Vehicle trunk control method, device, vehicle and storage medium

Abstract

The present application relates to a control method, device, vehicle and storage medium for a vehicle trunk, wherein the method comprises: obtaining a sound signal collected by at least one sound collection device in a current collection period; performing data processing on each frame of the audio signal in the sound signal to obtain a corresponding time domain signal and frequency domain signal, and calculating the time-frequency domain features of the sound signal based on the time domain signal and the frequency domain signal, so as to use the time-frequency domain features to determine whether the sound signal meets a preset knocking judgment condition; if the preset knocking judgment condition is met, extracting the knocking signal from the sound signal, and inputting the knocking signal into a pre-built knocking model to output a vehicle trunk control instruction, so as to open the vehicle trunk when the trunk control instruction is an opening instruction. The embodiment of the present application does not require a complex opening action, has a higher action safety, conforms to the user's behavioral rules for the door opening action, reduces the probability of false touches, and takes into account both convenience and safety.

Description

Technical Field

[0001] The present application relates to the field of intelligent automobile technology, and in particular to a control method, device, vehicle and storage medium for a vehicle trunk. Background Art

[0002] With the upgrade and iteration of the software and hardware of new energy vehicles, the way to open the trunk is also gradually being optimized. From manual opening at the beginning, to opening with the car key, to opening by kicking near the trunk, to voice wake-up and voice command opening, all of these reflect the automobile manufacturers' exploration of products, exploration of intelligence and exploration of consumer convenience.

[0003] However, the method of opening the trunk in the related art does not conform to the user's first reaction to opening the trunk, and kicking the trunk may easily cause unstable standing and falling, and voice wake-up may be easily exploited by others to cause safety problems, etc., which needs to be improved. Summary of the invention

[0004] The present application provides a control method, device, vehicle and storage medium for a vehicle trunk to solve the technical problems in the related art that the trunk opening method does not conform to the user's behavior pattern, there are certain safety hazards when the user operates, and it is easy to cause damage to the outside of the vehicle.

[0005] A first aspect of the present application provides a method for controlling a vehicle trunk, wherein a sound collection device for waking up an external function is provided at the rear of the vehicle, wherein the method comprises the following steps: obtaining a sound signal collected by at least one sound collection device in a current collection period; performing data processing on each frame of an audio signal in the sound signal to obtain a corresponding time domain signal and a frequency domain signal, and calculating the time and frequency domain features of the sound signal based on the time domain signal and the frequency domain signal, so as to use the time and frequency domain features to determine whether the sound signal meets a preset knock judgment condition; if the preset knock judgment condition is met, extracting a knock signal from the sound signal, and inputting the knock signal into a pre-constructed knock model to output a trunk control instruction of the vehicle, so as to open the trunk of the vehicle when the trunk control instruction is an opening instruction.

[0006] According to the above-mentioned technical means, the embodiment of the present application can process the sound signal collected by the sound collection device, so as to determine whether the sound signal meets the preset knocking judgment condition, and after determining that the judgment condition is met, further determine whether it is the control intention to open the trunk according to the knocking data, thereby realizing the automatic opening of the trunk of the vehicle without the need for complicated opening actions, with higher action safety, and in line with the user's behavioral rules for door opening actions, and reduce the probability of accidental touch, while taking into account convenience and safety.

[0007] Optionally, in one embodiment of the present application, the data processing of each frame of audio signal in the sound signal includes: framing and windowing the sound signal to obtain the time domain signal; and performing Fourier transform on each frame signal in the time domain signal to obtain the frequency domain signal.

[0008] According to the above technical means, the embodiment of the present application can process the sound signal to facilitate subsequent sound recognition.

[0009] Optionally, in one embodiment of the present application, the time-frequency domain features of the sound signal are calculated based on the time domain signal and the frequency domain signal, including: counting the peak amplitude features and the dynamic range features of the time domain signal; judging whether a knock signal exists in the audio signal based on the peak amplitude features and the dynamic range features; if the knock signal exists in the audio signal, calculating the time-frequency domain features based on the frequency domain signal.

[0010] According to the above technical means, the embodiment of the present application can perform feature extraction on the processed sound signal to distinguish between environmental noise and knocking sound.

[0011] Optionally, in one embodiment of the present application, before the knock signal is input into a pre-constructed knock model, it also includes: performing signal processing on the human knock sample signal and the car knock sample signal in the sample training set respectively to obtain a first training knock sample signal and a second training knock sample signal that meet a preset energy condition; constructing an initial knock model using a Gaussian mixture model, the first training knock sample signal and the second training knock sample signal, and gradually iterating the initial knock model through an EM clustering algorithm and a sample test set to obtain the knock model.

[0012] According to the above technical means, the embodiment of the present application can construct an initial knocking model, and train the initial knocking model through sample data to obtain a knocking model that can be used to identify the probability of knocking.

[0013] Optionally, in one embodiment of the present application, the knock signal is input into a pre-built knock model to output the trunk control instruction of the vehicle, including: inputting the knock signal into the knock model to obtain a knock probability; if the knock probability is greater than a preset probability threshold, determining that the trunk control instruction is the opening instruction.

[0014] According to the above-mentioned technical means, the embodiment of the present application can use the knocking model to secondary judge the knocking signal in the sound signal, reduce the probability of misrecognition, and thus ensure the property safety of the user.

[0015] A second aspect of the present application provides a control device for a vehicle trunk, wherein a sound collection device for waking up an external function is provided at the rear of the vehicle, wherein the device comprises: an acquisition module, for acquiring a sound signal collected by at least one sound collection device in a current collection period; a judgment module, for performing data processing on each frame of an audio signal in the sound signal to obtain a corresponding time domain signal and a frequency domain signal, and calculating the time and frequency domain features of the sound signal based on the time domain signal and the frequency domain signal, so as to use the time and frequency domain features to judge whether the sound signal satisfies a preset knock judgment condition; a control module, for extracting a knock signal from the sound signal when the preset knock judgment condition is met, and inputting the knock signal into a pre-constructed knock model to output a trunk control instruction of the vehicle, so as to open the trunk of the vehicle when the trunk control instruction is an opening instruction.

[0016] Optionally, in one embodiment of the present application, the judgment module includes: a processing unit, used to frame and window the sound signal to obtain the time domain signal; and a first calculation unit, used to perform Fourier transform on each frame signal in the time domain signal to obtain the frequency domain signal.

[0017] Optionally, in one embodiment of the present application, the judgment module includes: a statistical unit, used to count the peak amplitude characteristics and dynamic range characteristics of the time domain signal; a judgment unit, used to judge whether there is a knock signal in the audio signal based on the peak amplitude characteristics and dynamic range characteristics; a second calculation unit, used to calculate the time-frequency domain characteristics based on the frequency domain signal when the knock signal exists in the audio signal.

[0018] Optionally, in one embodiment of the present application, it also includes: a signal processing module, used to perform signal processing on the human knocking sample signals and the vehicle knocking sample signals in the sample training set respectively, to obtain the first training knocking sample signal and the second training knocking sample signal that meet the preset energy conditions; a construction module, used to use the Gaussian mixture model, the first training knocking sample signal and the second training knocking sample signal to construct an initial knocking model, and gradually iterate the initial knocking model through the EM clustering algorithm and the sample test set to obtain the knocking model.

[0019] Optionally, in one embodiment of the present application, the control module includes: an input unit, used to input the knock signal into the knock model to obtain a knock probability; and a determination unit, used to determine that the trunk control instruction is the opening instruction when the knock probability is greater than a preset probability threshold.

[0020] A third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle trunk control method as described in the above embodiment.

[0021] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the vehicle trunk control method as described in the above embodiment.

[0022] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned vehicle trunk control method.

[0023] Beneficial effects of the embodiments of the present application:

[0024] (1) The embodiment of the present application can process the sound signal collected by the sound collection device to determine whether the sound signal meets the preset knocking judgment condition, and after determining that the judgment condition is met, further determine whether it is a control intention to open the trunk based on the knocking data, thereby realizing the automatic opening of the vehicle trunk without the need for complex opening actions, with higher action safety, and in line with the user's behavioral rules for door opening actions, and reduce the probability of accidental touch, while taking into account convenience and safety.

[0025] (2) The embodiment of the present application can process the sound signal and perform feature extraction on the processed sound signal to distinguish between environmental noise and knocking sound.

[0026] (3) The embodiment of the present application can use the knocking model to secondary judge the knocking signal in the sound signal, thereby reducing the probability of misidentification and thus protecting the user's property safety.

[0027] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0029] Figure 1 A flowchart of a vehicle trunk control method provided according to an embodiment of the present application;

[0030] Figure 2 It is a schematic diagram of the principle of a method for controlling a vehicle trunk according to an embodiment of the present application;

[0031] Figure 3is a flow chart of a method for controlling a vehicle trunk according to an embodiment of the present application;

[0032] Figure 4 The figure is a schematic diagram of the construction principle of the knocking model according to one embodiment of the present application.

[0033] Figure 5 It is a structural schematic diagram of a control device for a vehicle trunk provided according to an embodiment of the present application;

[0034] Figure 6 It is a schematic diagram of the structure of a vehicle provided according to an embodiment of the present application.

[0035] Among them, 10 is a control device for the vehicle trunk, 100 is an acquisition module, 200 is a judgment module, 300 is a control module; 601 is a memory, 602 is a processor, and 603 is a communication interface. DETAILED DESCRIPTION

[0036] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0037] The following describes the control method, device, vehicle and storage medium of the vehicle trunk of the embodiment of the present application with reference to the accompanying drawings. In view of the technical problems that the trunk opening method does not conform to the user behavior law in the related art mentioned in the above background technology, there are certain safety hazards when the user operates, and it is easy to cause damage to the exterior of the vehicle, the present application provides a control method for the vehicle trunk, in which the sound signal collected by the sound collection device can be processed to determine whether the sound signal meets the preset knocking judgment condition, and after determining that the judgment condition is met, it is further determined whether it is the control intention to open the trunk according to the knocking data, thereby realizing the automatic opening of the vehicle trunk, without the need for complex opening actions, the action safety is higher, and it conforms to the user's behavioral law for the door opening action, and reduces the probability of false touch, while taking into account convenience and safety. Thus, the technical problems that the trunk opening method does not conform to the user behavior law in the related art, there are certain safety hazards when the user operates, and it is easy to cause damage to the exterior of the vehicle are solved.

[0038] Specifically, Figure 1 A schematic flow chart of a method for controlling a vehicle trunk provided in an embodiment of the present application.

[0039] like Figure 1 As shown, the control method of the vehicle trunk, the rear of the vehicle is provided with a sound collection device for waking up the vehicle function outside the vehicle, wherein the method comprises the following steps:

[0040] In step S101, a sound signal collected by at least one sound collecting device in a current collecting period is obtained.

[0041] In the actual implementation process, the embodiment of the present application can set a sound collection device at the rear of the vehicle, such as an external microphone. When the user needs to open the trunk, he knocks on the trunk area, and the sound collection device can collect the sound signal.

[0042] Among them, in the embodiment of the present application, the number of sound collection devices can be multiple, so that the user's tapping position range is larger.

[0043] In some embodiments, the sound collection device of the embodiment of the present application can also determine the sound collection time period of the sound collection device according to the user's historical car use time period to save energy consumption. For example, during the working time period of the user on weekdays, the embodiment of the present application can collect sound at longer time periods, and during the user's rest days, the embodiment of the present application can collect sound at shorter time periods.

[0044] In other embodiments, in order to avoid the situation where the user knocks during the non-collection period and the trunk of the vehicle fails to open, the embodiments of the present application can also set the sound loudness, that is, control the sound collection device to continue to operate in a low power consumption state, and when the loudness of any sound collected is higher than a certain value, it starts to continuously collect sound signals for a period of time.

[0045] In step S102, data processing is performed on each frame of the audio signal in the sound signal to obtain corresponding time domain signals and frequency domain signals, and the time-frequency domain features of the sound signal are calculated based on the time domain signals and frequency domain signals, so as to use the time-frequency domain features to determine whether the sound signal meets the preset knocking judgment conditions.

[0046] Furthermore, the embodiment of the present application can perform data processing on each frame of the audio signal in the sound signal, for example, enhance and reduce the noise of the audio signal, so as to subsequently determine and extract the knocking sound.

[0047] The embodiment of the present application can obtain the time domain signal and frequency domain signal corresponding to each frame of the audio signal after data processing, thereby calculating the time and frequency domain characteristics of the sound signal, and using the time and frequency domain characteristics to determine whether the sound signal meets the preset knock judgment conditions.

[0048] Optionally, in one embodiment of the present application, data processing is performed on each frame of audio signal in the sound signal, including: framing and windowing the sound signal to obtain a time domain signal; and performing Fourier transform on each frame signal in the time domain signal to obtain a frequency domain signal.

[0049] As a possible implementation method, the embodiment of the present application can perform frame and window processing on the audio signal x. The frame length can be 16ms, 32ms, etc., the frame shift can be 1 / 4 or 1 / 2 of the frame length, etc., and the windowing can be a Hamming window, a Hanning window, etc.

[0050] For example, the frame length selected in the embodiment of the present application is 32ms, and the frame shift is 1 / 2 of the frame length. The longer the frame length is, the more voice information is obtained per unit time, and the windowing uses the Hanning window. The specific formula is as follows:

[0051] x(m,n)=x_((m-1)*hop+n)*w(n),0≤n≤(L-1),

[0052] Wherein, m represents the frame number index, n represents the data point index of the frame, x(m,n) represents the nth time domain sampling frequency point of the mth frame, hop is the frame shift length, and L is the frame length.

[0053] The conversion relationship between the number of frame length points and the frame length is: L=fl*fs, where fl is the frame length, which is 32ms as mentioned above, and fs is the sampling rate. The sampling rate of the embodiment of the present application is 8000Hz.

[0054] In the embodiment of the present application, Fourier transform can be performed on each frame signal to obtain a frequency domain signal X(m, k), and the calculation formula thereof can be as follows:

[0055]

[0056] Among them, k is the frequency point in the frequency domain, and its range is

[0057] Optionally, in one embodiment of the present application, time-frequency domain features of a sound signal are calculated based on a time domain signal and a frequency domain signal, including: counting the peak amplitude features and dynamic range features of the time domain signal; judging whether a knock signal exists in the audio signal based on the peak amplitude features and the dynamic range features; and if a knock signal exists in the audio signal, calculating the time-frequency domain features based on the frequency domain signal.

[0058] In the actual execution process, the time domain signal x(m,n) and the frequency domain signal X(m,k) will be uniformly represented by x(m) and X(m) in the following description unless necessary, indicating the time domain signal of the mth frame and the frequency domain signal of the mth frame.

[0059] The embodiment of the present application utilizes a hard connection between the sound collection device and the tailgate, and the vibration generated by the user's tapping can be directly transmitted to the sound collection device, generating vibration signal characteristics that are different from the voice signal characteristics, which can be distinguished mainly through peak amplitude characteristics, dynamic range characteristics, and histogram characteristics. For each frame of the time domain signal x(m), specifically:

[0060] Statistical peak amplitude characteristics, the calculation formula is as follows:

[0061] F_peak(m)=max(20*log10(|x(m)|)).

[0062] Statistical dynamic range characteristics, the calculation formula is as follows:

[0063]

[0064] Among them, mins is a minimum value, generally 1×10 -6 or 1×10 -12 wait.

[0065] Based on the peak amplitude characteristics and dynamic range characteristics, if and only if F_peak(m)>T 1 (m)&&F_dr(m)>=T 2 (m), it is believed that there may be a knocking signal, then h 1 (m)=1, otherwise h 1 (m) = 0. At the same time, T 1 (m) and T 2 (m) When a knock signal is considered, smooth update is performed, and the formula is as follows:

[0066]

[0067] in, You can take 0.99.

[0068] The statistical histogram features are calculated as follows, where l is the different energy segments of different distributions.

[0069] F_hi(m,l)=count(20*log10(|x(m)|).

[0070] Based on the hl and F_hi features, if and only if F_hi(m,0) <T 3 (m)&&h 1 When (m) = 1, it is determined that there is a knock signal. 3 (m) can be updated smoothly or set to a fixed value.

[0071] Based on the acquired frequency domain signal X(m), the calculation characteristics are shown in the formula:

[0072]

[0073] When the determination of the dynamic range feature and the determination of the histogram feature are satisfied, the embodiment of the present application can extract and detect the frequency domain feature, as shown in the above formula, when It is judged as a knock signal.

[0074] The embodiment of the present application can obtain a first-stage judgment result by judging the time domain and frequency domain characteristics of all the above features, that is, judging whether the preset knocking judgment conditions are met. If the preset knocking judgment conditions are not met, continue to process other sound signals.

[0075] In step S103, if the preset knock determination condition is met, the knock signal is extracted from the sound signal and input into a pre-built knock model to output a vehicle trunk control instruction to open the vehicle trunk when the trunk control instruction is an opening instruction.

[0076] When it is judged as a knock signal, in order to optimize the false detection effect and provide users with a safer and more reliable experience, the embodiment of the present application can extract the knock signal from the sound signal and input the knock signal into a pre-built knock model, thereby using the knock model for secondary judgment, thereby improving the accuracy of knock signal detection and reducing misrecognition of knock signals.

[0077] Optionally, in one embodiment of the present application, before the knock signal is input into a pre-constructed knock model, it also includes: performing signal processing on the human knock sample signal and the vehicle knock sample signal in the sample training set respectively to obtain a first training knock sample signal and a second training knock sample signal that meet a preset energy condition; constructing an initial knock model using a Gaussian mixture model, the first training knock sample signal and the second training knock sample signal, and gradually iterating the initial knock model through the EM clustering algorithm and the sample test set to obtain a knock model.

[0078] In the actual implementation process, the embodiment of the present application can use a Gaussian mixture model to model the knocking signal, wherein the training set includes no less than 10 vehicles, 20 people in each vehicle, and more than 30 knocking data per person, as well as several data of non-target knocking signals such as raindrop knocking, door opening, and single knocking. Its steps such as framing and windowing for Fourier transform are consistent with the above part, and the sub-band energy needs to count the energy of 5 frequency bands such as 80Hz-200Hz, 200Hz-600Hz, 600Hz-1200Hz, 1200Hz-2400Hz, and 2400Hz-4000Hz. The Gaussian mixture model is gradually iterated by the EM clustering algorithm, and the mean vector μ, covariance matrix ∑ and weight vector φ of the Gaussian mixture model are finally obtained through a large amount of data.

[0079] When the user actually uses it, the probability of a knock signal is calculated by the following formula. When the probability exceeds a certain threshold, it is classified as a knock signal, otherwise it is a non-knock signal.

[0080]

[0081] in,

[0082] Optionally, in one embodiment of the present application, a knock signal is input into a pre-built knock model to output a trunk control instruction of the vehicle, including: inputting the knock signal into the knock model to obtain a knock probability; if the knock probability is greater than a preset probability threshold, determining that the trunk control instruction is an open instruction.

[0083] The embodiment of the present application can input a knock signal into a knock model to obtain a knock probability, and when a preset knock determination condition is met and the knock probability is greater than a certain value, determine that the trunk control instruction is an opening instruction, thereby using the opening instruction to control the opening of the trunk.

[0084] The preset probability threshold may be set by those skilled in the art according to actual conditions and is not specifically limited here.

[0085] Combination Figure 2-Figure 4 As shown, the working principle of the vehicle trunk control method of the embodiment of the present application is described in detail with an embodiment.

[0086] like Figure 2 As shown, it is a schematic diagram of the principle of an embodiment of the present application, with an external microphone with functions such as external wake-up as a sound collection device. When a user needs to open the trunk of the car and knocks on the trunk area, the external microphone of the embodiment of the present application collects a knocking signal and sends the signal to the central processing unit. The central processing unit uses a two-stage knock detection algorithm to detect that the collected signal meets the modeled knock signal characteristics, outputs a signal to open the trunk, and then opens the trunk. If the knock detection algorithm detects that the collected signal does not meet the modeled knock signal characteristics, it does not output a signal to open the trunk.

[0087] like Figure 3 As shown, taking the case where a user taps the same area of ​​the trunk three times in a row as an example of a tapping signal for opening the trunk, the embodiment of the present application may include the following steps:

[0088] Step S1: The microphone outside the vehicle collects the audio signal x in real time.

[0089] Step S2: The embodiment of the present application can perform frame division and windowing processing on the audio signal x. The frame length can be 16ms, 32ms, etc. The frame shift can be 1 / 4 or 1 / 2 of the frame length, etc. The windowing can be a Hamming window, a Hanning window, etc.

[0090] The frame length selected in the embodiment of the present application is 32ms, and the frame shift is 1 / 2 of the frame length. The longer the frame length, the more voice information is obtained per unit time, and the windowing uses the Hanning window. The specific formula is as follows:

[0091] x(m,n)=x_((m-1)*hop+n)*w(n),0≤n≤(L-1),

[0092] Wherein, m represents the frame number index, n represents the data point index of the frame, x(m,n) represents the nth time domain sampling frequency point of the mth frame, hop is the frame shift length, and L is the frame length.

[0093] The conversion relationship between the number of frame length points and the frame length is: L=fl*fs, where fl is the frame length, which is 32ms as mentioned above, and fs is the sampling rate. The sampling rate of the embodiment of the present application is 8000Hz.

[0094] In the embodiment of the present application, Fourier transform can be performed on each frame signal to obtain a frequency domain signal X(m, k), and the calculation formula thereof can be as follows:

[0095]

[0096] Among them, k is the frequency point in the frequency domain, and its range is

[0097] Step S3: In the actual execution process, the time domain signal x(m,n) and the frequency domain signal X(m,k) obtained in step S2 will be uniformly represented by x(m) and X(m) in the subsequent description unless necessary, indicating the time domain signal of the mth frame and the frequency domain signal of the mth frame.

[0098] The embodiment of the present application utilizes a hard connection between the sound collection device and the tailgate, and the vibration generated by the user's tapping can be directly transmitted to the sound collection device, generating vibration signal characteristics that are different from the voice signal characteristics, which can be distinguished mainly through peak amplitude characteristics, dynamic range characteristics, and histogram characteristics. For each frame of the time domain signal x(m), specifically:

[0099] Statistical peak amplitude characteristics, the calculation formula is as follows:

[0100] F_peak(m)=max(20*log10(|x(m)|)).

[0101] Statistical dynamic range characteristics, the calculation formula is as follows:

[0102]

[0103] Among them, mins is a minimum value, generally 1×10 -6 or 1×10 -12 wait.

[0104] Based on the peak amplitude characteristics and dynamic range characteristics, if and only if F_peak(m)>T1 (m)&&F_dr(m)>=T 2 (m), it is believed that there may be a knocking signal, then h 1 (m)=1, otherwise h 1 (m) = 0. At the same time, T 1 (m) and T 2 (m) When a knock signal is considered, smooth update is performed, and the formula is as follows:

[0105]

[0106] in, You can take 0.99.

[0107] The statistical histogram features are calculated as follows, where l is the different energy segments of different distributions.

[0108] F_hi(m,l)=count(20*log10(|x(m)|).

[0109] Based on the hl and F_hi features, if and only if F_hi(m,0) <T 3 (m)&&h 1 When (m) = 1, it is determined that there is a knock signal. 3 (m) can be updated smoothly or set to a fixed value.

[0110] Based on the acquired frequency domain signal X(m), the calculation characteristics are shown in the formula:

[0111]

[0112] When the determination of the dynamic range feature and the determination of the histogram feature are satisfied, the embodiment of the present application can extract and detect the frequency domain feature, as shown in the above formula, when It is judged as a knock signal.

[0113] The embodiment of the present application can obtain a first-stage judgment result by judging the time domain and frequency domain characteristics of all the above features, that is, judging whether the preset knocking judgment conditions are met. If the preset knocking judgment conditions are not met, continue to process other sound signals.

[0114] Step S4: When it is judged as a knocking signal, in order to optimize the false detection effect and provide users with a safer and more reliable experience, the embodiment of the present application can collect a large number of knocking signals from people and vehicles based on the time-frequency domain characteristics of stage one, model the knocking signals, improve the accuracy of knocking signal detection, and reduce false recognition of knocking signals.

[0115] In the actual implementation process, Figure 4As shown, the embodiment of the present application can use a Gaussian mixture model to model the knocking signal, wherein the training set includes no less than 10 vehicles, 20 people in each vehicle, and more than 30 knocking data per person, as well as several data of non-target knocking signals such as raindrop knocking, door opening, and single knocking. Its steps such as framing and windowing for Fourier transform are consistent with the above part. The subband energy needs to count the energy of 5 frequency bands such as 80Hz-200Hz, 200Hz-600Hz, 600Hz-1200Hz, 1200Hz-2400Hz, and 2400Hz-4000Hz. The specific formula can be shown in step S302. The Gaussian mixture model is gradually iterated by the EM clustering algorithm, and the mean vector μ, covariance matrix Σ and weight vector φ of the Gaussian mixture model are finally obtained through a large amount of data.

[0116] When the user actually uses it, the probability of a knock signal is calculated by the following formula. When the probability exceeds a certain threshold, it is classified as a knock signal, otherwise it is a non-knock signal.

[0117]

[0118] in,

[0119] Finally, the trunk opening command will be output only after the comprehensive judgment output of the first stage (preset knocking judgment condition stage) and the second stage (knocking model stage) is a knocking signal.

[0120] In summary, compared with voice unlocking, car key unlocking and other methods, the embodiments of the present application are more in line with the user's first intuition, convenient and without learning cost. Through a two-stage knock detection algorithm, compared with a single-stage time-frequency domain signal processing method, it can effectively filter out false detections and ensure the user's safety.

[0121] Through the embodiments of the present application, when the user approaches the tailgate of a car, he or she only needs to knock on the tailgate to open the trunk, without having to speak or press a button in a fixed position; when the user rides in the target vehicle for the first time and needs to open the trunk to put or take items, he or she does not need to search back and forth for the trunk opening button in the trunk to open the trunk.

[0122] According to the vehicle trunk control method proposed in the embodiment of the present application, the sound signal collected by the sound collection device can be processed to determine whether the sound signal meets the preset knocking judgment condition, and after determining that the judgment condition is met, it is further determined whether it is the control intention to open the trunk according to the knocking data, thereby realizing the automatic opening of the vehicle trunk, without the need for complex opening actions, with higher action safety, and in line with the user's behavioral rules for door opening actions, and reducing the probability of false touches, while taking into account convenience and safety. Thus, the technical problem in the related art that the trunk opening method does not conform to the user's behavioral rules, there are certain safety hazards when the user operates, and it is easy to cause damage to the outside of the vehicle is solved.

[0123] Next, the control device for the vehicle trunk proposed in accordance with the embodiment of the present application will be described with reference to the accompanying drawings.

[0124] Figure 5 It is a block diagram of a control device for a vehicle trunk according to an embodiment of the present application.

[0125] like Figure 5 As shown, the control device 10 of the vehicle trunk is provided with a sound collection device for waking up the external vehicle function at the rear of the vehicle, wherein the device 10 includes: an acquisition module 100, a judgment module 200 and a control module 300.

[0126] Specifically, the acquisition module 100 is used to acquire a sound signal collected by at least one sound collection device in a current collection period.

[0127] The judgment module 200 is used to perform data processing on each frame of the audio signal in the sound signal to obtain the corresponding time domain signal and frequency domain signal, and calculate the time and frequency domain characteristics of the sound signal based on the time domain signal and the frequency domain signal, so as to use the time and frequency domain characteristics to determine whether the sound signal meets the preset knock judgment conditions.

[0128] The control module 300 is used to extract a knock signal from the sound signal when a preset knock determination condition is met, and input the knock signal into a pre-built knock model to output a vehicle trunk control instruction to open the vehicle trunk when the trunk control instruction is an opening instruction.

[0129] Optionally, in one embodiment of the present application, the determination module 200 includes: a processing unit and a first calculation unit.

[0130] The processing unit is used to perform frame division and windowing processing on the sound signal to obtain a time domain signal.

[0131] The first calculation unit is used to perform Fourier transform on each frame signal in the time domain signal to obtain a frequency domain signal.

[0132] Optionally, in one embodiment of the present application, the judgment module 200 includes: a statistical unit, a judgment unit and a second calculation unit.

[0133] The statistical unit is used to count the peak amplitude characteristics and dynamic range characteristics of the time domain signal.

[0134] The judging unit is used to judge whether there is a knock signal in the audio signal based on the peak amplitude characteristic and the dynamic range characteristic.

[0135] The second calculation unit is used to calculate the time-frequency domain features based on the frequency domain signal when there is a knock signal in the audio signal.

[0136] Optionally, in one embodiment of the present application, the control device 10 of the vehicle trunk further includes: a signal processing module and a construction module.

[0137] Among them, the signal processing module is used to perform signal processing on the human knocking sample signal and the car knocking sample signal in the sample training set respectively, to obtain the first training knocking sample signal and the second training knocking sample signal that meet the preset energy conditions.

[0138] A construction module is used to construct an initial knocking model using a Gaussian mixture model, a first training knocking sample signal and a second training knocking sample signal, and to gradually iterate the initial knocking model through an EM clustering algorithm and a sample test set to obtain a knocking model.

[0139] Optionally, in one embodiment of the present application, the control module 300 includes: an input unit and a determination unit.

[0140] Among them, the input unit is used to input the knocking signal into the knocking model to obtain the knocking probability.

[0141] The determination unit is used to determine that the trunk control instruction is an opening instruction when the knock probability is greater than a preset probability threshold.

[0142] It should be noted that the above explanation of the embodiment of the vehicle trunk control method is also applicable to the vehicle trunk control device of this embodiment, and will not be repeated here.

[0143] According to the control device for the trunk of a vehicle proposed in the embodiment of the present application, the sound signal collected by the sound collection device can be processed to determine whether the sound signal meets the preset knocking judgment condition, and after determining that the judgment condition is met, it is further determined whether it is the control intention to open the trunk according to the knocking data, thereby realizing the automatic opening of the trunk of the vehicle, without the need for complex opening actions, with higher action safety, and in line with the user's behavioral rules for door opening actions, and reducing the probability of false touches, while taking into account convenience and safety. Thus, the technical problem in the related art that the trunk opening method does not conform to the user's behavioral rules, there are certain safety hazards when the user operates, and it is easy to cause damage to the outside of the vehicle is solved.

[0144] Figure 6 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0145] A memory 601 , a processor 602 , and a computer program stored in the memory 601 and executable on the processor 602 .

[0146] When the processor 602 executes the program, the vehicle trunk control method provided in the above embodiment is implemented.

[0147] Furthermore, the vehicle also includes:

[0148] The communication interface 603 is used for communication between the memory 601 and the processor 602 .

[0149] The memory 601 is used to store computer programs that can be executed on the processor 602 .

[0150] The memory 601 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0151] If the memory 601, the processor 602 and the communication interface 603 are implemented independently, the communication interface 603, the memory 601 and the processor 602 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0152] Optionally, in a specific implementation, if the memory 601, the processor 602 and the communication interface 603 are integrated on a chip, the memory 601, the processor 602 and the communication interface 603 can communicate with each other through an internal interface.

[0153] The processor 602 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0154] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned vehicle trunk control method is implemented.

[0155] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the vehicle trunk control method provided by an embodiment of the present invention.

[0156] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0157] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0158] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0159] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0160] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0161] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0162] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0163] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for controlling a vehicle trunk, characterized in that: The rear of the vehicle is provided with a sound collection device for waking up the external vehicle function, wherein the method comprises the following steps: Acquire a sound signal collected by at least one sound collection device in a current collection period; Performing data processing on each frame of the audio signal in the sound signal to obtain a corresponding time domain signal and a frequency domain signal, and calculating the time-frequency domain features of the sound signal based on the time domain signal and the frequency domain signal, so as to use the time-frequency domain features to determine whether the sound signal meets a preset knocking determination condition; If the preset knock determination condition is met, a knock signal is extracted from the sound signal and input into a pre-built knock model to output a trunk control instruction of the vehicle, so as to open the trunk of the vehicle when the trunk control instruction is an opening instruction.

2. The method according to claim 1, characterized in that The performing data processing on each frame of the audio signal in the sound signal comprises: Performing frame division and windowing processing on the sound signal to obtain the time domain signal; Perform Fourier transform on each frame signal in the time domain signal to obtain the frequency domain signal.

3. The method according to claim 1, characterized in that The calculating the time-domain and frequency-domain features of the sound signal based on the time-domain signal and the frequency-domain signal comprises: Counting the peak amplitude characteristics and dynamic range characteristics of the time domain signal; Determining whether there is a knock signal in the audio signal based on the peak amplitude feature and the dynamic range feature; If the knock signal exists in the audio signal, the time-frequency domain feature is calculated based on the frequency domain signal.

4. The method according to claim 1, characterized in that: Before inputting the knock signal into the pre-built knock model, the method further includes: Perform signal processing on the human knocking sample signal and the vehicle knocking sample signal in the sample training set respectively to obtain a first training knocking sample signal and a second training knocking sample signal that meet a preset energy condition; An initial knocking model is constructed using a Gaussian mixture model, the first training knocking sample signal and the second training knocking sample signal, and the initial knocking model is gradually iterated using an EM clustering algorithm and a sample test set to obtain the knocking model.

5. The method according to claim 1, characterized in that The step of inputting the knock signal into a pre-built knock model to output a trunk control instruction of the vehicle includes: Inputting the knock signal into the knock model to obtain a knock probability; If the knock probability is greater than a preset probability threshold, the trunk control instruction is determined to be the opening instruction.

6. A control device for a vehicle trunk, characterized in that: The rear of the vehicle is provided with a sound collection device for waking up the external function, wherein the device comprises: An acquisition module, used to acquire a sound signal collected by at least one sound collection device in a current collection period; A judgment module, used to perform data processing on each frame of the audio signal in the sound signal to obtain a corresponding time domain signal and a frequency domain signal, and calculate the time-frequency domain features of the sound signal based on the time domain signal and the frequency domain signal, so as to use the time-frequency domain features to judge whether the sound signal meets a preset knocking judgment condition; A control module is used to extract a knock signal from the sound signal when the preset knock judgment condition is met, and input the knock signal into a pre-built knock model to output a trunk control instruction of the vehicle, so as to open the trunk of the vehicle when the trunk control instruction is an opening instruction.

7. The device according to claim 6, characterized in that The judgment module includes: A processing unit, used for performing frame division and windowing processing on the sound signal to obtain the time domain signal; The first calculation unit is used to perform Fourier transform on each frame signal in the time domain signal to obtain the frequency domain signal.

8. The device according to claim 6, characterized in that The judging module comprises: A statistical unit, used for counting the peak amplitude characteristics and dynamic range characteristics of the time domain signal; A judging unit, configured to judge whether there is a knock signal in the audio signal based on the peak amplitude feature and the dynamic range feature; The second calculation unit is used to calculate the time-frequency domain feature based on the frequency domain signal when the knock signal exists in the audio signal.

9. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle trunk control method as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the vehicle trunk control method as described in any one of claims 1 to 5.

Images