, give priority to selecting deployment points that are far from the center point of the vehicle's internal space structure model, until the number of deployment points selected from the pre-selected deployment points reaches the set number.

[0014] Furthermore, the audio spectrum based on the Audacity output is marked with the source audio sensor position information of the corresponding noise audio mark and stored on the computer device driving Audacity to run.

[0015] Furthermore, in the stage of analyzing the complexity of the audio spectrum, the audio sensor position information marked by each group of audio spectrum is identified, and the audio spectrum corresponding to the group of audio sensor position information farthest from the center point of the interior space structure model of the car is used as the first analysis target, and the audio spectrum corresponding to the audio sensor position information second farthest from the center point of the interior space structure model of the car is used as the second analysis target, and so on, so as to construct an audio spectrum complexity analysis queue, and perform complexity analysis on the audio in sequence based on the analysis queue;

[0016] The complexity analysis logic of the audio spectrum is expressed as:

[0017] ;

[0018] Where: is the complexity of the audio spectrum; is the total amount of peak points in the audio spectrum; is the corresponding value of the i-th peak point in the audio spectrum; is the corresponding value of the jth peak point in the audio spectrum; is the weight;

[0019] Among them, the complexity of the audio spectrum The larger the value, the more complex the audio spectrum is, and vice versa, the lower the complexity of the audio spectrum is.

[0020] Furthermore, the peak point i comes from the total amount of peak points in the audio spectrum , Represents the set of peak points in the audio spectrum that does not contain peak point i;

[0021] Weight The value of is subject to:

[0022] ;

[0023] Where: is the cumulative reference value; , is the length of the wave lines on both sides of the corresponding peak point; is the total length of the band in the audio spectrum;

[0024] in, represents the reference value of peak point i, represents the reference value of peak point j,

[0025] Weight ∈(0, 1), and Inversely proportional.

[0026] Furthermore, the speaker is electrically connected to the vehicle center control panel through a dielectric, and the vehicle center control panel is connected to the computer device driving Audacity through a network;

[0027] The three groups of speakers, the vehicle's central control panel and the computer equipment that drives Audacity perform real-time data interaction through the network.

[0028] Furthermore, the time from when the audio sensor senses the noise audio to when the speaker starts to play the noise reduction audio does not exceed 20ms, and when the audio sensor senses the noise audio, the time span of the noise audio sensed each time is consistent with the time from when the audio sensor senses the noise audio to when the speaker starts to play the noise reduction audio;

[0029] Among them, the time from the time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio is customized by the user end, and after the time from the time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio is set by the user end, the set time from the time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio is further coordinated based on the coordination logic.

[0030] Furthermore, the coordination logic is expressed as:

[0031] ;

[0032] Where: It is the time from the next time the audio sensor senses the noise audio to the time when the speaker starts playing the noise reduction audio; The time from when the current audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio; for The corresponding vehicle speed; is the current vehicle speed; is the normalization factor.

[0033] Furthermore, the coordination logic calculates the time from the next time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio based on formula (1). Then, based on formula (2), the time from the next time the audio sensor senses the noise audio to the time when the speaker starts to play the noise reduction audio is calculated. Execute the decision of specific value.

[0034] Furthermore, the coordination logic in equation (1) only considers situation, When the audio sensor senses the noise audio and the speaker starts playing the noise reduction audio, no coordination operation is performed.

[0035] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0036] The present invention provides a multi-sensor based automobile active noise reduction method. During the execution of the method, multiple groups of audio sensors are adaptively deployed to collect noise audio inside the automobile, and the complexity of the noise audio is further analyzed to make a decision and select the most complex noise audio. Then, noise reduction audio is generated accordingly based on the decided and selected noise audio, and then played through the vehicle-mounted speakers to achieve the goal of active noise reduction in the car, provide a more comfortable noise reduction experience for automobile driving users, and thereby realize a more intelligent and effective noise reduction service. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 The figure is a flow chart of a multi-sensor based automotive active noise reduction method. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] The present invention will be further described below in conjunction with the embodiments. Example 1

[0041] A multi-sensor based active noise reduction method for an automobile in this embodiment, such as Figure 1 As shown, including:

[0042] Step 1: Select deployment points inside the car, deploy audio sensors based on the selected deployment points, and use the audio sensors to sense noise audio;

[0043] In the stage of selecting the deployment point inside the car, the car structure and interior accessories specifications are uploaded simultaneously, a 3D model of the car is constructed based on the car structure and interior accessories specifications, the car interior space structure model is picked up in the 3D model, the inner corner is captured on the car interior space structure model, and the inner corner is recorded as a pre-selected deployment point. The number of deployment points selected is further customized by the user end, and the audio sensor deployment point is selected from the pre-selected deployment points based on the number of deployment points selected;

[0044] Each deployment point has a set of audio sensors deployed. When selecting a deployment point from the pre-selected deployment points, the following conditions apply:

[0045] Identify the distance between each pre-selected deployment point and the center point of the vehicle interior space structure model, and select the deployment point that is far from the center point of the vehicle interior space structure model based on the number of deployment points selected, and select the deployment point until the number of deployment points selected from the pre-selected deployment points reaches the set number;

[0046] Step 2: Obtain the noise audio sensed by the audio sensor, mark the location information of the audio sensor from which each noise audio comes, and transfer the marked noise audio to Audacity for format conversion to output the corresponding audio spectrum;

[0047] Step 3: Obtain each group of audio spectra output by Audacity, take each group of audio spectra as the analysis target, and analyze the complexity of each group of audio spectra;

[0048] Step 4: Obtain the audio spectrum complexity analysis results, select a group of audio spectra with the highest complexity as the processing target, install speakers inside the car, connect the speakers to the car's central control panel, and connect the car's central control panel to the computer device that drives Audacity to run;

[0049] Step 5: Take the most complex audio spectrum as the editing target, edit a set of audio spectrum with the opposite waveform to the most complex audio spectrum through Audacity, convert the edited audio spectrum into noise reduction audio through Audacity again, and transmit the noise reduction audio to the speaker through the vehicle center control panel;

[0050] Step 6: After receiving the noise reduction audio transmitted by the vehicle's central control panel, the speaker plays the noise reduction audio in real time;

[0051] The time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio does not exceed 20ms, and when the audio sensor senses the noise audio, the time span of the noise audio sensed each time is consistent with the time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio;

[0052] The time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio is customized by the user end, and after the time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio is set by the user end, the time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio is further coordinated based on the coordination logic;

[0053] The coordination logic is expressed as:

[0054] ;

[0055] Where: It is the time from the next time the audio sensor senses the noise audio to the time when the speaker starts playing the noise reduction audio; The time from when the current audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio; for The corresponding vehicle speed; is the current vehicle speed; is the normalization factor.

[0056] In this embodiment, by executing the method in the above embodiment, the existing active noise reduction technology for automobiles is further improved, so that the automobile noise reduction technology can make decisions and select the preferred noise reduction target noise based on the noise complexity analysis, so that the automobile noise reduction effect can play its role to a greater extent, providing a more comfortable noise reduction experience for automobile driving users, and thus realizing a more intelligent and effective noise reduction service; it should be noted that in actual application, the above formula Usually less than one second, so here Although it can be understood as a time threshold, its corresponding Can be regarded as the only one. Example 2

[0057] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The multi-sensor based active noise reduction method for automobiles in Example 1 is further described in detail:

[0058] The audio spectrum outputted by Audacity is labeled with the source audio sensor position information of the corresponding noise audio label and stored in the computer device driving Audacity;

[0059] In the stage of analyzing the complexity of the audio spectrum, the audio sensor position information marked by each group of audio spectrum is identified, and the audio spectrum corresponding to the group of audio sensor position information farthest from the center point of the interior space structure model of the car is used as the first analysis target, and the audio spectrum corresponding to the audio sensor position information second farthest from the center point of the interior space structure model of the car is used as the second analysis target, and so on, so as to construct an audio spectrum complexity analysis queue, and perform complexity analysis on the audio in sequence based on the analysis queue;

[0060] The complexity analysis logic of the audio spectrum is expressed as:

[0061] ;

[0062] Where: is the complexity of the audio spectrum; is the total amount of peak points in the audio spectrum; is the corresponding value of the i-th peak point in the audio spectrum; is the corresponding value of the jth peak point in the audio spectrum; is the weight;

[0063] Among them, the complexity of the audio spectrum The larger the value, the more complex the audio spectrum is, and vice versa, the less complex the audio spectrum is.

[0064] The source of peak i is the total number of peaks in the audio spectrum , Represents the set of peak points in the audio spectrum that does not contain peak point i;

[0065] Weight The value of is subject to:

[0066] ;

[0067] Where: is the cumulative reference value; , is the length of the wave lines on both sides of the corresponding peak point; is the total length of the band in the audio spectrum;

[0068] in, represents the reference value of peak point i, Indicates the reference value of peak point j, weight ∈(0, 1), and Inversely proportional.

[0069] In this embodiment, through the above-mentioned settings, further step execution data support is provided for the method in Example 1 to ensure stable execution of the steps of the method in Example 1. At the same time, a complexity analysis logic of a specified audio spectrum is provided to ensure that the method in Example 1 is further supported by data and a more effective and critical noise elimination target is determined. Example 3

[0070] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The multi-sensor based active noise reduction method for automobiles in Example 1 is further described in detail:

[0071] The speaker is electrically connected to the vehicle's central control panel through a dielectric, and the vehicle's central control panel is connected to the computer device driving Audacity through a network;

[0072] The three groups of speakers, the vehicle's central control panel, and the computer device that drives Audacity to perform real-time data interaction through the network;

[0073] In the coordination logic, based on formula (1), the time from the next time the audio sensor senses the noise audio to the time when the speaker starts playing the noise reduction audio is calculated. Then, based on formula (2), the time from the next time the audio sensor senses the noise audio to the time when the speaker starts to play the noise reduction audio is calculated. Execute the decision of specific value;

[0074] In coordination logic, only situation, When the audio sensor senses the noise audio and the speaker starts playing the noise reduction audio, no coordination operation is performed.

[0075] In this embodiment, through the above settings, further step execution data support is provided for the method in Example 1, and the application conditions of the coordination logic are further limited, ensuring that the method in Example 1 can serve the in-vehicle noise reduction task with higher matching degree and better adaptability based on the above settings.

[0076] In summary, during the execution of the method in the above embodiment, multiple groups of audio sensors are adaptively deployed to collect the noise audio inside the car, and the complexity of the noise audio is further analyzed to decide and select the most complex noise audio, and then generate noise reduction audio according to the decided and selected noise audio, and then play it through the car speakers to achieve the goal of active noise reduction in the car, provide a more comfortable noise reduction experience for car drivers, and thereby realize a more intelligent and effective noise reduction service.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-sensor based active noise reduction method for automobiles, characterized in that: include: Step 1: Select deployment points inside the car, deploy audio sensors based on the selected deployment points, and use the audio sensors to sense noise audio; Step 2: Obtain the noise audio sensed by the audio sensor, mark the location information of the audio sensor from which each noise audio comes, and transfer the marked noise audio to Audacity for format conversion to output the corresponding audio spectrum; Step 3: Obtain each group of audio spectra output by Audacity, take each group of audio spectra as the analysis target, and analyze the complexity of each group of audio spectra; Step 4: Obtain the audio spectrum complexity analysis results, select a group of audio spectra with the highest complexity as the processing target, install speakers inside the car, connect the speakers to the car's central control panel, and connect the car's central control panel to the computer device that drives Audacity to run; Step 5: Take the most complex audio spectrum as the editing target, edit a set of audio spectrum with the opposite waveform to the most complex audio spectrum through Audacity, convert the edited audio spectrum into noise reduction audio through Audacity again, and transmit the noise reduction audio to the speaker through the vehicle center control panel; Step 6: After receiving the noise reduction audio transmitted by the vehicle's central control panel, the speaker plays the noise reduction audio in real time; The analysis of the complexity of each group of audio spectra includes: identifying the audio sensor position information marked by each group of audio spectra, taking the audio spectrum corresponding to the group of audio sensor position information farthest from the center point of the interior space structure model of the car as the first analysis target, taking the audio spectrum corresponding to the audio sensor position information second farthest from the center point of the interior space structure model of the car as the second analysis target, and so on, so as to construct an audio spectrum complexity analysis queue, and perform complexity analysis on the audio in sequence based on the analysis queue; The complexity analysis logic of the audio spectrum is expressed as: Where: δ is the complexity of the audio spectrum; n is the total number of peaks in the audio spectrum; h i is the corresponding value of the i-th peak point in the audio spectrum; h j is the corresponding value of the jth peak point in the audio spectrum; ω is the weight; The larger the value of the audio spectrum complexity δ is, the more complex the audio spectrum is, and vice versa, the lower the audio spectrum complexity is. The peak point i comes from the total number of peak points n in the audio spectrum, and ni represents the set of peak points in the audio spectrum that does not contain the peak point i; The value of weight ω follows: Where: q is the cumulative reference value; L1 and L2 are the lengths of the wave lines on both sides of the corresponding peak point; L ALL is the total length of the band in the audio spectrum; in, represents the reference value of peak point i, Represents the reference value of peak point j, with weight ω∈(0,1), and q is inversely proportional to ω.

2. The multi-sensor based active noise reduction method for automobiles according to claim 1, characterized in that: The deployment point is located in the car interior selection stage, and the car structure and interior accessories specification parameters are uploaded synchronously, and the car three-dimensional model is constructed based on the car structure and interior accessories specification parameters, and the car interior space structure model is picked up in the car three-dimensional model, and the inner corner is captured on the car interior space structure model. The inner corner is recorded as a pre-selected deployment point, and the number of deployment point selections is further customized by the user end, and the audio sensor deployment point is selected from the pre-selected deployment points based on the number of deployment point selections; Each deployment point has a set of audio sensors deployed. When selecting a deployment point from the pre-selected deployment points, the following conditions apply: Identify the distance between each pre-selected deployment point and the center point of the vehicle's internal space structure model, and based on the number of deployment point selections, give priority to selecting deployment points that are far from the center point of the vehicle's internal space structure model, until the number of deployment points selected from the pre-selected deployment points reaches the set number.

3. The multi-sensor based active noise reduction method for automobiles according to claim 1, characterized in that: The audio spectrum based on the Audacity output is labeled with the source audio sensor position information of its corresponding noise audio label and stored on the computer device driving Audacity.

4. The multi-sensor based active noise reduction method for automobiles according to claim 1, characterized in that: The speaker is electrically connected to the vehicle-mounted central control panel through a dielectric, and the vehicle-mounted central control panel is connected to a computer device driving Audacity to run through a network; The three groups of speakers, the vehicle's central control panel and the computer equipment that drives Audacity perform real-time data interaction through the network.

5. The multi-sensor based active noise reduction method for automobiles according to claim 1, characterized in that: The time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio does not exceed 20ms, and when the audio sensor senses the noise audio, the time span of the noise audio sensed each time is consistent with the time from when the audio sensor senses the noise audio to when the speaker starts playing the noise reduction audio; Among them, the time from the time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio is customized by the user end, and after the time from the time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio is set by the user end, the set time from the time the audio sensor senses the noise audio to the time the speaker starts playing the noise reduction audio is further coordinated based on the coordination logic.

6. The multi-sensor based active noise reduction method for automobiles according to claim 5, characterized in that: The coordination logic is expressed as: Where: s next It is the time from the next time the audio sensor senses the noise audio to the time when the speaker starts playing the noise reduction audio; s0 is the time from when the current audio sensor senses the noise audio to when the speaker starts to play the noise reduction audio; v0 is the vehicle speed corresponding to s0; v now is the current vehicle speed; γ is the normalization factor.

7. The multi-sensor based active noise reduction method for automobiles according to claim 6, characterized in that: In the coordination logic, the time s from the next time the audio sensor senses the noise audio to the time when the speaker starts to play the noise reduction audio is calculated based on formula (1): next Then, based on formula (2), the time s from the next time the audio sensor senses the noise audio to the time when the speaker starts to play the noise reduction audio is calculated. next Execute the decision of specific value.

8. The multi-sensor based active noise reduction method for automobiles according to claim 6, characterized in that: In the coordination logic, only v0<v is considered in formula (1) now In the case of v0≥v now When the audio sensor senses the noise audio and the speaker starts playing the noise reduction audio, no coordination operation is performed.

Images