Intelligent driving method, external audio acquisition device, domain controller and automobile

Abstract

The invention relates to the technical field of automatic driving, and discloses an intelligent driving method, an out-of-vehicle audio collection device, a domain controller and an automobile, the method is applied to the out-of-vehicle audio collection device in the automobile, the automobile further comprises the domain controller, the out-of-vehicle audio collection device is connected with the domain controller, and the method comprises the steps that out-of-vehicle audio outside the automobile is collected; screening the audio outside the vehicle to obtain a target audio, and performing sound source positioning on the target audio; and transmitting the sound source positioning result and the target audio to a domain controller, so that the domain controller determines a corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the automobile according to the target driving strategy. Compared with existing intelligent driving which does not consider the sound in the external environment, the method can be used for driving in combination with the sound outside the vehicle, so that the safety is improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to an intelligent driving method, an off-vehicle audio acquisition device, a domain controller, and a vehicle. Background Art

[0002] With the development of automobile intelligence, many vehicles have intelligent driving functions such as automatic driving or assisted driving. At present, automatic driving or assisted driving mainly relies on optical cameras, millimeter wave and lidar technology, or a combination of the above technologies to collect data about the external environment, and transmit the collected results to the domain controller for analysis, so as to drive according to the corresponding driving strategy.

[0003] However, existing autonomous driving or assisted driving does not take into account the sound conditions in the external environment, resulting in poor safety. Summary of the invention

[0004] The main purpose of this application is to provide an intelligent driving method, an off-vehicle audio collection device, a domain controller and a car, aiming to solve the technical problem that the existing technology does not take the sound in the external environment into consideration during intelligent driving and has poor safety.

[0005] To achieve the above object, the present application provides an intelligent driving method, which is applied to an external audio collection device in a car, the car also includes a domain controller, the external audio collection device is connected to the domain controller, and the method includes:

[0006] collecting external audio outside the car;

[0007] Filtering the external audio to obtain target audio, and locating the sound source of the target audio;

[0008] The sound source positioning result and the target audio are transmitted to the domain controller, so that the domain controller determines a corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

[0009] In one embodiment, the step of filtering the external audio to obtain the target audio includes:

[0010] Extracting features of the signal of the external audio, wherein the signal features include at least one of a frequency feature, an amplitude feature, a duration feature or a spectrum feature;

[0011] Marking the external audio according to the feature extraction result to obtain vehicle audio and human voice audio;

[0012] The vehicle audio and the human voice audio are used as target audio.

[0013] In one embodiment, before the step of locating the sound source of the target audio, the method further includes:

[0014] Performing signal analysis on the target audio, and determining audio characteristics of the target audio based on the analysis result;

[0015] Determining a corresponding target proportional amplification ratio based on the audio feature, and performing signal enhancement on the target audio according to the target proportional amplification ratio;

[0016] The step of locating the sound source of the target audio comprises:

[0017] Perform sound source positioning on the target audio after signal enhancement.

[0018] In one embodiment, the step of locating the sound source of the target audio includes:

[0019] Performing partition processing on the target audio according to each preset partition, obtaining the audio intensity and arrival time point of the target audio relative to each preset partition;

[0020] Determine the sound source direction and sound source distance of the target audio according to each of the audio intensities and each of the arrival time points;

[0021] The sound source of the target audio is located based on the sound source orientation and the sound source distance.

[0022] In addition, to achieve the above purpose, the present application also proposes an intelligent driving method, which is applied to a domain controller in a car, the car also includes an external audio collection device, the domain controller is connected to the external audio collection device, and the method includes:

[0023] Receiving the sound source positioning result and the target audio sent by the external audio collection device, wherein the target audio is obtained by filtering the external audio collected outside the car by the external audio collection device, and the sound source positioning result is obtained by performing sound source positioning on the target audio by the external audio collection device;

[0024] A corresponding target driving strategy is determined according to the target audio and the sound source positioning result, and the car is driven according to the target driving strategy.

[0025] In one embodiment, the step of determining a corresponding target driving strategy according to the target audio and the sound source localization result, and driving the car according to the target driving strategy includes:

[0026] Determine the audio type of the target audio through a preset classification model, and determine the sound source direction and sound source distance of the target audio according to the sound source positioning result;

[0027] Determine a distance change of the target audio based on the sound source orientation and the sound source distance;

[0028] When the distance change is a close distance, a corresponding target driving strategy is determined based on the audio type, the sound source orientation, and the sound source distance, and the car is driven according to the target driving strategy.

[0029] In one embodiment, the domain controller is also connected to a prompting device of the car;

[0030] The step of determining a corresponding target driving strategy according to the target audio and the sound source positioning result, and driving the car according to the target driving strategy includes:

[0031] Determine the current driving mode;

[0032] When the current driving mode is the assisted driving mode, determining a corresponding target prompt strategy according to the target audio and the sound source positioning result, and controlling the prompt device to give a prompt according to the target prompt strategy;

[0033] When the current driving mode is the automatic driving mode, a corresponding target avoidance strategy is determined according to the target audio and the sound source positioning result, and the car is controlled to avoid according to the target avoidance strategy.

[0034] In addition, to achieve the above-mentioned purpose, the present application also proposes an off-vehicle audio collection device, the off-vehicle audio collection device comprising: a main collection component and at least one slave collection component;

[0035] The master acquisition component is connected to the slave acquisition component and the domain controller respectively;

[0036] The secondary collection component is used to collect the external audio outside the car;

[0037] The main collecting component is used to collect the external audio outside the car and receive the external audio transmitted by the slave collecting component;

[0038] The main acquisition component is also used to filter the external audio of the vehicle, obtain the target audio, and locate the sound source of the target audio;

[0039] The main acquisition component is also used to transmit the sound source positioning result and the target audio to the domain controller, so that the domain controller determines the corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

[0040] In addition, to achieve the above-mentioned purpose, the present application also proposes a domain controller, wherein the domain controller is connected to an external audio collection device;

[0041] The domain controller is used to receive the sound source positioning result and the target audio sent by the external audio collection device, the target audio is obtained by filtering the external audio collected by the external audio collection device, and the sound source positioning result is obtained by performing sound source positioning on the target audio by the external audio collection device;

[0042] The domain controller is further used to determine a corresponding target driving strategy according to the target audio and the sound source positioning result, and drive the car according to the target driving strategy.

[0043] In addition, to achieve the above-mentioned purpose, the present application also proposes a car, which includes the external audio collection device as described above and the domain controller as described above.

[0044] The present application provides an intelligent driving method, an external audio collection device, a domain controller and a car. The method is applied to an external audio collection device in a car. The car also includes a domain controller, and the external audio collection device is connected to the domain controller. The method includes: collecting external audio outside the car; screening the external audio to obtain target audio, and locating the sound source of the target audio; transmitting the sound source positioning result and the target audio to the domain controller, so that the domain controller determines a corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

[0045] Since the present application can be provided with an external audio collection device in the car, and the external audio collection device is connected to the domain controller, in actual use, the external audio collection device can collect the external audio outside the car, and filter the external audio to obtain the target audio that affects the driving of the car, and then locate the sound source of the target audio, and transmit the obtained sound source location result and the target audio to the domain controller together, so that the domain controller determines the corresponding target driving strategy to drive the car according to the sound source location result and the target audio. Compared with the existing intelligent driving that does not consider the sound in the external environment, the present application can drive in combination with the sound outside the car, which improves safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0048] Figure 1 This is a flow chart of the first embodiment of the intelligent driving method of the present application;

[0049] Figure 2 This is a schematic diagram of the structure of the external audio collection device in the first embodiment of the intelligent driving method of this application;

[0050] Figure 3 This is a basic principle diagram of the first embodiment of the intelligent driving method of this application;

[0051] Figure 4 This is a processing diagram of the first embodiment of the intelligent driving method of the present application;

[0052] Figure 5 This is a flow chart of the second embodiment of the intelligent driving method of the present application;

[0053] Figure 6 This is a schematic diagram of preset partitions in the second embodiment of the intelligent driving method of this application;

[0054] Figure 7 This is a flowchart of the third embodiment of the intelligent driving method of the present application;

[0055] Figure 8 This is a schematic diagram of the algorithm identification process in the third embodiment of the intelligent driving method of this application.

[0056] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0057] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] It should be noted that with the development of automobile intelligence, many vehicles have intelligent driving functions such as automatic driving or assisted driving. At present, automatic driving or assisted driving mainly relies on optical cameras, millimeter wave and lidar technologies, or a combination of the above technologies to collect data about the external environment, and transmit the collected results to the domain controller for analysis, so as to drive according to the corresponding driving strategy.

[0059] However, existing autonomous driving or assisted driving does not take into account the sound conditions in the external environment, resulting in poor safety.

[0060] Therefore, in order to solve the above defects, the present embodiment provides an intelligent driving method, which is applied to an external audio collection device in a car. The car also includes a domain controller, and the external audio collection device is connected to the domain controller. Since the present embodiment can be provided with an external audio collection device in the car, and the external audio collection device is connected to the domain controller, in actual use, the external audio collection device can collect external audio outside the car, and filter the external audio to obtain the target audio that affects the driving of the car, and then locate the sound source of the target audio, and transmit the obtained sound source location result and the target audio to the domain controller together, so that the domain controller determines the corresponding target driving strategy to drive the car according to the sound source location result and the target audio. Compared with the existing intelligent driving that does not take into account the sound in the external environment, the present embodiment can drive in combination with the sound outside the car, thereby improving safety.

[0061] For ease of understanding, the following combination Figures 1 to 8 The intelligent driving method provided in the embodiments of the present application is specifically introduced.

[0062] The present application embodiment provides an intelligent driving method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the intelligent driving method of the present application.

[0063] In this embodiment, the intelligent driving method is applied to an external audio collection device in a car. The car also includes a domain controller, and the external audio collection device is connected to the domain controller.

[0064] It is understandable that the above-mentioned external audio collection device can be any device for audio collection, such as a microphone, etc., and this embodiment does not limit this. The above-mentioned domain controller can be a device for controlling the intelligent driving of the car. During the intelligent driving process, the domain controller is usually connected to multiple cameras, millimeter-wave radars or laser radars, etc. The domain controller analyzes and processes the external image data collected by the camera, millimeter-wave radar or laser radar, and outputs the final driving strategy to the car's actuator, thereby controlling the lateral and longitudinal movement of the vehicle through the actuator.

[0065] In this embodiment, the external audio acquisition device can be connected to the domain controller, so that the domain controller can obtain not only external image data, but also external audio data, and analyze and process the external image data and the external audio data together to determine the final driving strategy, thereby improving safety.

[0066] In this embodiment, the intelligent driving method includes the following steps:

[0067] Step S10: collecting the external audio outside the car.

[0068] It should be understood that the method of this embodiment can be executed by the above-mentioned vehicle-external audio acquisition device. In this embodiment, the vehicle-external audio acquisition device may be provided with an audio pickup sensor for collecting audio, such as a microphone or a vibration acceleration sensor; a processor for processing and analysis may be provided, and the processor may be a digital signal processor (DSP). Of course, other processors may also be used, and this embodiment is not limited to this; a power supply module may also be provided to power the audio pickup sensor and the processor, such as a battery; and, in order to transmit data to the domain controller, the vehicle-external audio acquisition device in this embodiment may also include a transmission module, which may be a module for data transmission. In this embodiment, an automotive audio bus (Automotive Audio Bus, A2B) may be used to realize data transmission, that is, the DSP is connected to the domain controller via the automotive audio bus.

[0069] Furthermore, considering that in order to collect the external audio from different directions, refer to Figure 2 , Figure 2 This is a schematic diagram of the structure of the external audio collection device in the first embodiment of the intelligent driving method of this application. Figure 2 As shown, in this embodiment, the above-mentioned vehicle external audio collection device includes: a main collection component and at least one slave collection component;

[0070] The master acquisition component is connected to the slave acquisition component and the domain controller respectively;

[0071] The secondary collection component is used to collect the external audio outside the car;

[0072] The main collecting component is used to collect the external audio outside the car and receive the external audio transmitted by the slave collecting component.

[0073] It should be noted that the above-mentioned main acquisition component and the slave acquisition component can be components that both have audio acquisition functions, and the above-mentioned power supply module (i.e. Figure 2 POWER in the main acquisition component and the components of each slave acquisition component are powered by bus power supply; an audio pickup sensor (i.e. Figure 2 VACC / VPU / MIC), which may be a vibration acceleration sensor for collecting the above-mentioned audio outside the vehicle; a processor for performing analysis and processing (i.e. Figure 2DSP) and a transmission module for data transmission (i.e. Figure 2 A2B); the above-mentioned processor and power supply module do not need to be set in the acquisition component. Of course, the power supply module can also be set in the acquisition component for independent power supply. This embodiment does not limit this.

[0074] It should also be noted that in the present embodiment, at least one main acquisition component may be provided in the above-mentioned vehicle external audio acquisition device to implement the solution of the present embodiment. Of course, in order to collect multi-directional vehicle external audio, at least one slave acquisition component may also be provided. Figure 2 There are two acquisition components shown in the figure, namely Figure 2 From collection component 1 and from collection component 2.

[0075] Then, each audio pickup sensor in the acquisition component is connected to the corresponding transmission module (ie Figure 2 In PDM), the acquisition components can be set at different positions in the car to collect the external audio at different positions, such as the front and rear windshields and near the triangular windows of the car. The external audio is transmitted to the car by resonating with the glass, and the acquisition component indirectly collects the external audio by collecting the glass vibration signal.

[0076] Furthermore, considering that there is a certain difference between the collected vibration signal and the external audio when the external microphone is actually used to collect the audio outside the car, the present embodiment can convert the two through an empirical conversion function. At the same time, an additional microphone can be installed, with the sound hole facing the interior of the car, mainly used to collect the audio signal inside the car, further suppressing the influence of the audio inside the car on the external audio collected by the vibration acceleration sensor.

[0077] After the audio outside the vehicle is collected by the audio pickup sensor, it can be transmitted to the transmission module. The transmission modules in each acquisition component can be connected in series (i.e. Figure 2 A2B BUS), and set a transmission module of the slave acquisition component to connect with the transmission module of the master acquisition component, Figure 2 The transmission module of the slave acquisition component 2 is connected to the transmission module of the slave acquisition component 1, and the transmission module of the slave acquisition component 1 is connected to the transmission module of the main acquisition component, so that the external audio collected by each slave acquisition component can be transmitted to the main acquisition module.

[0078] The transmission module in the main acquisition module can transmit the received external audio from each acquisition component to the processor in the main acquisition component (i.e. Figure 2In the TDM), the main acquisition component can also collect the audio outside the car through the internal audio pickup sensor and transmit it to the processor, so that the processor can obtain the audio outside the car collected by the main acquisition component and each slave acquisition component, and perform subsequent processing on the audio outside the car. Therefore, the subsequent steps of this embodiment that are performed by the external audio acquisition device can be performed by the processor in the main acquisition component, but for the convenience of subsequent description, the external audio acquisition device is still used for description.

[0079] Furthermore, considering that the acquisition module includes a vibration acceleration sensor and a microphone, the vibration acceleration sensor is arranged near the car glass to collect the vibration of the car glass, and the microphone can collect the sound inside the car and transmit them to the processor. The processor can subtract the sound inside the car from the vibration signal to obtain the audio outside the car.

[0080] Step S20: Filter the external audio to obtain target audio, and locate the sound source of the target audio.

[0081] It is understandable that the target audio may be audio that interferes with the driving of the car. For example, when the car is in intelligent driving, if a special vehicle such as an ambulance honks behind the car, the car should give way to it, so the honking of the special vehicle can be used as the target audio; if a pedestrian speaks to the side of the car, the car should give way to it, so the pedestrian speaks can be used as the target audio, etc. This embodiment does not limit this.

[0082] Therefore, in actual use, when the above-mentioned vehicle external audio collection device obtains the vehicle external audio, specifically, when the processor in the main collection component obtains the vehicle external audio, the vehicle external audio can be screened to determine whether there is target audio that affects driving, and the target audio is located to determine the specific location of the target audio. At the same time, the above-mentioned collection component is set in the car to prevent the device from being damaged by bad weather and prolong its service life.

[0083] At the same time, when collecting, in order to ensure the synchronization of collection, in this embodiment, the master collection component can provide a synchronous clock for each slave collection component. Of course, the domain controller can also provide a synchronous clock for the master collection component and the slave collection component to achieve timing synchronization. This embodiment does not impose any restrictions on this.

[0084] Furthermore, in order to obtain the above-mentioned target audio, the step of filtering the external audio to obtain the target audio includes:

[0085] Step S21: extracting features of the signal features of the external vehicle audio, wherein the signal features include at least one of frequency features, amplitude features, duration features or spectrum features.

[0086] For ease of understanding, refer to Figure 3 as well as Figure 4 , Figure 3 This is the basic principle diagram of the first embodiment of the intelligent driving method of this application. Figure 4 This is a processing diagram of the first embodiment of the intelligent driving method of this application. Figure 3 as well as Figure 4 As shown, it should be understood that after the processor in the main acquisition component obtains the above-mentioned external audio (ie Figure 3 Data record, that is Figure 4 The master node audio data and the slave node audio data) can be acquired by the processor of the main acquisition component (i.e. Figure 3 DSP, that is Figure 4 The main node DSP in the middle first pre-processes the external audio (i.e. Figure 3 Data preprocessing), which may include denoising, filtering, etc. (i.e. Figure 4 medium noise suppression).

[0087] After preprocessing, the preprocessed external audio can be subjected to feature extraction (i.e. Figure 3 In the feature processing, the feature extraction can be based on at least one of the frequency feature, the amplitude feature, the duration feature or the spectrum feature; wherein the frequency feature can be the audio component feature of the external audio, the amplitude feature can be the amplitude change feature of the external audio, the duration feature can be the feature corresponding to the start time and the end time of the external audio, and the spectrum feature can be the feature corresponding to the frequency domain of the external audio;

[0088] Therefore, the processor can perform feature extraction on the external vehicle audio according to at least one of the above-mentioned frequency features, amplitude features, duration features or spectrum features to obtain corresponding feature extraction results.

[0089] Step S22: marking the external audio according to the feature extraction result to obtain vehicle audio and human voice audio;

[0090] Step S23: taking the vehicle audio and the human voice audio as target audio.

[0091] It is understandable that the above-mentioned vehicle audio may be the audio emitted by surrounding vehicles, and the above-mentioned human voice audio may be the audio emitted by surrounding pedestrians.

[0092] After obtaining the feature extraction results, analysis can be performed based on the feature extraction results. Since the sounds emitted by different sound sources have different signal characteristics, vehicle audio and human voice audio can be extracted from them according to the corresponding signal characteristics and used as the above-mentioned target audio (i.e. Figure 3 Target audio extraction).

[0093] After obtaining the target audio, the target audio can be located (i.e. Figure 3 and Figure 4 The specific positioning method can be implemented by the time difference of arrival (TDOA). Of course, it can also be implemented by other methods, which is not limited in this embodiment.

[0094] Step S30: Transmitting the sound source positioning result and the target audio to the domain controller, so that the domain controller determines a corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

[0095] Continue as Figure 2 As shown, the transmission module in the main acquisition component in this embodiment can also be connected to the domain controller (ie Figure 2 After the sound source is located, the sound source location result and the target audio can be transmitted to the domain controller (i.e. Figure 3 Output position information, that is, Figure 4 (audio data + position information output).

[0096] It should be noted that the above target driving strategy may be a strategy for controlling the driving of a car. In actual use, after obtaining the target audio and the sound source positioning result, the domain controller can analyze and determine the corresponding target driving strategy, and drive the car according to the target driving strategy.

[0097] In this embodiment, an external audio collection device may be provided in the car, and the external audio collection device is connected to the domain controller. In actual use, the external audio collection device may collect external audio outside the car, and filter the external audio to obtain the target audio that affects the driving of the car, and then locate the sound source of the target audio, and transmit the obtained sound source location result and the target audio to the domain controller together, so that the domain controller determines the corresponding target driving strategy to drive the car according to the sound source location result and the target audio. Compared with the existing intelligent driving that does not consider the sound in the external environment, this embodiment can drive in combination with the sound outside the car, which improves safety.

[0098] refer to Figure 5 , Figure 5 The flowchart of the second embodiment of the intelligent driving method of the present application is shown in FIG. Based on the above first embodiment, the second embodiment of the intelligent driving method of the present application is proposed.

[0099] In order to further improve the quality of the target audio for subsequent processing, such as Figure 5 As shown, in this embodiment, before the step of locating the sound source of the target audio, the step further includes:

[0100] Step S201: performing signal analysis on the target audio, and determining the audio characteristics of the target audio based on the analysis result.

[0101] It should be noted that the above audio characteristics may be features of an audio signal, such as loudness, frequency, etc., which is not limited in this embodiment.

[0102] In actual use, after determining the target audio, the processor of the main acquisition component may first perform signal analysis on the target audio, specifically analyzing its loudness, frequency distribution, etc., to obtain corresponding audio characteristics.

[0103] Step S202: determining a corresponding target proportional amplification ratio based on the audio feature, and performing signal enhancement on the target audio according to the target proportional amplification ratio;

[0104] The step of locating the sound source of the target audio comprises:

[0105] Step S20': localizing the sound source of the target audio after the signal enhancement.

[0106] It is understandable that the above target proportional amplification ratio may be a ratio for enhancing the audio signal. In this embodiment, the signal strength may be enhanced by proportional amplification. In actual use, after the main acquisition component obtains the audio characteristics, a suitable proportional amplification ratio may be selected according to the audio characteristics as the above target proportional amplification ratio, and the target audio signal may be enhanced according to the target proportional amplification ratio (i.e. Figure 3 At the same time, the audio outside the car audio can be suppressed except for the target audio. Finally, the sound source of the target audio after signal enhancement can be located.

[0107] Furthermore, in order to realize sound source localization, in this embodiment, the step of performing sound source localization on the target audio includes:

[0108] Step S21: partition the target audio according to each preset partition to obtain the audio intensity and arrival time point of the target audio relative to each preset partition.

[0109] It should be noted that the preset partition may be the area corresponding to the installation position of the main acquisition component and each slave acquisition component in this embodiment. The audio intensity may be the loudness of the audio. The arrival time point may be the time point when the audio arrives at the acquisition module in the preset partition.

[0110] Reference Figure 6 , Figure 6 This is a schematic diagram of preset partitions in the second embodiment of the intelligent driving method of this application, such as Figure 6As shown, if this embodiment is provided with a main acquisition component and three slave acquisition components, and they are respectively arranged at the front, rear, left and right positions of the car, there can be four preset partitions (i.e. Figure 6 The target audio is then processed in different zones according to the beamforming principle to determine the audio intensity and arrival time of the target audio in each preset zone.

[0111] Step S22: determining the sound source direction and sound source distance of the target audio according to each of the audio intensities and each of the arrival time points;

[0112] Step S23: localizing the sound source of the target audio based on the sound source orientation and the sound source distance.

[0113] It is understandable that the above-mentioned sound source direction can be the direction of the sound source emitting the target audio relative to the car, and the above-mentioned sound source distance can be the distance between the sound source emitting the target audio and the car.

[0114] After obtaining the audio intensity and arrival time point in each preset partition, the processor of the main acquisition component can compare the audio intensity and the corresponding arrival time point to locate the sound source, obtain the sound source direction and sound source distance of the above-mentioned target audio, and complete the sound source positioning.

[0115] It should be emphasized that the above-mentioned sound source localization can also be achieved through a controllable beamforming localization method or a localization method based on high-resolution spectrum estimation, etc., and this embodiment does not limit this.

[0116] In addition, to achieve the above purpose, the present application embodiment also provides an intelligent driving method, referring to Figure 7 , Figure 7 This is a flowchart of the third embodiment of the intelligent driving method of the present application;

[0117] It should be noted that, in this embodiment, the above-mentioned intelligent driving method is applied to a domain controller in a car, and the car also includes an external audio collection device, and the domain controller is connected to the external audio collection device;

[0118] The method includes:

[0119] Step S40: receiving the sound source positioning result and the target audio sent by the external audio collection device, wherein the target audio is obtained by screening the external audio collected by the external audio collection device, and the sound source positioning result is obtained by performing sound source positioning on the target audio by the external audio collection device;

[0120] Step S50: determining a corresponding target driving strategy according to the target audio and the sound source positioning result, and driving the car according to the target driving strategy.

[0121] It is understandable that the method of this embodiment can be executed by the above-mentioned domain controller. And the domain controller can be connected to the transmission module in the above-mentioned main acquisition component, so as to receive the sound source positioning result and the target audio obtained by the main acquisition component. The specific process can refer to the above-mentioned embodiments, and this embodiment will not be repeated.

[0122] Furthermore, considering that there may be audios in the target audio that are gradually moving away from the car, and these audios will not affect the driving of the car, in this embodiment, the above step S50 includes:

[0123] Step S51: determining the audio type of the target audio through a preset classification model, and determining the sound source direction and sound source distance of the target audio according to the sound source positioning result.

[0124] It should be understood that the above-mentioned preset classification model can be a model for determining the audio type, which may include but is not limited to: a model composed of a decision tree algorithm, a model composed of a hidden Markov algorithm, etc., or a model composed of a convolutional neural network algorithm based on deep learning, a model composed of a time-delay neural network algorithm, etc. This embodiment does not limit this.

[0125] It should also be understood that, considering that the target audio in this embodiment includes vehicle audio and human voice audio, and the vehicle audio may include social vehicle audio and special vehicle audio (such as ambulance, fire truck, etc.). Figure 8 , Figure 8 This is a flow chart of algorithm identification in the third embodiment of the intelligent driving method of this application. Figure 8 As shown, the domain controller may be provided with a preset classification model, through which the target audio is classified to obtain the corresponding audio type (i.e. Figure 8 The algorithm application stage includes the acquisition and preprocessing of external audio signals, feature extraction, classification decision and output recognition results).

[0126] Then, the sound source direction and sound source distance of the target audio can be determined according to the received sound source positioning result.

[0127] It should be noted that the above-mentioned preset classification model can be obtained by pre-training the initial classification model with different types of audio, that is, obtaining sample audio and corresponding type labels, and inputting them into the initial classification model for training after pre-processing and other processes (i.e. Figure 8 In the algorithm learning stage, the external vehicle audio signal collection and processing, feature selection and extraction, sample selection, classifier design, and model learning and optimization are performed to obtain the above-mentioned preset classification model.

[0128] Step S52: determining a distance change of the target audio based on the sound source orientation and the sound source distance;

[0129] Step S53: When the distance change is a close distance, a corresponding target driving strategy is determined based on the audio type, the direction of the sound source, and the distance of the sound source, and the car is driven according to the target driving strategy.

[0130] It is understandable that the above distance change may be a change in the distance between the sound source emitting the target audio and the car.

[0131] In actual use, after obtaining the distance change, the domain controller can determine whether the distance is getting closer; if the distance is getting closer, it means that the sound source is approaching and may need to be avoided. Therefore, the domain controller can determine the corresponding target driving strategy based on the audio type, sound source direction and sound source distance to drive the car to avoid, etc.

[0132] Furthermore, considering that the intelligent driving mode includes an assisted driving mode and an automatic driving mode, the assisted driving mode generally requires driving by a user, while the automatic driving mode is driven by the domain controller. Different driving modes correspond to different driving strategies. Therefore, in this embodiment, the domain controller is also connected to the prompt device of the car.

[0133] It should be noted that the above-mentioned prompting device may be a speaker, an ambient light or other device in a car for prompting the user, and this embodiment does not impose any limitation on this.

[0134] The above step S50 includes:

[0135] Step S54: determining the current driving mode;

[0136] Step S55: When the current driving mode is the assisted driving mode, a corresponding target prompt strategy is determined according to the target audio and the sound source positioning result, and the prompt device is controlled to provide prompts according to the target prompt strategy.

[0137] It is understandable that the above-mentioned target prompt strategy can be a strategy for prompting the user. In this embodiment, the corresponding target prompt strategy can be set according to different audio types, sound source directions and sound source distances.

[0138] In actual use, when the domain controller determines that the current driving mode is the assisted driving mode, it can determine the specific type of vehicle the sound source is based on the audio type, determine the specific position of the vehicle based on the direction of the sound source, and determine the distance from the vehicle based on the distance of the sound source.

[0139] If the present embodiment adopts the atmosphere light as the above-mentioned prompting device, the atmosphere light can be set to display different colors according to different audio types (for example, red for an ambulance and blue for a fire truck); the atmosphere light can be displayed at a specific position according to different sound source directions (for example, the left side of the car is displayed by the atmosphere light on the left A-pillar of the car); the flashing frequency of the atmosphere light can be determined according to the distance from the sound source (for example, the closer the distance, the faster the flashing frequency of the atmosphere light);

[0140] If the present embodiment uses a loudspeaker as the above-mentioned prompting device, the loudspeaker can be controlled to directly broadcast according to the audio type, the direction of the sound source and the distance of the sound source, for example, there is a fire truck 3 meters to the left, etc. The specific broadcasting method is not limited in the present embodiment (i.e. Figure 3 In the prompt, that is Figure 4 Of course, both the ambient light and the speaker can be used, which will not be described in detail in this embodiment.

[0141] Step S56: When the current driving mode is the automatic driving mode, a corresponding target avoidance strategy is determined according to the target audio and the sound source positioning result, and the car is controlled to avoid according to the target avoidance strategy.

[0142] It should be understood that the above-mentioned target avoidance strategy may be a strategy for controlling a car to perform avoidance. In this embodiment, a corresponding target avoidance strategy may be set according to different audio types, sound source directions, and sound source distances.

[0143] In actual use, when the preprocessor determines that the current driving mode is the automatic driving mode, it can determine the type of vehicle that the sound source is based on the audio type, determine the specific location of the vehicle based on the direction of the sound source, and determine the distance from the vehicle based on the distance of the sound source. Then, the width to be avoided is determined based on the audio type, the direction to be avoided is determined based on the direction of the sound source, and the time to be avoided is determined based on the distance of the sound source, thereby generating the above-mentioned target avoidance strategy.

[0144] Then, according to the target avoidance strategy and the autonomous driving system, the system will choose lane change, acceleration and deceleration according to the actual traffic conditions to give priority to special vehicles (i.e. Figure 3 Avoid Figure 4 (automatic avoidance).

[0145] In addition, to achieve the above-mentioned purpose, the embodiment of the present application further provides an external vehicle audio collection device, the external vehicle audio collection device comprising: a main collection component and at least one slave collection component;

[0146] The master acquisition component is connected to the slave acquisition component and the domain controller respectively;

[0147] The secondary collection component is used to collect the external audio outside the car;

[0148] The main collecting component is used to collect the external audio outside the car and receive the external audio transmitted by the slave collecting component;

[0149] The main acquisition component is also used to filter the external audio of the vehicle, obtain the target audio, and locate the sound source of the target audio;

[0150] The main acquisition component is also used to transmit the sound source positioning result and the target audio to the domain controller, so that the domain controller determines the corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

[0151] In this embodiment, an external audio collection device may be provided in the car, and the external audio collection device is connected to the domain controller. In actual use, the external audio collection device may collect external audio outside the car, and filter the external audio to obtain the target audio that affects the driving of the car, and then locate the sound source of the target audio, and transmit the obtained sound source location result and the target audio to the domain controller together, so that the domain controller determines the corresponding target driving strategy to drive the car according to the sound source location result and the target audio. Compared with the existing intelligent driving that does not consider the sound in the external environment, this embodiment can drive in combination with the sound outside the car, which improves safety.

[0152] As an implementation mode, the main acquisition component is also used to extract feature features of the signal features of the external vehicle audio, where the signal features include at least one of frequency features, amplitude features, duration features or spectrum features; mark the external vehicle audio according to the feature extraction results to obtain vehicle audio and human voice audio; and use the vehicle audio and the human voice audio as target audio.

[0153] As an implementation mode, the main acquisition component is also used to perform signal analysis on the target audio, and determine the audio characteristics of the target audio based on the analysis results; determine the corresponding target proportional amplification ratio based on the audio characteristics, and perform signal enhancement on the target audio according to the target proportional amplification ratio; and perform sound source positioning on the target audio after signal enhancement.

[0154] As an embodiment, the main acquisition component is also used to perform partition processing on the target audio according to each preset partition, obtain the audio intensity and arrival time point of the target audio relative to each preset partition; determine the sound source direction and sound source distance of the target audio according to each audio intensity and each arrival time point; and locate the sound source of the target audio based on the sound source direction and the sound source distance.

[0155] Other embodiments or specific implementations of the vehicle external audio collection device of the present application can refer to the above-mentioned method embodiments, which will not be repeated here.

[0156] In addition, to achieve the above-mentioned purpose, the embodiment of the present application further provides a domain controller, wherein the domain controller is connected to an external audio collection device of the vehicle;

[0157] The domain controller is used to receive the sound source positioning result and the target audio sent by the external audio collection device, the target audio is obtained by filtering the external audio collected by the external audio collection device, and the sound source positioning result is obtained by performing sound source positioning on the target audio by the external audio collection device;

[0158] The domain controller is further used to determine a corresponding target driving strategy according to the target audio and the sound source positioning result, and drive the car according to the target driving strategy.

[0159] As an embodiment, the domain controller is also used to determine the audio type of the target audio through a preset classification model, and determine the sound source direction and sound source distance of the target audio according to the sound source positioning result; determine the distance change of the target audio based on the sound source direction and the sound source distance; when the distance change is a close distance, determine the corresponding target driving strategy based on the audio type, sound source direction and sound source distance, and drive the car according to the target driving strategy.

[0160] As an implementation mode, the domain controller is also connected to a prompting device of the car;

[0161] The domain controller is further used to determine the current driving mode;

[0162] When the current driving mode is the assisted driving mode, a corresponding target prompt strategy is determined according to the target audio and the sound source positioning result, and the prompt device is controlled to give prompts according to the target prompt strategy; when the current driving mode is the automatic driving mode, a corresponding target avoidance strategy is determined according to the target audio and the sound source positioning result, and the car is controlled to avoid according to the target avoidance strategy.

[0163] Other embodiments or specific implementations of the domain controller of the present application can refer to the above-mentioned method embodiments and will not be described in detail here.

[0164] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0165] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0166] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0167] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An intelligent driving method, characterized in that: The method is applied to an external audio collection device in a car, the car further comprising a domain controller, the external audio collection device is connected to the domain controller, and the method comprises: collecting external audio outside the car; Filtering the external audio to obtain target audio, and locating the sound source of the target audio; The sound source positioning result and the target audio are transmitted to the domain controller, so that the domain controller determines a corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

2. The method according to claim 1, characterized in that The step of screening the external audio to obtain the target audio includes: Extracting features of the signal of the external audio, wherein the signal features include at least one of a frequency feature, an amplitude feature, a duration feature or a spectrum feature; Marking the external audio according to the feature extraction result to obtain vehicle audio and human voice audio; The vehicle audio and the human voice audio are used as target audio.

3. The method according to claim 1, characterized in that Before the step of locating the sound source of the target audio, the method further includes: Performing signal analysis on the target audio, and determining audio characteristics of the target audio based on the analysis result; Determining a corresponding target proportional amplification ratio based on the audio feature, and performing signal enhancement on the target audio according to the target proportional amplification ratio; The step of locating the sound source of the target audio comprises: Perform sound source positioning on the target audio after signal enhancement.

4. The method according to claim 1, characterized in that The step of locating the sound source of the target audio comprises: Performing partition processing on the target audio according to each preset partition, obtaining the audio intensity and arrival time point of the target audio relative to each preset partition; Determine the sound source direction and sound source distance of the target audio according to each of the audio intensities and each of the arrival time points; The sound source of the target audio is located based on the sound source orientation and the sound source distance.

5. An intelligent driving method, characterized in that: The method is applied to a domain controller in a car, the car further comprising an external audio collection device, the domain controller being connected to the external audio collection device, and the method comprising: Receiving the sound source positioning result and the target audio sent by the external audio collection device, wherein the target audio is obtained by filtering the external audio collected outside the car by the external audio collection device, and the sound source positioning result is obtained by performing sound source positioning on the target audio by the external audio collection device; A corresponding target driving strategy is determined according to the target audio and the sound source positioning result, and the car is driven according to the target driving strategy.

6. The method according to claim 5, characterized in that The step of determining a corresponding target driving strategy according to the target audio and the sound source positioning result, and driving the car according to the target driving strategy includes: Determine the audio type of the target audio through a preset classification model, and determine the sound source direction and sound source distance of the target audio according to the sound source positioning result; Determine a distance change of the target audio based on the sound source orientation and the sound source distance; When the distance change is a close distance, a corresponding target driving strategy is determined based on the audio type, the sound source orientation, and the sound source distance, and the car is driven according to the target driving strategy.

7. The method according to claim 5, characterized in that The domain controller is also connected to a prompting device of the car; The step of determining a corresponding target driving strategy according to the target audio and the sound source positioning result, and driving the car according to the target driving strategy includes: Determine the current driving mode; When the current driving mode is the assisted driving mode, determining a corresponding target prompt strategy according to the target audio and the sound source positioning result, and controlling the prompt device to give a prompt according to the target prompt strategy; When the current driving mode is the automatic driving mode, a corresponding target avoidance strategy is determined according to the target audio and the sound source positioning result, and the car is controlled to avoid according to the target avoidance strategy.

8. An external audio collection device, characterized in that: The vehicle external audio collection device comprises: a main collection component and at least one slave collection component; The master acquisition component is connected to the slave acquisition component and the domain controller respectively; The secondary collection component is used to collect the external audio outside the car; The main collecting component is used to collect the external audio outside the car and receive the external audio transmitted by the slave collecting component; The main acquisition component is also used to filter the external audio of the vehicle, obtain the target audio, and locate the sound source of the target audio; The main acquisition component is also used to transmit the sound source positioning result and the target audio to the domain controller, so that the domain controller determines the corresponding target driving strategy according to the target audio and the sound source positioning result, and drives the car according to the target driving strategy.

9. A domain controller, characterized in that: The domain controller is connected to an external audio collection device; The domain controller is used to receive the sound source positioning result and the target audio sent by the external audio collection device, the target audio is obtained by filtering the external audio collected by the external audio collection device, and the sound source positioning result is obtained by performing sound source positioning on the target audio by the external audio collection device; The domain controller is further used to determine a corresponding target driving strategy according to the target audio and the sound source positioning result, and drive the car according to the target driving strategy.

10. An automobile, characterized in that: The car includes the off-vehicle audio collection device as claimed in claim 8 and the domain controller as claimed in claim 9.

Images