Dangerous driving early warning method and device, vehicle and storage medium

Abstract

The invention relates to a dangerous driving early warning method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining the electroencephalogram data of a driver; processing the electroencephalogram data, and determining data features corresponding to the electroencephalogram data; inputting the data features into a driving behavior recognition model, so that the driving behavior recognition model outputs a target driving type of the driving behavior; and if the target driving type needs to be pre-warned, adopting a pre-warning mode corresponding to the target driving type to perform pre-warning. According to the method and the device, the electroencephalogram data can reflect the condition of the driver more directly, so that the driving type can be identified more timely and accurately, the early warning mode corresponding to the target driving type can be adopted to carry out early warning when the driving type is identified and early warning needs to be carried out, and the driving efficiency is improved. Therefore, the driving danger degree can be determined in real time, corresponding early warning is performed, the time delay is effectively reduced, and the driving safety is improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a dangerous driving warning method, device, vehicle and storage medium. Background Art

[0002] Developing a dangerous driving warning method is of great significance and has broad application prospects. In order to avoid traffic safety problems caused by dangerous driving behaviors, major car manufacturers have successively developed safety warning functions. However, the dangerous driving warning methods in related technologies cannot correctly judge the inconsistency between the driver's inner thoughts and external performance, and are easily interfered by factors such as the driver's appearance characteristics, and some methods have a certain delay. Summary of the invention

[0003] One of the purposes of this application is to provide a dangerous driving warning method, which can provide dangerous driving warning more timely and accurately; the second purpose of this application is to provide a dangerous driving warning device; the third purpose of this application is to provide a vehicle; the fourth purpose of this application is to provide a storage medium.

[0004] In order to achieve the above objectives, in a first aspect, the present application provides a dangerous driving warning method, the dangerous driving warning method comprising:

[0005] Obtain the driver's EEG data;

[0006] Processing the EEG data to determine data features corresponding to the EEG data;

[0007] Inputting the data features into a driving behavior recognition model so that the driving behavior recognition model outputs a target driving type of the driving behavior;

[0008] If the target driving type requires a warning, a warning method corresponding to the target driving type is used for the warning.

[0009] Furthermore, the driving behavior recognition model is determined in the following manner:

[0010] Constructing a training sample set; wherein the training sample set includes a plurality of training sample pairs, the training sample pairs include data feature samples and driving type samples, the EEG data samples corresponding to the data feature samples are the EEG data corresponding to the driving behavior corresponding to the driving type samples;

[0011] The original classification model is trained based on the training sample set to determine the driving behavior recognition model.

[0012] Furthermore, the constructing of the training sample set includes:

[0013] Obtain the standard value of driving data corresponding to normal driving behavior;

[0014] Collecting multiple induced test data in the dangerous driving behavior induced test; wherein the induced test data includes multiple groups of mutually corresponding induced test driving data and induced test EEG data;

[0015] Determining a driving type sample corresponding to the induced test driving data according to the induced test driving data and the driving data standard value;

[0016] Processing the induced test EEG data to determine data feature samples corresponding to the induced test EEG data;

[0017] Determining the driving type samples and the data feature samples corresponding to each other based on the induced test driving data and the induced test EEG data corresponding to each other;

[0018] Constructing the driving type samples and the data feature samples corresponding to each other into a training sample pair;

[0019] A plurality of the training sample pairs are constructed as the training sample set.

[0020] Furthermore, the obtaining of the standard value of driving data corresponding to the normal driving behavior includes:

[0021] Collecting non-interference test data in the non-interference test; wherein the non-interference test data includes non-interference driving data;

[0022] Determining data other than the abnormal driving data in the non-interference driving data as the normal driving data;

[0023] Based on a plurality of sets of the normal driving data, the driving data standard value is determined.

[0024] Furthermore, determining the driving type sample corresponding to the induced test driving data according to the induced test driving data and the driving data standard value includes:

[0025] Determining a target Euclidean distance interval corresponding to the induced test driving data according to the induced test driving data and the driving data standard value;

[0026] The driving type corresponding to the target Euclidean distance interval in the setting configuration information is determined as the driving type sample; wherein the setting configuration information includes a correspondence between a plurality of Euclidean distance intervals and driving types.

[0027] Furthermore, the collecting of multiple induced test data in the dangerous driving behavior induced test includes:

[0028] During the dangerous driving behavior induction test, the test data within a set time interval is used as the induction test data.

[0029] Furthermore, the data feature includes at least one of the following:

[0030] Amplitude of alpha waves, amplitude of beta waves, amplitude of gamma waves, amplitude of delta waves, amplitude of theta waves, logarithm of power, power spectral density, differential entropy, energy in frequency domain.

[0031] In order to achieve the above-mentioned purpose, in a second aspect, the present application further provides a dangerous driving warning device, the dangerous driving warning device comprising:

[0032] An acquisition module, used for acquiring the driver's EEG data;

[0033] A determination module, used to process the EEG data and determine data features corresponding to the EEG data;

[0034] A recognition module, used for inputting the data features into a driving behavior recognition model so that the driving behavior recognition model outputs a target driving type of the driving behavior;

[0035] The warning module is used to provide a warning in a warning manner corresponding to the target driving type if the target driving type requires a warning.

[0036] To achieve the above objectives, in a third aspect, the present application also provides a vehicle, comprising: a processor and a memory, wherein the processor is used to execute a control program stored in the memory to implement the dangerous driving warning method as described above.

[0037] To achieve the above objectives, in a fourth aspect, the present application also provides a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the dangerous driving warning method as described above.

[0038] Beneficial effects of this application:

[0039] In this application, the driving type corresponding to the driver's driving behavior can be determined based on the driver's EEG data and the driving behavior recognition model. Among them, the EEG data can more directly reflect the driver's situation, so that the driving type can be identified more timely and accurately. Then, when the driving type needs to be warned, the warning method corresponding to the target driving type can be used to warn, so that the degree of driving danger can be determined in real time and corresponding warnings can be issued, effectively reducing delays and improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram showing a flow chart of a dangerous driving warning method provided in an embodiment of the present application;

[0041] Figure 2 A schematic diagram showing the structure of a dangerous driving warning device provided in an embodiment of the present application is shown;

[0042] Figure 3 A schematic diagram of the structure of a vehicle provided by an embodiment of the present application is shown;

[0043] in:

[0044] 10. Acquisition module; 20. Determination module; 30. Identification module; 40. Warning module;

[0045] 100. Vehicle; 101. Processor; 102. Memory; 1021. Operating system; 1022. Application; 103. User interface; 104. Network interface; 105. Bus system. DETAILED DESCRIPTION

[0046] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0047] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0048] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0049] To facilitate understanding of the embodiments of the present application, further explanation will be given below with reference to specific embodiments in conjunction with the accompanying drawings. The embodiments do not constitute a limitation on the embodiments of the present application.

[0050] This embodiment provides a dangerous driving warning method, which can be applied to vehicles. Figure 1 As shown, the method may include:

[0051] S110, obtaining the driver's EEG data;

[0052] S120, processing the EEG data to determine data features corresponding to the EEG data;

[0053] S130, inputting the data features into a driving behavior recognition model so that the driving behavior recognition model outputs a target driving type of the driving behavior;

[0054] S140: If the target driving type requires a warning, a warning method corresponding to the target driving type is used for the warning.

[0055] In step S110, the EEG data may also be referred to as brain waves, which are bioelectric signals generated when information is transmitted between neurons. There are five types of brain waves, which are arranged in order from low to high frequency. These five types of brain waves are alpha waves (alpha waves), beta waves (beta waves), gamma waves (gamma waves), delta waves (delta waves) and theta waves (theta waves).

[0056] In this step, the driver can wear a wireless EEG acquisition device during driving to monitor his EEG data. According to the international 10-20 system arrangement, the wireless EEG acquisition device in this application can refer to an 8-channel, 16-channel, 32-channel or 64-channel wireless EEG acquisition device. The wireless EEG acquisition device collects the voltage value of each point.

[0057] It should be noted that in addition to obtaining the driver's EEG data through wireless EEG acquisition equipment, the vehicle can also obtain the driver's EEG data through other means, without limitation.

[0058] In step S120, after the vehicle obtains the driver's EEG data, it can perform fast Fourier transform and wavelet transform on the EEG data to obtain data features corresponding to the EEG data.

[0059] The data characteristics may include at least one of the following:

[0060] Amplitude of alpha waves, amplitude of beta waves, amplitude of gamma waves, amplitude of delta waves, amplitude of theta waves, logarithm of power, power spectral density, differential entropy, energy in frequency domain.

[0061] For example, after the vehicle obtains the driver's real-time EEG data, it can process the EEG data by fast Fourier transform and wavelet transform, thereby obtaining the alpha wave amplitude, beta wave amplitude, gamma wave amplitude, delta wave amplitude, theta wave amplitude, power logarithm, power spectral density, differential entropy, frequency domain energy and other data characteristics corresponding to the EEG data.

[0062] It should be noted that, in addition to obtaining the data features corresponding to the EEG data in the above-mentioned manner, the data features corresponding to the EEG data may also be obtained in other manners, which are not limited to this.

[0063] In step S130, a driving behavior recognition model may be preset in the vehicle. In this step, after the vehicle obtains the data features corresponding to the EEG data, the data features may be input into the driving behavior recognition model, and the driving behavior recognition model may output the driving type corresponding to the above data features, and the driving type may be recorded as the target driving type.

[0064] The driving type may be identified by at least one number, at least one letter, or a combination of numbers and letters, and there is no limitation on this.

[0065] The driving behavior recognition model can be trained in advance and configured in the vehicle. It can be set according to the actual situation, and there is no limitation on its specific model architecture. The driving behavior recognition model can be set before the vehicle leaves the factory or after the vehicle leaves the factory, and there is no limitation on this. After the driving behavior recognition model is set, it can be modified later to better meet different needs.

[0066] It should be noted that the driving type that the driving behavior recognition model can be used to identify can be set according to actual needs and is not limited to this. For example, the driving behavior recognition model can be used to identify driving behaviors of multiple driving types, such as normal driving, mildly dangerous driving, and severely dangerous driving. Of course, the driving behavior recognition model can also be used to identify driving behaviors of other driving types.

[0067] In step S140, when the driving behavior recognition model outputs the target driving type, the vehicle can take corresponding treatment measures based on the specific situation of the target driving type. If the target driving type requires an early warning, an early warning method corresponding to the target driving type is used for early warning.

[0068] Among them, when the driving type may include normal driving, mildly dangerous driving and severely dangerous driving. The corresponding treatment measure for normal driving is no warning; the corresponding treatment measure for mildly dangerous driving is to issue an alarm (this treatment measure is the warning method corresponding to mildly dangerous driving); the corresponding treatment measure for severely dangerous driving is to issue an alarm and take automatic driving and other forms to stop the vehicle in a safe area (this treatment measure is the warning method corresponding to severely dangerous driving). It should be noted that the specific content and form of the alarm can be set according to actual needs and is not limited to this.

[0069] Among them, during the driving process, when the driver's current driving behavior is determined to be normal driving, there will be no warning; when the driver's current driving behavior is determined to be mildly dangerous driving, an alarm will be issued; when the driver's current driving behavior is determined to be severely dangerous driving, an alarm will be issued and automatic driving will be adopted to stop the vehicle in a safe area.

[0070] It should be noted that in addition to the above-mentioned warning methods, other warning methods can also be set, and there is no limitation on this.

[0071] This method can determine the driving type corresponding to the driver's driving behavior based on the driver's EEG data and the driving behavior recognition model. Among them, EEG data can more directly reflect the driver's situation, so that the driving type can be recognized more timely and accurately. Then, when the driving type needs to be warned, the warning method corresponding to the target driving type can be used to warn, so that the degree of driving danger can be determined in real time and corresponding warnings can be issued, effectively reducing delays and improving driving safety.

[0072] This embodiment provides a dangerous driving warning method, which can be applied to vehicles. In this method, the driving behavior recognition model is determined by:

[0073] S210, constructing a training sample set; wherein the training sample set includes a plurality of training sample pairs, the training sample pairs include data feature samples and driving type samples, the EEG data samples corresponding to the data feature samples are EEG data corresponding to the driving behavior corresponding to the driving type samples;

[0074] S220: Train the original classification model based on the training sample set to determine a driving behavior recognition model.

[0075] In step S210, the driving data corresponding to different driving behaviors may be obtained first, and then the driving type corresponding to the driving behavior may be determined based on the driving data, which is a driving type sample. In addition, the EEG data of the driver corresponding to the above-mentioned driving behavior may be obtained at the same time, and the EEG data may be processed to determine the data features, which is a data feature sample. In this step, the data feature samples and driving type samples corresponding to the same driving behavior may be recorded as a training sample pair. Among them, the data feature samples serve as the input samples of the training sample pairs, and the driving type samples serve as the output samples of the training sample pairs. In the above manner, multiple training sample pairs may be determined, thereby obtaining a training sample set consisting of multiple training sample pairs.

[0076] It should be noted that, generally speaking, the richer the data in the training sample set, the better the reliability of the driving behavior recognition model trained based on the training sample set.

[0077] In step S220, the architecture of the original classification model can be set according to actual needs, and its specific architecture may not be limited. For example, a support vector machine (SVM) may be selected as the original classification model.

[0078] In this step, after obtaining the training sample set, the original classification model can be trained based on the training sample set to obtain the required driving behavior recognition model.

[0079] In this method, a driving behavior recognition model is determined by training sample pairs consisting of corresponding data features and driving types, and a large number of training sample pairs are used to train the original classification model. This can better ensure the reliability of the driving behavior recognition model and improve the reliability of the dangerous driving warning method, so that the degree of driving danger can be determined in real time and corresponding warnings can be issued, effectively reducing delays and improving driving safety.

[0080] This embodiment provides a dangerous driving warning method, which can be applied to vehicles. In this method, constructing a training sample set may include:

[0081] S310, obtaining a standard value of driving data corresponding to a normal driving behavior;

[0082] S320, collecting multiple induced test data in the dangerous driving behavior induced test; wherein the induced test data includes multiple groups of mutually corresponding induced test driving data and induced test EEG data;

[0083] S330, determining a driving type sample corresponding to the induced test driving data according to the induced test driving data and the driving data standard value;

[0084] S340, processing the induced test EEG data to determine data feature samples corresponding to the induced test EEG data;

[0085] S350, determining corresponding driving type samples and data feature samples based on the corresponding induced test driving data and induced test EEG data;

[0086] S360, constructing a training sample pair from the corresponding driving type samples and data feature samples;

[0087] S370: Construct multiple training sample pairs into a training sample set.

[0088] In step S310, the standard value of driving data corresponding to normal driving behavior can be determined based on experiments, or obtained from other locations (such as the cloud or other vehicles), without limitation.

[0089] When determining the standard value of driving data based on the test, the non-interference test data in the non-interference test may be collected first; wherein the non-interference test data includes the non-interference driving data. Then, the abnormal driving data in the non-interference driving data is eliminated, that is, the data other than the abnormal driving data in the non-interference driving data is determined as the normal driving data. Then, based on multiple groups of normal driving data, the standard value of driving data is determined. For example, the average value of the multiple groups of normal driving data may be aggregated to obtain the standard value of driving data corresponding to the normal driving behavior.

[0090] In some embodiments,

[0091] A 20-minute non-interference test can be conducted, that is, a driving simulation test without any interference. The subjects continue to simulate driving, and the subjects are assumed to be in a neutral, sober, and focused driving mood. During the test, a questionnaire survey can be conducted every 5 minutes (or other durations can be set according to needs) to determine whether the subjects think the current state is normal driving.

[0092] In the undisturbed driving data collected during the undisturbed test period, the driver questionnaire results are displayed as data collected during the normal driving period and marked as normal driving data. Then, the normal driving data is aggregated and the average value after aggregation is taken as the standard value of the driving data corresponding to normal driving behavior.

[0093] It should be noted that, in addition to determining the standard value of driving data by the above method, the standard value of driving data may also be determined by other methods, which are not limited to this.

[0094] In step S320, a dangerous driving behavior induction test may be conducted on the driver (also referred to as the subject), and the induced test driving data and induced test EEG data during the above test process are collected and recorded as induced test data.

[0095] In the process of the dangerous driving behavior induction test, the test data within the set time interval is used as an induced test data, thereby obtaining a plurality of induced test data. The set time interval can be set according to actual needs, and its specific value may not be limited.

[0096] In some embodiments,

[0097] The set time interval can be 60s. In the dangerous driving behavior induction test, the subject continues to use the driving simulator to simulate driving, and randomly adds visual n-back and auditory n-back interference to induce the subject's distraction, and randomly adds negative music and audio to induce the subject's negative emotions. During the continuous simulated driving process, the KSS scale can be used every 20 minutes to confirm the subject's fatigue state. According to experience, 90 minutes of simulated driving is sufficient to induce subject fatigue. After the end, a questionnaire survey is conducted to determine whether the driver believes that there is dangerous driving behavior in this experiment. If it is believed that there is no dangerous driving behavior, the degree of interference should be increased and the experiment should be repeated until the test data when dangerous driving behavior exists is obtained. In the test data, the test data of every 60 seconds can be used as an induced test data, thereby obtaining multiple induced test data. The induced test data may include corresponding induced test driving data and induced test EEG data.

[0098] Among them, regarding the collection of evoked test EEG data, a 0.1-30Hz FIR filter can be used to digitally filter the data to reduce low-frequency drift and high-frequency noise. The downsampling is 200Hz. The re-reference is the average of the whole head. When there are enough electrodes to cover the entire head, it is considered to be one of the best re-reference methods. Among them, the electrodes judged to be bad tracks are replaced by a spherical interpolation algorithm. Independent component analysis removes artifacts such as eye movements, muscles, and head movements. After the preprocessing is completed, the data with data loss and data quality problems will be further eliminated according to the preprocessing results, so as to obtain the final multiple evoked test EEG data.

[0099] It should be noted that, in addition to obtaining the induced test data through the above-mentioned test method, the induced test data can also be obtained through other methods, which are not limited to this.

[0100] In step S330, after the induced test data is obtained, the driving type corresponding to the induced test driving data can be determined based on the induced test driving data in the induced test data and the driving data standard value. The driving type can be recorded as a driving type sample.

[0101] The data can be divided by the Euclidean distance threshold, so as to determine the driving type corresponding to the induced test driving data.

[0102] In some embodiments,

[0103] Driving data may include acceleration a, speed b, headway c, lane departure d and other data. The standard value of driving data can be recorded as (a 0 , b 0 、c 0 d 0 ), the driving data of the nth induced test can be recorded as (a n , b n 、c n d n ).

[0104] It should be noted that under normal driving conditions, the frontal, parietal, occipital and temporal regions of the brain are all active, beta waves and the logarithm of beta wave power are significantly correlated with driving behavior, and acceleration, speed and headway are the most relevant driving behavior characteristic quantities for brain perception activities. In addition, the most concerned issue in the study of dangerous driving behavior is lane deviation caused by following too closely, i.e. headway and abnormal lane change. Therefore, this implementation method can select four items, i.e. acceleration, speed, headway and lane deviation, as driving data.

[0105] If the driving data fluctuates within 20% based on the standard value, it can be recorded as normal driving; if the driving data fluctuates between 20%-50% based on the standard value, it can be recorded as mildly dangerous driving; if the driving data fluctuates by more than 50% based on the standard value, it can be recorded as severely dangerous driving.

[0106] The absolute value of the driving data difference when the driving data standard value fluctuates by 20% is recorded as (a 1 , b 1 、c 1 d 1 ), the absolute value of the driving data difference when the driving data standard value fluctuates by 50% is (a 2 , b 2 、c 2 d 2 ).

[0107] The Euclidean distance threshold can be determined as follows:

[0108]

[0109] Among them, x 1 is the Euclidean distance threshold corresponding to the driving data when the standard value of the driving data fluctuates by 20%, x 2 It is the Euclidean distance threshold corresponding to the driving data when the standard value of the driving data fluctuates by 50%.

[0110] In this implementation, the Euclidean distance corresponding to the nth induced test driving data is:

[0111]

[0112] Where d is the Euclidean distance.

[0113] After obtaining the Euclidean distance corresponding to the induced test driving data, the Euclidean distance threshold (x 1 and x 2 ) to determine the driving type corresponding to the induced test driving data.

[0114] The driving type corresponding to each induced test driving data can be determined based on Table 1. Referring to Table 1, when the Euclidean distance of the induced test driving data is less than or equal to x 1 When the driving type corresponding to the induced test driving data is normal driving, its label can be recorded as 0; when the Euclidean distance of the induced test driving data is greater than x 1 and less than or equal to x 2 When the driving type corresponding to the induced test driving data is mildly dangerous driving, its label can be recorded as 1; when the Euclidean distance of the induced test driving data is greater than x 2 When , the driving type corresponding to the induced test driving data is severe dangerous driving, and its label can be recorded as 2.

[0115] Table 1

[0116] Euclidean distance interval Dangerous driving level Label <![CDATA[d≤x 1 ]]> Normal driving 0 <![CDATA[x 1 <d≤x 2 ]]> Mildly dangerous driving 1 <![CDATA[x 2 <d]]> Severely dangerous driving 2

[0117] It should be noted that, in the present application, in addition to determining the driving type corresponding to the induced test driving data in the above manner, it can also be determined in other ways, which are not limited to this.

[0118] In step S340, the evoked test EEG data is processed to obtain corresponding data features, which can be recorded as data feature samples. The processing method of the evoked test EEG data can refer to step S120 in other embodiments, which will not be described in detail.

[0119] In step S350, after obtaining the driving type samples corresponding to the induced test driving data and the data feature samples corresponding to the induced test EEG data, the driving type samples corresponding to the induced test driving data and the data feature samples corresponding to the induced test EEG data that correspond to each other can be determined as the corresponding driving type samples and data feature samples.

[0120] That is to say, after obtaining the driving type samples corresponding to the induced test driving data and the data feature samples corresponding to the induced test EEG data, the driving type samples corresponding to the induced test driving data and the data feature samples corresponding to the induced test EEG data in the same induced test data can be determined as mutually corresponding driving type samples and data feature samples.

[0121] In step S360, after obtaining the corresponding driving type samples and data feature samples, the data feature samples can be used as input samples of the training sample pair, and the driving type samples can be used as output samples of the training sample pair, thereby obtaining a training sample pair.

[0122] In step S370, after obtaining a plurality of training sample pairs, the plurality of training sample pairs may be constructed into a required training sample set, so as to facilitate subsequent model training based on the training sample set, thereby obtaining a driving behavior recognition model.

[0123] It should be noted that for the non-interference test and the dangerous driving behavior induction test, the test environment can be a room with no noise and no large electromagnetic radiation source. In addition, the test environment can be large enough to accommodate driving simulators and EEG acquisition equipment (such as wireless EEG acquisition equipment). The subjects are healthy people, without a history of basic cardiovascular and cerebrovascular diseases and neurological diseases, and have not stayed up late in the past two days. Before the test, a fixed scene needs to be set for the driving simulator, and the subjects can wear a 64-channel wireless EEG acquisition device to monitor the driver's EEG data in real time. In addition, the subjects can be tested to perform the psychomotor vigilance task (PVT) and the Karolinka sleepiness test (KDT), and the Karolinka sleepiness scale and sleepiness scale tests. At the same time, after the collected data is input into the EEG fatigue algorithm, the classification result is awake to ensure that the subject is in a sober state. In addition, the subject can be evaluated by the emotional self-assessment SAM scale, and at the same time, after the collected data is input into the emotional algorithm, the classification result is positive to ensure that the driver's emotions are positive. Through the above preparations, the reliability of the test data can be better guaranteed.

[0124] In this method, the EEG data is calibrated by driving data, and the degree of dangerous driving is directly judged by the EEG data, which improves the real-time warning of dangerous driving behavior while ensuring verifiability. In addition, the training data in this method comes from simulated driving tests, and the subjective and objective combination ensures that the collected data contains normal driving and dangerous driving situations. Among them, normal driving data can be obtained through non-interference tests, and the standard value of driving data can be obtained after clustering and averaging. Then, the driving type can be divided by the Euclidean distance threshold, and the corresponding EEG data can be calibrated by the driving type to ensure the accurate correspondence between the EEG data and the driving type, so as to obtain a more reliable training sample pair. In addition, the data collection and processing method provided by this method, the selected driving data may include acceleration, speed, head spacing and lane deviation, etc., which include the driving behavior performance of most driving safety accidents, and in the dangerous driving behavior induction test, the induction stage of dangerous driving behavior may include fatigue, negative emotions, and distraction state random induction, the obtained driving data and EEG data are close to the real abnormal driving situation, and the data can be applied in a wide range.

[0125] This method can obtain a more reliable driving behavior recognition model. Based on the above driving behavior recognition model and the driver's real-time EEG data, the degree of driving danger can be determined in real time and corresponding warnings can be issued, effectively reducing delays and improving driving safety.

[0126] This embodiment provides a dangerous driving warning device, which can be applied to a vehicle. The device can be used to implement the above-mentioned dangerous driving warning method. For example, refer to Figure 2 As shown, the device may include an acquisition module 10 , a determination module 20 , an identification module 30 and an early warning module 40 .

[0127] The acquisition module 10 is used to acquire the driver's EEG data.

[0128] The determination module 20 is used to process the EEG data and determine data features corresponding to the EEG data.

[0129] The recognition module 30 is used to input the data features into the driving behavior recognition model so that the driving behavior recognition model outputs the target driving type of the driving behavior.

[0130] The warning module 40 is used to provide a warning in a warning manner corresponding to the target driving type if a warning is required for the target driving type.

[0131] This embodiment provides a dangerous driving warning device, which can be applied to a vehicle. In the device, the data features include at least one of the following:

[0132] Amplitude of alpha waves, amplitude of beta waves, amplitude of gamma waves, amplitude of delta waves, amplitude of theta waves, logarithm of power, power spectral density, differential entropy, energy in frequency domain.

[0133] This embodiment provides a vehicle, which may be a pure electric vehicle, a hybrid electric vehicle, or other vehicles including a motor, without limitation.

[0134] refer to Figure 3 As shown, the vehicle 100 includes: at least one processor 101, a memory 102, at least one network interface 104 and other user interfaces 103. The various components in the vehicle 100 are coupled together through a bus system 105. It can be understood that the bus system 105 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 105 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, all buses are labeled as bus system 105.

[0135] The user interface 103 may include a display, a keyboard, or a pointing device (eg, a mouse, a trackball, a touch pad, or a touch screen).

[0136] It can be understood that the memory 102 in the embodiment of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM). The memory 102 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0137] In some implementations, the memory 102 stores the following elements, executable units or data structures, or a subset thereof, or an extended set thereof: an operating system 1021 and application programs 1022 .

[0138] Among them, the operating system 1021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application 1022 includes various application programs, such as a media player (Media Player), a browser (Browser), etc., which are used to implement various application services. The program that implements the method of the embodiment of the present application can be included in the application 1022.

[0139] In the embodiment of the present application, by calling the program or instruction stored in the memory 102, specifically, the program or instruction stored in the application 1022, the processor 101 is used to execute the method provided by each method embodiment.

[0140] The method disclosed in the above embodiment of the present application can be applied to the processor 101, or implemented by the processor 101. The processor 101 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 101 or the instruction in the form of software. The above processor 101 can be a general processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to execute, or the hardware and software units in the decoding processor can be executed. The software unit can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 102, and the processor 101 reads the information in the memory 102 and completes the above method in combination with its hardware.

[0141] It is understood that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit may be implemented in one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing vehicles (DSPDevice, DSPD), programmable logic vehicles (PLD), field programmable gate arrays (FPGA), general purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described in the present application, or a combination thereof.

[0142] For software implementation, the technology described herein can be implemented by a unit that performs the functions described herein. The software code can be stored in a memory and executed by a processor. The memory can be implemented in the processor or outside the processor.

[0143] The embodiment of the present application also provides a storage medium (computer-readable storage medium). The storage medium here stores one or more programs. The storage medium may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a read-only memory, a flash memory, a hard disk or a solid-state drive; the memory may also include a combination of the above-mentioned types of memory.

[0144] When one or more programs in the storage medium can be executed by one or more processors to implement the above-mentioned method executed in the vehicle.

[0145] The processor is used to execute the vehicle control program stored in the memory to implement the above-mentioned method of execution in the vehicle.

[0146] The professionals should also be further aware that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0147] It should be noted that the phrases "one implementation", "an embodiment", "an exemplary embodiment", "some embodiments", etc. mentioned in the specification indicate that the described embodiments may include certain features, structures or characteristics, but not every embodiment may include the certain features, structures or characteristics. In addition, such phrases do not necessarily refer to the same embodiment. In addition, when describing certain features, structures or characteristics in conjunction with an embodiment, it is within the knowledge of those skilled in the art to implement such features, structures or characteristics in conjunction with other embodiments, whether explicitly or not explicitly described.

[0148] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or vehicle 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 vehicle. In the absence of more limitations, the elements defined by the sentence "comprises one..." do not exclude the presence of other identical elements in the process, method, article or vehicle including the elements.

[0149] The above embodiments are only preferred embodiments for fully illustrating the present application, and the protection scope of the present application is not limited thereto. Any equivalent substitution or change made by a person skilled in the art based on the present application is within the protection scope of the present application.

Claims

1. A dangerous driving warning method, characterized in that: The dangerous driving warning method comprises: Obtain the driver's EEG data; Processing the EEG data to determine data features corresponding to the EEG data; Inputting the data features into a driving behavior recognition model so that the driving behavior recognition model outputs a target driving type of the driving behavior; If the target driving type requires a warning, a warning method corresponding to the target driving type is used for the warning.

2. The dangerous driving warning method according to claim 1, characterized in that: The driving behavior recognition model is determined by: Constructing a training sample set; wherein the training sample set includes a plurality of training sample pairs, the training sample pairs include data feature samples and driving type samples, the EEG data samples corresponding to the data feature samples are the EEG data corresponding to the driving behavior corresponding to the driving type samples; The original classification model is trained based on the training sample set to determine the driving behavior recognition model.

3. The dangerous driving warning method according to claim 2, characterized in that: The step of constructing a training sample set includes: Obtain the standard value of driving data corresponding to normal driving behavior; Collecting multiple induced test data in the dangerous driving behavior induced test; wherein the induced test data includes multiple groups of mutually corresponding induced test driving data and induced test EEG data; Determining a driving type sample corresponding to the induced test driving data according to the induced test driving data and the driving data standard value; Processing the induced test EEG data to determine data feature samples corresponding to the induced test EEG data; Determining the driving type samples and the data feature samples corresponding to each other based on the induced test driving data and the induced test EEG data corresponding to each other; Constructing the driving type samples and the data feature samples corresponding to each other into a training sample pair; A plurality of the training sample pairs are constructed as the training sample set.

4. The dangerous driving warning method according to claim 3, characterized in that: The step of obtaining the standard value of driving data corresponding to the normal driving behavior includes: Collecting non-interference test data in the non-interference test; wherein the non-interference test data includes non-interference driving data; Determining data other than the abnormal driving data in the non-interference driving data as the normal driving data; Based on a plurality of sets of the normal driving data, the driving data standard value is determined.

5. The dangerous driving warning method according to claim 3, characterized in that: The step of determining the driving type sample corresponding to the induced test driving data according to the induced test driving data and the driving data standard value includes: Determining a target Euclidean distance interval corresponding to the induced test driving data according to the induced test driving data and the driving data standard value; The driving type corresponding to the target Euclidean distance interval in the setting configuration information is determined as the driving type sample; wherein the setting configuration information includes a correspondence between a plurality of Euclidean distance intervals and driving types.

6. The dangerous driving warning method according to claim 3, characterized in that: The collecting of multiple induced test data in the dangerous driving behavior induced test includes: During the dangerous driving behavior induction test, the test data within a set time interval is used as the induction test data.

7. The dangerous driving warning method according to any one of claims 1 to 6, characterized in that: The data features include at least one of the following: Amplitude of alpha waves, amplitude of beta waves, amplitude of gamma waves, amplitude of delta waves, amplitude of theta waves, logarithm of power, power spectral density, differential entropy, energy in frequency domain.

8. A dangerous driving warning device, characterized in that: The dangerous driving warning device comprises: An acquisition module, used for acquiring the driver's EEG data; A determination module, used to process the EEG data and determine data features corresponding to the EEG data; A recognition module, used for inputting the data features into a driving behavior recognition model so that the driving behavior recognition model outputs a target driving type of the driving behavior; The warning module is used to provide a warning in a warning manner corresponding to the target driving type if the target driving type requires a warning.

9. A vehicle, characterized in that: include: A processor and a memory, wherein the processor is used to execute a control program stored in the memory to implement the dangerous driving warning method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the dangerous driving warning method described in any one of claims 1-7.

Images