In-car sound field zoning control method, device and car speaker

Abstract

The present application relates to the field of audio processing technology, and discloses a method, device and car speaker for controlling the zoning of the sound field in a vehicle. The method comprises: performing zoning and data collection of the sound field in the vehicle interior space based on the speaker distribution cluster and multiple seat receiving terminals, and obtaining multiple regional sound field distribution data; performing sound field distribution feature detection, and obtaining a sound field distribution feature subset; performing multi-dimensional sound field evaluation index calculation and vector mapping, and obtaining multiple sound field distribution feature vectors and creating a first sound field zoning control parameter set; creating an in-vehicle sound field space propagation network; performing state feature extraction, and obtaining a state feature parameter set; generating a sound field parameter control execution strategy according to the state feature parameter set and optimizing and solving the control parameters of the first sound field zoning control parameter set, and obtaining a second sound field zoning control parameter set. The present application improves the accuracy of the in-vehicle sound field zoning control.

Description

Technical Field

[0001] The present application relates to the field of audio processing technology, and in particular to a method and device for controlling the zoning of sound fields in a vehicle, and a vehicle speaker. Background Art

[0002] Traditional in-car audio systems mostly use a single audio configuration. The sound quality distributed within the entire vehicle space of this configuration often cannot meet the auditory needs of all passengers, especially under various noise conditions, such as wind noise and road noise during driving.

[0003] In addition, with the rise of intelligent driving technology and personalized services, the in-car audio system not only plays the role of entertainment, but also needs to play a greater role in ensuring driving safety and providing in-car information services. For example, the sound field control in different areas can help drivers receive navigation sounds more clearly without interfering with the music experience of other passengers. However, the existing sound field control technology still has limitations in the precise control of sound propagation and regional division, and it is often difficult to achieve the desired effect in a complex in-car environment. Summary of the invention

[0004] The present application provides a method and device for controlling the zoning of the sound field in a vehicle, and a car speaker, which are used to improve the accuracy of controlling the zoning of the sound field in the vehicle.

[0005] In a first aspect, the present application provides a method for controlling a sound field in a vehicle by partition, the method comprising:

[0006] Based on the speaker distribution cluster and the multiple seat receiving terminals, the interior space of the vehicle is partitioned into a vehicle sound field to obtain multiple seat sound field areas, and data is collected for the multiple seat sound field areas to obtain multiple area sound field distribution data;

[0007] Performing sound field distribution feature detection on the sound field distribution data of the plurality of regions to obtain a sound field distribution feature subset of each seat sound field region;

[0008] Performing multi-dimensional sound field evaluation index calculation and vector mapping on the multiple seat sound field areas according to the sound field distribution feature subset to obtain multiple sound field distribution feature vectors, and creating a first sound field partition control parameter set according to the multiple sound field distribution feature vectors and a particle swarm algorithm;

[0009] Creating an in-vehicle sound field spatial propagation network of the vehicle interior space according to the speaker distribution cluster, the plurality of seat receiving terminals and the plurality of seat sound field areas;

[0010] Extracting state features of multiple propagation path nodes in the in-vehicle sound field spatial propagation network to obtain a state feature parameter set of each propagation path node;

[0011] A sound field parameter control execution strategy is generated according to the state characteristic parameter set, and the control parameter optimization solution is performed on the first sound field partition control parameter set to obtain a second sound field partition control parameter set.

[0012] In a second aspect, the present application provides a device for controlling a sound field in a vehicle by partitions, the device comprising:

[0013] A collection module, used for partitioning the interior space of the vehicle into a sound field based on the speaker distribution cluster and the multiple seat receiving terminals to obtain multiple seat sound field areas, and collecting data on the multiple seat sound field areas to obtain multiple area sound field distribution data;

[0014] A detection module, used to perform sound field distribution feature detection on the sound field distribution data of the plurality of regions, and obtain a sound field distribution feature subset of each seat sound field region;

[0015] a calculation module, configured to perform multi-dimensional sound field evaluation index calculation and vector mapping on the plurality of seat sound field areas according to the sound field distribution feature subset, obtain a plurality of sound field distribution feature vectors, and create a first sound field partition control parameter set according to the plurality of sound field distribution feature vectors and a particle swarm algorithm;

[0016] A creation module, configured to create an in-vehicle sound field spatial propagation network of the vehicle interior space according to the speaker distribution cluster, the plurality of seat receiving terminals and the plurality of seat sound field areas;

[0017] An extraction module is used to extract state features of multiple propagation path nodes in the in-vehicle sound field spatial propagation network to obtain a state feature parameter set of each propagation path node;

[0018] A solution module is used to generate a sound field parameter control execution strategy according to the state characteristic parameter set and perform control parameter optimization solution on the first sound field partition control parameter set to obtain a second sound field partition control parameter set.

[0019] A third aspect of the present application provides a car speaker, which is used to execute the above-mentioned in-vehicle sound field zoning control method.

[0020] A fourth aspect of the present application provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes the above-mentioned in-vehicle sound field zoning control method.

[0021] In the technical solution provided by the present application, the method can accurately control and optimize the sound effects of various areas in the car through the sound field partition based on the speaker distribution cluster and multiple seat receiving ends. This fine partitioning enables each seat area to obtain the most suitable audio output, thereby maintaining the consistency of the sound quality of the whole vehicle while also meeting the personalized needs of passengers for the sound environment. Through data collection and sound field distribution feature detection of multiple sound field areas in the car, the system can effectively understand and analyze the acoustic characteristics of each area. Using technologies such as wavelet filtering and sound field propagation feature analysis, the interference of environmental noise can be removed and the accuracy and practicality of sound field data can be improved. The particle swarm algorithm is used to analyze and optimize the sound field distribution feature vector, and the sound field parameters can be automatically adjusted to adapt to the dynamic environmental changes inside the vehicle. The dynamic control supported by this algorithm not only improves the efficiency of sound field control, but also optimizes the sound effect in real time according to the actual situation, ensuring that each passenger can get the best auditory experience. The construction of the in-car sound field space propagation network takes into account the relative position between the speaker and the seat and the sound field propagation path, and effectively integrates the acoustic resources inside the vehicle. By precisely controlling the direction and intensity of sound field propagation, the network ensures the uniformity of sound distribution and the sound quality of the target area, thereby improving the accuracy of in-vehicle sound field zoning control. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0023] Figure 1 A schematic diagram of an embodiment of a method for controlling the zoning of the in-vehicle sound field in an embodiment of the present application;

[0024] Figure 2 It is a schematic diagram of an embodiment of the in-vehicle sound field zoning control device in the embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application provide a method, device and car speaker for controlling the zoning of the sound field in a car. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the method for controlling the zoning of the in-vehicle sound field includes:

[0027] Step S101, partitioning the interior space of the vehicle into vehicle sound fields based on the speaker distribution cluster and the multiple seat receiving terminals to obtain multiple seat sound field regions, and collecting data on the multiple seat sound field regions to obtain multiple regional sound field distribution data;

[0028] It is understandable that the execution subject of the present application may be a vehicle interior sound field zoning control device, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0029] Specifically, multiple speakers are deployed in the interior space of the vehicle to obtain a speaker distribution cluster. At the same time, the interior space is partitioned according to the preset layout of multiple seat receiving terminals in the vehicle to ensure that each seat can become an independent sound field area. Through partitioning, the acoustic performance of each area is accurately controlled to meet the auditory needs of different passengers. The position relationship analysis of the speaker distribution cluster and multiple seat sound field areas is performed to determine the distance and relative position between each speaker and each seat, and obtain the position proximity relationship information. The position proximity relationship information helps the system predict and adjust the propagation effect of sound in the car to optimize the auditory experience of each area. According to the obtained position proximity relationship information, a sound pressure level propagation model is constructed for each seat sound field area. The model is based on the physical and acoustic principles of sound propagation and takes into account the path characteristics from the sound source (speaker) to the receiving point (seat), including factors such as sound attenuation, reflection and absorption. Based on the constructed sound pressure level propagation model, sound field distribution data is collected. Including sound pressure level measurement from each speaker to each seat area to ensure that the initial sound field distribution data of each seat sound field area is obtained. The initial sound field distribution data of each seat sound field area is processed by wavelet filtering. Wavelet filtering is an effective signal processing technology that can separate and remove noise components in sound field data, retain key sound field features, and obtain sound field distribution data for multiple areas.

[0030] Step S102, performing sound field distribution feature detection on the sound field distribution data of multiple regions to obtain a sound field distribution feature subset of each seat sound field region;

[0031] Specifically, the sound field propagation distribution characteristics of multiple regional sound field distribution data are analyzed to obtain the sound field propagation distribution characteristics of each seat sound field area. The analysis focuses on understanding how sound propagates in different areas of the vehicle, including the propagation path of the sound, the intensity change and its interaction with the internal structure of the vehicle, to help determine the efficiency and effect of sound propagation. The time series feature extraction of multiple regional sound field distribution data is performed to obtain the time series distribution characteristics of the sound field in each seat sound field area, including the dynamic response of the sound field, such as the rapid change of the sound pressure level and the duration and other parameters. Time series feature extraction can reveal the performance of the sound field on the time axis, which helps to understand the clarity and stability of the audio content, especially in dynamic environments such as the audio performance when the vehicle is driving. At the same time, the antinode and node feature identification of multiple regional sound field distribution data is performed to obtain the sound field wave distribution characteristics of each seat sound field area. Identifying the peak and trough positions in the sound field, which identify the areas with the highest and lowest sound pressure levels, helps to adjust the speaker output and sound field balance. By identifying the antinodes and nodes, the sound field can be adjusted more accurately to make it more evenly distributed inside the vehicle, avoiding over-concentration or blank areas, and optimizing the auditory experience of each passenger. The sound field propagation distribution characteristics, sound field timing distribution characteristics, and sound field wave distribution characteristics are standardized and converted into sets. All complex sound field data are converted into standardized feature subsets to simplify the subsequent sound field control and optimization process. The sound field distribution feature subset of each seat sound field area serves as the basis for optimizing the sound field control strategy, ensuring accurate and personalized acoustic adjustment in the car.

[0032] Step S103, performing multi-dimensional sound field evaluation index calculation and vector mapping on multiple seat sound field areas according to the sound field distribution feature subset to obtain multiple sound field distribution feature vectors, and creating a first sound field partition control parameter set according to the multiple sound field distribution feature vectors and a particle swarm algorithm;

[0033] Specifically, a multi-dimensional evaluation index is calculated for the sound field of the interior space of the vehicle according to the subset of sound field distribution characteristics. The multi-dimensional evaluation includes the uniformity, clarity, dynamic range and spatial resolution of the sound field, etc. These indicators jointly define the acoustic quality of each seat sound field area. The multi-dimensional sound field evaluation results of each seat sound field area are converted into a sound field distribution feature vector. Through vector mapping, various sound field evaluation indicators are converted into a set of numerical vectors, which are used for subsequent parameter analysis and optimization. Multiple control parameter ranges for each seat sound field area are obtained. These parameters include the output power of the speaker, frequency response adjustment, and directivity control of the sound wave. The range of these parameters defines the possibility and flexibility of sound field control. The sound field distribution feature vector and the control parameter range are analyzed and optimized by the particle swarm algorithm. The particle swarm algorithm is an optimization technology based on swarm intelligence, which can find the optimal solution in a multi-dimensional search space. Through algorithm processing, the optimal sound field partition control parameter value is determined for each seat sound field area, and these parameter values ​​are intended to maximize the overall performance of the sound field and the auditory experience of the passengers. According to the control parameter values ​​of each seat sound field area, the first sound field partition control parameter set of the interior space of the vehicle is generated. This parameter set serves as the basis for adjusting and controlling the sound field in each area of ​​the vehicle, ensuring that each area achieves the preset acoustic performance targets.

[0034] Step S104, creating an in-vehicle sound field space propagation network of the vehicle interior space according to the speaker distribution cluster, the plurality of seat receiving terminals and the plurality of seat sound field areas;

[0035] Specifically, multiple first nodes are defined according to the speaker distribution cluster, and these nodes represent the emission points of the sound, that is, the specific locations of the speakers. Multiple second nodes are defined according to multiple seat receiving ends, and these nodes serve as the receiving points of the sound and represent the target positions of the sound field propagation. The sound field propagation direction of each seat sound field area in the car is analyzed to determine the propagation direction of the sound from the first node (speaker) to the second node (seat receiving end), including the straight path and possible reflection path of the sound propagation. The sound field propagation direction is based on the physical acoustic principle and takes into account the specific conditions of the in-car environment, such as interior materials, seat layout and other factors that may affect the propagation of sound waves. According to the sound field propagation direction obtained by the analysis, target propagation paths from multiple first nodes to multiple second nodes are created. These paths constitute the basic architecture of the in-car sound field space propagation network. Each path represents a possible propagation route of the sound field and is also an important basis for subsequent control strategies. Through settings, the distribution and intensity of the sound field in the car are controlled to ensure that each seat area can obtain the best auditory experience. Multiple first nodes and second nodes are used as propagation path nodes, combined with the defined target propagation path, to construct the in-car sound field space propagation network of the vehicle interior space. The network maps the propagation dynamics of the sound field and can also provide necessary data support for sound field control, such as the distance between nodes, the uniformity of sound field coverage, and propagation efficiency.

[0036] Step S105, extracting state features of multiple propagation path nodes in the in-vehicle sound field spatial propagation network to obtain a state feature parameter set of each propagation path node;

[0037] Specifically, the state features of multiple propagation path nodes in the in-vehicle sound field spatial propagation network are extracted. The acoustic data of each node, such as sound pressure level, frequency response, reflection characteristics and absorption rate, are collected. These data together constitute the first characteristic parameter set of each node. The first characteristic parameter of each node is subjected to feature screening. The most critical features for sound field control are identified, and parameters with little or redundant effects on sound field regulation are excluded. The screening is based on the analysis of the impact on sound field propagation efficiency and quality, ensuring that the second characteristic parameter set that is finally retained contains the features that best represent the state of the node sound field. After screening, the second characteristic parameter of each propagation path node is normalized in characteristic scale. The characteristic parameters are standardized to a unified metric to eliminate the dimensional effects between different parameters, making data processing more effective in subsequent sound field control algorithms. The normalized state characteristic parameter set provides a clear and consistent data basis for each node, which is convenient for algorithm analysis and sound field adjustment.

[0038] Step S106 , generating a sound field parameter control execution strategy according to the state characteristic parameter set and performing control parameter optimization on the first sound field partition control parameter set to obtain a second sound field partition control parameter set.

[0039] Specifically, the initial parameter control execution strategy corresponding to each propagation path node is generated according to the state characteristic parameter set. Based on the sound field state data of each node, a preliminary sound field control plan is formulated, including the output level of the speaker, frequency adjustment, and the directivity control of the sound wave. The reward feedback parameter of each strategy is calculated through the reward function. The reward function is designed based on the comparison between the sound field control target and the actual output effect. The main considerations include the integrity of the sound field coverage, the maintenance of sound quality, and the control effect on a specific frequency band. The calculation result of the reward feedback parameter directly reflects the execution effect of each initial control strategy and provides a basis for adjusting these strategies. The initial parameter control execution strategy is adjusted according to the reward feedback parameter to generate a sound field parameter control execution strategy. By modifying the relevant parameters of the sound field control, such as adjusting the sound pressure level setting or changing the propagation angle of the sound wave, it can better adapt to the in-vehicle environment and passenger needs. By applying the adjusted sound field parameter control execution strategy, the first sound field partition control parameter set is optimized and solved to generate the second sound field partition control parameter set. The solution process involves iterative operations of the algorithm model, with the aim of finding a new set of control parameters so that the sound field distribution inside the car achieves the predetermined optimal effect.

[0040] In the embodiment of the present application, by partitioning the sound field based on the speaker distribution cluster and multiple seat receiving ends, the method can accurately control and optimize the sound effects of each area in the car. This fine partitioning enables each seat area to obtain the most suitable audio output, thereby maintaining the consistency of the sound quality of the whole vehicle while also meeting the personalized needs of passengers for the sound environment. Through data collection and sound field distribution feature detection of multiple sound field areas in the car, the system can effectively understand and analyze the acoustic characteristics of each area. Using technologies such as wavelet filtering and sound field propagation feature analysis, the interference of environmental noise can be removed and the accuracy and practicality of sound field data can be improved. The particle swarm algorithm is used to analyze and optimize the sound field distribution feature vector, and the sound field parameters can be automatically adjusted to adapt to the dynamic environmental changes inside the vehicle. The dynamic control supported by this algorithm not only improves the efficiency of sound field control, but also optimizes the sound effect in real time according to the actual situation, ensuring that each passenger can get the best auditory experience. The construction of the in-car sound field space propagation network takes into account the relative position between the speaker and the seat and the sound field propagation path, and effectively integrates the acoustic resources inside the vehicle. By precisely controlling the direction and intensity of sound field propagation, the network ensures the uniformity of sound distribution and the sound quality of the target area, thereby improving the accuracy of in-vehicle sound field zoning control.

[0041] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0042] (1) Deploy multiple speakers in the interior space of the vehicle to obtain a speaker distribution cluster, and partition the interior space of the vehicle into vehicle sound field zones according to multiple preset seat receiving terminals to obtain multiple seat sound field areas;

[0043] (2) Analyze the position relationship between the speaker distribution cluster and multiple seat sound field areas to obtain position proximity relationship information;

[0044] (3) Construct a sound pressure level propagation model for each seat sound field area based on the position proximity relationship information;

[0045] (4) Based on the sound pressure level propagation model, the sound field distribution data of the speaker distribution cluster and multiple seat sound field areas are collected to obtain the initial sound field distribution data of each seat sound field area;

[0046] (5) Perform wavelet filtering on the initial sound field distribution data of each seat sound field area to obtain sound field distribution data of multiple areas.

[0047] Specifically, multiple speakers are deployed in the interior space of the vehicle to form a speaker distribution cluster. The speakers are arranged in a specific configuration, such as on the roof, door, and dashboard, to ensure that the sound can cover the entire cabin in all directions. According to the layout of the seats in the car, the space corresponding to each seat is divided into an independent sound field area. The partition design enables the system to adjust the acoustic output according to the specific needs of each area to meet the auditory preferences of different passengers. The position relationship analysis of the speaker distribution cluster and each seat sound field area is performed. The distance and relative position between each speaker and each seat are identified, and the position proximity relationship information is helpful for the subsequent construction of the sound pressure level propagation model. For example, the closer the speaker is to a seat, the greater its impact on the sound field of the seat, while the farther the distance, the smaller the impact. The sound may need to compensate for the energy loss during the propagation process through higher output power or signal enhancement. According to the position proximity relationship information, a sound pressure level propagation model is constructed for each seat sound field area. The model is based on acoustic principles and considers the path characteristics of sound waves propagating from the speaker to the receiving end (seat area), including factors such as reflection, absorption, and diffraction of sound waves in the car. For example, when sound waves are transmitted from the front speakers to the back seats, they may be absorbed by the materials in the car and hindered by the seat layout, resulting in reduced sound intensity and clarity. Based on the sound pressure level propagation model, the sound field distribution data of the speaker distribution cluster and multiple seat sound field areas are collected. By installing measurement equipment such as sound level meters in each seat sound field area, the initial sound field distribution data of each area is obtained. These data reflect how the sound pressure level of each seat responds to sounds of different frequencies and how the sound is distributed in space in the actual vehicle environment. The initial sound field distribution data of each seat sound field area is processed by wavelet filtering to obtain multiple regional sound field distribution data. Wavelet filtering can decompose the sound field data, remove non-target signals, such as road noise, wind noise and other background sounds, and leave only clear audio signals.

[0048] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0049] (1) Analyze the sound field propagation distribution characteristics of the sound field distribution data of multiple regions respectively to obtain the sound field propagation distribution characteristics of each seat sound field region;

[0050] (2) Extracting time series features from the sound field distribution data of multiple regions respectively to obtain the time series distribution features of the sound field in each seat sound field region;

[0051] (3) Identify the antinode and node characteristics of the sound field distribution data of multiple regions respectively to obtain the sound field wave distribution characteristics of each seat sound field region;

[0052] (4) The sound field propagation distribution characteristics, the sound field time series distribution characteristics, and the sound field wave distribution characteristics are standardized and transformed into sets to obtain a subset of the sound field distribution characteristics of each seat sound field area.

[0053] Specifically, the sound field propagation distribution characteristics are analyzed to understand how sound propagates in the in-vehicle environment, including the propagation path, propagation speed, and degree of sound attenuation from the speaker to each seat area. For example, the speakers may be located in the door panels, trunk, and roof, and the sound needs to pass through a variety of interior materials such as leather, fabric, and hard plastic. The different acoustic properties of these materials will affect the propagation path and attenuation of the sound waves. By analyzing these characteristics, it is determined which areas may have uneven sound coverage problems so that further adjustments can be made. The time series feature extraction is performed on the sound field distribution data of each area, and the change of the sound field data over time is analyzed to capture the dynamic characteristics of the sound field. For example, when the vehicle starts or accelerates, the background noise level in the car will change, which may affect the clarity and volume perception of music playback. Through the extraction of time series features, a detailed picture of how the sound field of each seat area changes over time is obtained, so as to design a sound field control strategy that can adapt to changes. The antinode and node feature identification of the sound field distribution data of multiple areas is performed separately to understand the distribution pattern of sound waves in the in-vehicle space. During the propagation of sound waves, antinodes (the highest sound pressure points) and nodes (the lowest sound pressure points) are formed in certain areas. Analyzing these characteristics helps adjust the output of the speakers to ensure uniform sound distribution. For example, if nodes frequently appear in a certain seating area, the angle or output of adjacent speakers may need to be adjusted to eliminate these dead zones and ensure that every passenger has a balanced auditory experience. The sound field propagation distribution characteristics, sound field time series distribution characteristics, and sound field wave distribution characteristics are standardized and transformed into sets to ensure that the comparison and comprehensive application of different characteristics are consistent and comparable in technical processing. Standardization ensures that all data are operated on the same scale, eliminating differences in units and dimensions, while set transformation converts the processed data into feature subsets, which are directly used in the algorithms of the sound field control system to achieve precise sound field adjustment.

[0054] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0055] (1) Calculate the multi-dimensional sound field evaluation index of the vehicle interior space based on the sound field distribution feature subset to obtain the multi-dimensional sound field evaluation of each seat sound field area;

[0056] (2) Performing vector mapping on the multi-dimensional sound field evaluation of each seat sound field area to obtain multiple sound field distribution feature vectors;

[0057] (3) Obtain multiple control parameter ranges for each seat sound field area;

[0058] (4) using a particle swarm algorithm, performing sound field partition control parameter analysis on multiple seat sound field areas according to multiple sound field distribution feature vectors and multiple control parameter ranges, and obtaining a sound field partition control parameter value for each seat sound field area;

[0059] (5) Generate a first sound field partition control parameter set for the vehicle interior space according to the sound field partition control parameter value of each seat sound field area.

[0060] Specifically, the multi-dimensional sound field evaluation index of the vehicle interior space is calculated according to the sound field distribution feature subset to ensure that each seat sound field area can meet specific acoustic performance requirements. The calculation includes evaluating indicators such as sound pressure level, sound clarity, timbre balance and sound positioning accuracy of each area. For example, it is ensured that the front seat area has high standards in dialogue clarity, while the back seat may pay more attention to the surround sound effect of music playback. The multi-dimensional sound field evaluation of each seat sound field area is vector mapped to obtain multiple sound field distribution feature vectors. The multi-dimensional evaluation of the sound field is quantified into a set of operational values ​​so that these data can be used by subsequent optimization algorithms. Multiple control parameter ranges are obtained for each seat sound field area. These parameters include the output power of the speaker, the frequency adjustment range, the delay setting and possible phase control. The purpose of determining the parameter range is to provide operational flexibility for the sound field control and ensure that the implementation of the control can be carried out within the technically feasible range. Through the particle swarm algorithm, the sound field partition control parameter analysis of multiple seat sound field areas is performed according to multiple sound field distribution feature vectors and multiple control parameter ranges to obtain the sound field partition control parameter value of each seat sound field area. Particle swarm algorithm is an effective optimization tool for dealing with complex optimization problems. Particle swarm algorithm simulates the search behavior of a group of particles (possible solution set) in the solution space and finds the optimal solution through iteration. In this process, each particle represents a set of possible sound field control parameters. The particles update their own positions by tracking the historical optimal solutions of individuals and groups, thereby approaching the global optimal solution. The first sound field partition control parameter set of the vehicle interior space is generated according to the sound field partition control parameter values ​​of each seat sound field area. This parameter set includes all necessary acoustic adjustment values, such as speaker output level, frequency distribution, sound field balance and timing settings.

[0061] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0062] (1) defining a plurality of first nodes according to a speaker distribution cluster, and defining a plurality of second nodes according to a plurality of seat receiving terminals;

[0063] (2) Analyze the sound field propagation direction of multiple seat sound field areas to obtain multiple sound field propagation directions;

[0064] (3) creating a plurality of target propagation paths of a plurality of first nodes and a plurality of second nodes according to a plurality of sound field propagation directions;

[0065] (4) The plurality of first nodes and the plurality of second nodes are used as a plurality of propagation path nodes, and an in-vehicle sound field spatial propagation network of the interior space of the vehicle is created according to the plurality of propagation path nodes and the plurality of target propagation paths.

[0066] Specifically, multiple first nodes are defined according to the speaker distribution cluster, and multiple second nodes are defined according to multiple seat receiving ends. The speakers and seat receiving ends serve as the starting point and receiving point of the sound field propagation, respectively. For example, the speakers are usually installed in the door, dashboard, and possible rear seat area. The position of each speaker is regarded as a first node, and each seat is defined as a second node. The sound field propagation direction analysis is performed on multiple seat sound field areas inside the vehicle. Consider how the sound wave propagates from each speaker (first node) to each seat receiving end (second node). The purpose of the sound field propagation direction analysis is to determine the most effective path for sound propagation, while taking into account various acoustic obstacles in the car, such as seat layout, interior materials, and other structural features in the car. For example, in the analysis, it may be found that when the sound wave propagates from the front speaker to the back seat, the sound wave path will be significantly attenuated or deviate from the expected path due to the obstruction of the back seat backrest. After obtaining the sound field propagation direction, the target propagation path between multiple first nodes and multiple second nodes is created according to these directions. The path is based on the straight-line propagation model and takes into account the possible reflection and refraction paths of the sound wave to ensure that the design can adapt to the complex in-vehicle environment. For example, the propagation of sound waves can be improved by adding sound-absorbing materials in certain specific locations or using reflective panels. Multiple first nodes and multiple second nodes are integrated as multiple propagation path nodes, and an in-vehicle sound field spatial propagation network for the interior space of the vehicle is created based on these nodes and their target propagation paths. The network is a complex system that includes all possible sound field propagation paths, designed to optimize the acoustic performance of the entire vehicle. Use acoustic simulation software for simulation and verification to ensure the effectiveness of the design. The network design also needs to dynamically adapt to different driving environments, such as adjusting the output of the internal speakers to maintain the sound quality in the car when the external noise is high during high-speed driving.

[0067] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0068] (1) extracting state features of multiple propagation path nodes in the in-vehicle sound field spatial propagation network to obtain multiple first feature parameters of each propagation path node;

[0069] (2) performing feature screening on a plurality of first feature parameters of each propagation path node to obtain a plurality of second feature parameters of each propagation path node;

[0070] (3) Normalizing the characteristic scales of the multiple second characteristic parameters of each propagation path node respectively to obtain a state characteristic parameter set of each propagation path node.

[0071] Specifically, the state features of multiple propagation path nodes in the spatial propagation network of the in-vehicle sound field are extracted to obtain multiple first feature parameters of each propagation path node. The sound data at each node from the speaker (sound source) to the seat receiving end (listener position) is collected and analyzed. The sound data of each node includes features such as sound pressure level, frequency response, echo, noise level, and possible sound distortion. For example, if a speaker node located behind the driver's seat is considered, its first feature parameter may include the linear sound pressure propagation level from the position to the driver's seat, and the sound pressure change caused by the reflection of the material in the car. The multiple first feature parameters of each propagation path node are feature screened to select the most influential feature for the sound field regulation to obtain the second feature parameter. The purpose of feature screening is to simplify the model and remove redundant data or data that has little effect on the final sound field regulation through statistical analysis methods such as principal component analysis or correlation analysis. The screening process considers the relationship between each feature and the sound field quality (such as clarity, uniformity, etc.), and only retains features that contribute significantly to improving the sound field. For example, the sound pressure level from the speaker to different seats may be a key feature because it directly affects the auditory experience of each passenger. The second characteristic parameter of each propagation path node is scaled normalized. The characteristic values ​​are converted to the same scale for comparison and further analysis. This usually involves standardizing the characteristic data so that its mean is zero and the standard deviation is one, or scaling the data to a range of 0 to 1 to obtain a state characteristic parameter set, which provides a unified measurement standard for the establishment of the sound field model and parameter optimization. For example, the output of the speaker is adjusted according to the state characteristic parameter set obtained from each node to optimize the directivity and distribution of the sound wave to ensure that each seat can get the best listening experience.

[0072] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0073] (1) Generate the initial parameter control execution strategy corresponding to each propagation path node based on the state characteristic parameter set, and calculate the reward feedback parameter of the initial parameter control execution strategy through the reward function;

[0074] (2) Adjust the initial parameter control execution strategy according to the reward feedback parameters to generate the sound field parameter control execution strategy;

[0075] (3) The control parameters of the first sound field partition control parameter set are optimized and solved through the sound field parameter control execution strategy to obtain the second sound field partition control parameter set.

[0076] Specifically, the initial parameter control execution strategy corresponding to each propagation path node is generated according to the state characteristic parameter set. The initial strategy includes, for example, the output power of the speaker, frequency response adjustment, sound delay, and spatial sound effect configuration. For example, assuming that there are ten speaker positions as sound nodes, each speaker needs to adjust its output settings according to its distance to the listener, reflections, and other acoustic obstacles. The initial settings are based on theoretical calculations and preliminary simulations, with the goal of covering as wide a range of hearing uniformity and clarity as possible. In order to evaluate and optimize the initial parameter control execution strategy, the reward feedback parameters of the strategy are calculated through a reward function. The reward function is usually based on multiple preset performance indicators, such as the uniformity of the sound field, the suitability of the sound pressure level, and the passenger's hearing satisfaction. By applying the initial strategy in actual vehicle operation or in advanced simulation tests, the reward function can provide feedback based on the deviation of the actual performance from the expected target. For example, if the sound distribution in a certain seating area is too concentrated or deviates from the target hearing area, the reward function will give a low score for this, indicating that adjustment is required. The initial parameter control execution strategy is adjusted according to the reward feedback parameters. Analyze the deficiencies pointed out in the reward feedback and adjust the relevant sound field parameters to better meet the sound field quality standards. Strategy adjustments may include modifying the output power of the speaker, adjusting the frequency distribution, changing the time delay of sound propagation, or optimizing the sound field effect settings, such as the intensity and direction of 3D sound effects. The first sound field partition control parameter set is optimized and solved through the sound field parameter control execution strategy to obtain the second sound field partition control parameter set. Use advanced optimization algorithms, such as genetic algorithms or deep learning models, to ensure that the obtained parameters maximize the sound field performance. For example, after multiple iterative tests, it may be found that the low-frequency response of the rear seat area needs to be enhanced to provide a richer sound quality experience, and the corresponding parameter adjustments will be integrated into the final control parameter set.

[0077] The above describes the method for controlling the sound field in a vehicle by partition in the embodiment of the present application. The following describes the device for controlling the sound field in a vehicle by partition in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the in-vehicle sound field partition control device includes:

[0078] The acquisition module 201 is used to partition the interior space of the vehicle into vehicle sound fields based on the speaker distribution cluster and the multiple seat receiving terminals to obtain multiple seat sound field areas, and to collect data on the multiple seat sound field areas to obtain multiple area sound field distribution data;

[0079] A detection module 202 is used to perform sound field distribution feature detection on the sound field distribution data of multiple regions to obtain a sound field distribution feature subset of each seat sound field region;

[0080] A calculation module 203 is used to calculate multi-dimensional sound field evaluation indicators and vector mapping for multiple seat sound field areas according to the sound field distribution feature subset, obtain multiple sound field distribution feature vectors, and create a first sound field partition control parameter set according to the multiple sound field distribution feature vectors and a particle swarm algorithm;

[0081] A creation module 204 is used to create an in-vehicle sound field spatial propagation network of the vehicle interior space according to the speaker distribution cluster, the plurality of seat receiving terminals and the plurality of seat sound field areas;

[0082] An extraction module 205 is used to extract state features of multiple propagation path nodes in the in-vehicle sound field spatial propagation network to obtain a state feature parameter set of each propagation path node;

[0083] The solution module 206 is used to generate a sound field parameter control execution strategy according to the state characteristic parameter set and perform control parameter optimization on the first sound field partition control parameter set to obtain a second sound field partition control parameter set.

[0084] Through the collaboration of the above components, the method can accurately control and optimize the sound effects of various areas in the car through the sound field partitioning based on the speaker distribution cluster and multiple seat receiving ends. This fine partitioning enables each seat area to obtain the most suitable audio output, thereby maintaining the consistency of the sound quality of the whole vehicle while also meeting the personalized needs of passengers for the sound environment. Through data collection and sound field distribution feature detection of multiple sound field areas in the car, the system can effectively understand and analyze the acoustic characteristics of each area. Using technologies such as wavelet filtering and sound field propagation feature analysis, the interference of environmental noise can be removed and the accuracy and practicality of sound field data can be improved. The particle swarm algorithm is used to analyze and optimize the sound field distribution feature vector, and the sound field parameters can be automatically adjusted to adapt to the dynamic environmental changes inside the vehicle. The dynamic control supported by this algorithm not only improves the efficiency of sound field control, but also optimizes the sound effect in real time according to the actual situation, ensuring that every passenger can get the best auditory experience. The construction of the in-vehicle sound field space propagation network takes into account the relative position between the speaker and the seat and the sound field propagation path, and effectively integrates the acoustic resources inside the vehicle. By precisely controlling the direction and intensity of sound field propagation, the network ensures the uniformity of sound distribution and the sound quality of the target area, thereby improving the accuracy of in-vehicle sound field zoning control.

[0085] The present application also provides a car speaker, which is used to execute the steps of the in-vehicle sound field zoning control method in the above embodiments.

[0086] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the method for controlling the zoning of the in-vehicle sound field.

[0087] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0088] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0089] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for controlling the zoning of the sound field in a vehicle, characterized in that: The method for controlling the zoning of the in-vehicle sound field comprises: Based on the speaker distribution cluster and multiple seat receiving terminals, the vehicle interior space is partitioned into in-vehicle sound field zones to obtain multiple seat sound field areas, and data is collected on the multiple seat sound field areas to obtain multiple regional sound field distribution data; specifically comprising: deploying multiple speakers in the vehicle interior space to obtain a speaker distribution cluster, and partitioning the vehicle interior space into in-vehicle sound field zones according to a preset multiple seat receiving terminals to obtain multiple seat sound field areas; performing position relationship analysis on the speaker distribution cluster and the multiple seat sound field areas to obtain position proximity relationship information; constructing a sound pressure level propagation model for each seat sound field area according to the position proximity relationship information; based on the sound pressure level propagation model, collecting sound field distribution data on the speaker distribution cluster and the multiple seat sound field areas to obtain initial sound field distribution data for each seat sound field area; performing wavelet filtering on the initial sound field distribution data of each seat sound field area to obtain multiple regional sound field distribution data; Performing sound field distribution feature detection on the sound field distribution data of the multiple regions to obtain a sound field distribution feature subset of each seat sound field region; specifically comprising: performing sound field propagation distribution feature analysis on the sound field distribution data of the multiple regions to obtain sound field propagation distribution features of each seat sound field region; performing time series feature extraction on the sound field distribution data of the multiple regions to obtain sound field time series distribution features of each seat sound field region; performing antinode and node feature identification on the sound field distribution data of the multiple regions to obtain sound field wave distribution features of each seat sound field region; performing standardization processing and set conversion on the sound field propagation distribution features, the sound field time series distribution features and the sound field wave distribution features to obtain a sound field distribution feature subset of each seat sound field region; According to the sound field distribution feature subset, multi-dimensional sound field evaluation index calculation and vector mapping are performed on the multiple seat sound field areas to obtain multiple sound field distribution feature vectors, and a first sound field partition control parameter set is created according to the multiple sound field distribution feature vectors and a particle swarm algorithm; specifically comprising: according to the sound field distribution feature subset, multi-dimensional sound field evaluation index calculation is performed on the vehicle interior space to obtain a multi-dimensional sound field evaluation of each seat sound field area; vector mapping is performed on the multi-dimensional sound field evaluation of each seat sound field area to obtain multiple sound field distribution feature vectors; multiple control parameter ranges are obtained for each seat sound field area; through a particle swarm algorithm, sound field partition control parameter analysis is performed on the multiple seat sound field areas according to the multiple sound field distribution feature vectors and the multiple control parameter ranges to obtain a sound field partition control parameter value for each seat sound field area; and a first sound field partition control parameter set for the vehicle interior space is generated according to the sound field partition control parameter value of each seat sound field area; An in-vehicle sound field space propagation network for the interior space of the vehicle is created according to the speaker distribution cluster, the multiple seat receiving ends and the multiple seat sound field areas; specifically comprising: defining multiple first nodes according to the speaker distribution cluster, and defining multiple second nodes according to the multiple seat receiving ends; performing sound field propagation direction analysis on the multiple seat sound field areas to obtain multiple sound field propagation directions; creating multiple target propagation paths of the multiple first nodes and the multiple second nodes according to the multiple sound field propagation directions; using the multiple first nodes and the multiple second nodes as multiple propagation path nodes, and creating an in-vehicle sound field space propagation network for the interior space of the vehicle according to the multiple propagation path nodes and the multiple target propagation paths; wherein, according to the sound field propagation directions obtained by the analysis, target propagation paths from the multiple first nodes to the multiple second nodes are created, and the target propagation paths constitute the basic architecture of the in-vehicle sound field space propagation network, and each target propagation path represents a possible propagation route of the sound field; The state feature extraction is performed on multiple propagation path nodes in the in-vehicle sound field space propagation network to obtain a state feature parameter set of each propagation path node; specifically comprising: extracting the state feature of multiple propagation path nodes in the in-vehicle sound field space propagation network to obtain multiple first feature parameters of each propagation path node; performing feature screening on the multiple first feature parameters of each propagation path node to obtain multiple second feature parameters of each propagation path node; and performing feature scale normalization on the multiple second feature parameters of each propagation path node to obtain a state feature parameter set of each propagation path node; Generate a sound field parameter control execution strategy according to the state characteristic parameter set and optimize and solve the control parameters of the first sound field partition control parameter set to obtain a second sound field partition control parameter set; specifically including: generate an initial parameter control execution strategy corresponding to each propagation path node according to the state characteristic parameter set, and calculate the reward feedback parameter of the initial parameter control execution strategy through a reward function; adjust the strategy of the initial parameter control execution strategy according to the reward feedback parameter to generate a sound field parameter control execution strategy; optimize and solve the control parameters of the first sound field partition control parameter set through the sound field parameter control execution strategy to obtain a second sound field partition control parameter set.

2. A vehicle interior sound field zoning control device, characterized in that: The in-vehicle sound field partition control device comprises: A collection module, for partitioning the interior space of a vehicle into in-vehicle sound fields based on a speaker distribution cluster and a plurality of seat receiving terminals, obtaining a plurality of seat sound field areas, and collecting data on the plurality of seat sound field areas to obtain a plurality of regional sound field distribution data; specifically comprising: deploying a plurality of speakers in the interior space of a vehicle to obtain a speaker distribution cluster, and partitioning the interior space of the vehicle into in-vehicle sound fields according to a plurality of preset seat receiving terminals to obtain a plurality of seat sound field areas; performing position relationship analysis on the speaker distribution cluster and the plurality of seat sound field areas to obtain position proximity relationship information; constructing a sound pressure level propagation model for each seat sound field area respectively according to the position proximity relationship information; based on the sound pressure level propagation model, collecting sound field distribution data on the speaker distribution cluster and the plurality of seat sound field areas to obtain initial sound field distribution data for each seat sound field area; performing wavelet filtering on the initial sound field distribution data of each seat sound field area respectively to obtain a plurality of regional sound field distribution data; A detection module is used to detect the sound field distribution characteristics of the multiple regional sound field distribution data to obtain a subset of the sound field distribution characteristics of each seat sound field area; specifically comprising: respectively analyzing the sound field propagation distribution characteristics of the multiple regional sound field distribution data to obtain the sound field propagation distribution characteristics of each seat sound field area; respectively extracting the time series characteristics of the multiple regional sound field distribution data to obtain the time series distribution characteristics of the sound field of each seat sound field area; respectively identifying the antinode and node characteristics of the multiple regional sound field distribution data to obtain the sound field wave distribution characteristics of each seat sound field area; performing standardization processing and set conversion on the sound field propagation distribution characteristics, the sound field time series distribution characteristics and the sound field wave distribution characteristics to obtain a subset of the sound field distribution characteristics of each seat sound field area; A calculation module, used for performing multi-dimensional sound field evaluation index calculation and vector mapping on the multiple seat sound field areas according to the sound field distribution feature subset to obtain multiple sound field distribution feature vectors, and creating a first sound field partition control parameter set according to the multiple sound field distribution feature vectors and a particle swarm algorithm; specifically comprising: performing multi-dimensional sound field evaluation index calculation on the vehicle interior space according to the sound field distribution feature subset to obtain a multi-dimensional sound field evaluation of each seat sound field area; performing vector mapping on the multi-dimensional sound field evaluation of each seat sound field area respectively to obtain multiple sound field distribution feature vectors; obtaining multiple control parameter ranges for each seat sound field area; performing sound field partition control parameter analysis on the multiple seat sound field areas according to the multiple sound field distribution feature vectors and the multiple control parameter ranges through a particle swarm algorithm to obtain a sound field partition control parameter value for each seat sound field area; generating a first sound field partition control parameter set for the vehicle interior space according to the sound field partition control parameter value of each seat sound field area; A creation module, used for creating an in-vehicle sound field space propagation network of the vehicle interior space according to the speaker distribution cluster, the multiple seat receiving ends and the multiple seat sound field areas; specifically comprising: defining multiple first nodes according to the speaker distribution cluster, and defining multiple second nodes according to the multiple seat receiving ends; performing sound field propagation direction analysis on the multiple seat sound field areas to obtain multiple sound field propagation directions; creating multiple target propagation paths of the multiple first nodes and the multiple second nodes according to the multiple sound field propagation directions; using the multiple first nodes and the multiple second nodes as multiple propagation path nodes, and creating an in-vehicle sound field space propagation network of the vehicle interior space according to the multiple propagation path nodes and the multiple target propagation paths; wherein, according to the sound field propagation directions obtained by the analysis, target propagation paths from the multiple first nodes to the multiple second nodes are created, and the target propagation paths constitute the basic architecture of the in-vehicle sound field space propagation network, and each target propagation path represents a possible propagation route of the sound field; The extraction module is used to extract state features of multiple propagation path nodes in the in-vehicle sound field space propagation network to obtain a state feature parameter set of each propagation path node; specifically comprising: extracting state features of multiple propagation path nodes in the in-vehicle sound field space propagation network to obtain multiple first feature parameters of each propagation path node; performing feature screening on the multiple first feature parameters of each propagation path node to obtain multiple second feature parameters of each propagation path node; and performing feature scale normalization on the multiple second feature parameters of each propagation path node to obtain a state feature parameter set of each propagation path node; A solution module is used to generate a sound field parameter control execution strategy according to the state characteristic parameter set and optimize and solve the control parameters of the first sound field partition control parameter set to obtain a second sound field partition control parameter set; specifically comprising: generating an initial parameter control execution strategy corresponding to each propagation path node according to the state characteristic parameter set, and calculating a reward feedback parameter of the initial parameter control execution strategy through a reward function; adjusting the initial parameter control execution strategy according to the reward feedback parameter to generate a sound field parameter control execution strategy; optimizing and solving the control parameters of the first sound field partition control parameter set through the sound field parameter control execution strategy to obtain a second sound field partition control parameter set.

3. A car speaker, characterized in that: The car speaker is used to execute the in-vehicle sound field zoning control method as claimed in claim 1.

4. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the method for controlling the in-vehicle sound field by zoning as claimed in claim 1 is implemented.

Images