Driver driving anger evaluation and intervention system and method

Abstract

The invention discloses a driver driving anger evaluation and intervention system and method, and the system comprises a VR driving simulator and an information platform, and the VR driving simulator and the information platform are in real-time communication through network connection. The VR driving simulator serves as an operation platform of a driver and is used for displaying preset scene information to the driver through the VR head-mounted display device and collecting driving behavior data, electrocardio data and electroencephalogram data when the driver simulates driving. And the information platform is used for performing feature extraction according to the collected driving behavior data, the electrocardio data and the electroencephalogram data, obtaining the driving angry intensity level of the driver according to the extracted features, and determining the driver angry intervention measure according to the driving angry intensity level. According to the invention, the VR driving simulator is adopted for driving test, the driving anger condition encountered in an actual road can be more comprehensively simulated, and the method has the advantages of good stability, small interference, safety and the like, and can be widely applied to the field of driving anger evaluation.

Description

Technical Field

[0001] The present invention relates to the field of driving anger assessment, and in particular to a driver's driving anger assessment and intervention system and method. Background Art

[0002] In recent years, with the increase in the number of cars on the road and the increase in people's life pressure, drivers' road rage has become more and more serious, and traffic accidents caused by "road rage" have also increased year by year. Facing the same driving environment, different drivers will have different emotional reactions and different operating behaviors. Therefore, it is extremely valuable to detect and evaluate drivers' road rage, quantify the intensity of driving anger, and output targeted intervention measures.

[0003] At present, in the study of driver driving anger, on the one hand, scholars use EEG, face, voice, physiological information and driving operation behavior to identify or detect the driver's anger. For example, Chinese patent: CN116129405A (application number: 2022115002659) discloses a multimodal hybrid fusion method for driver anger emotion recognition, which collects the driver's ECG signal, car driving parameters and driver's facial image, and uses the driver emotion recognition model based on the random forest algorithm and the driver emotion recognition model based on the convolutional neural network and the sofemax classifier for emotion recognition. The driver emotion recognition results obtained by the two emotion recognition models are fused through the DS evidence theory, and finally the driver's emotion recognition results are obtained. On the other hand, scholars give real-time warnings to the driver's anger based on emotion recognition and output real-time appeasement measures or connect the vehicle to the automatic driving system. For example, Chinese patent: CN117584985B (application number: 202410072074X) discloses a driving anger index detection and vehicle control method and system based on driving status, which performs feature extraction, time series change point detection and forgetting, and calculation of driving anger index on vehicle driving status data, and timely controls the vehicle's throttle, brake, voice and aromatherapy system according to the anger index to adjust emotional soothing and vehicle control parameters.

[0004] In summary, in the relevant technologies, most of them are about identifying the driver's anger emotions, while ignoring the evaluation of the intensity of driving anger; and in existing research, it is usually carried out under real car driving conditions, which is dangerous, and the collected data is easily disturbed by other emotions such as tension, and the road driving anger scenes are limited, which cannot fully and fairly reflect the driver's driving anger level; at the same time, in existing research, there are few studies that evaluate the driver's driving anger intensity and output levels and corresponding intervention measures. Summary of the invention

[0005] In order to solve at least one of the technical problems existing in the prior art to a certain extent, the object of the present invention is to provide a driver's driving anger assessment and intervention system and method based on VR simulated driving.

[0006] The first technical solution adopted by the present invention is:

[0007] A driver driving anger assessment and intervention system, comprising a VR driving simulator and an information platform, wherein the VR driving simulator and the information platform are connected via a network for real-time communication;

[0008] The VR driving simulator serves as an operating platform for the driver, and is used to display preset scene information to the driver through a VR head display device, collect driving behavior data, electrocardiogram data, and electroencephalogram data of the driver during simulated driving, and perform feature extraction on the collected driving behavior data, electrocardiogram data, and electroencephalogram data;

[0009] The information platform is used to obtain the driver's driving anger intensity level according to the extracted features, and determine the driver's anger intervention measures according to the driving anger intensity level.

[0010] Furthermore, the VR driving simulator includes a data acquisition module and a data analysis module.

[0011] The data acquisition module includes an electrocardiogram acquisition device, an electroencephalogram acquisition device, and a plurality of sensors arranged on the driving control device, which are used to collect the driver's electrocardiogram, electroencephalogram and driving behavior data in real time;

[0012] The data analysis module is used to pre-process and extract features from the collected driver's electrocardiogram, electroencephalogram and driving behavior data;

[0013] The information platform includes a personal information module, a simulated driving module, a driving anger assessment module and an intervention module;

[0014] The personal information module is used to collect and store the driver's personal information, assessment results and intervention measures; the personal information includes name, age, gender, driving experience, driving mileage, and number of violations;

[0015] The simulated driving module is used to establish an electronic information database of VR road traffic driving anger scenes through three-dimensional modeling based on historical classic driving anger scenes, and display the scene information through a VR head display device;

[0016] The driving anger assessment module is used to input the extracted features into a pre-trained driving anger assessment model based on a Bayesian network to obtain the driving anger intensity level of the driver;

[0017] The intervention module is used to determine intervention measures for driver anger according to the anger intensity level output by the driving anger assessment module and the corresponding relationship between the anger intensity level and the emotion intervention method set in advance.

[0018] Furthermore, the connection method of each module of the system is: the personal information module is connected to the simulation driving module, the simulation driving module is connected to the data acquisition module, the data acquisition module is connected to the data analysis module, the data analysis module is connected to the driving anger assessment module and connected to the personal information module, and the driving anger assessment module is connected to the intervention module and connected to the personal information module.

[0019] Further, the data analysis module includes an electrocardiogram analysis unit, an electroencephalogram analysis unit, and a driving behavior analysis unit;

[0020] The ECG analysis unit is used to pre-process the ECG signal and extract the RR interval mean value in the ECG signal. RR , the percentage of the number of adjacent RR intervals whose difference is greater than the preset value to the total number of RR intervals pNN 20 , RR interval standard deviation H RR , the ratio of low frequency power to high frequency power LF\HF data characteristics;

[0021] The EEG analysis unit is used to pre-process the EEG signal and extract data features of the relative power spectrum β% of the β wave and the relative power spectrum θ% of the θ wave in the EEG signal;

[0022] The driving behavior analysis unit is used to pre-process the driving operation data and extract the vehicle distance ξ(t), steering wheel turning entropy H p , vehicle longitudinal acceleration |a|, vehicle lateral acceleration LA, and accelerator pedal speed SGP data characteristics.

[0023] Furthermore, the RR interval mean value Mean RR The calculation formula is:

[0024]

[0025] Where N is the number of RR intervals in the sampling period, RR i represents the i-th RR interval;

[0026] Percent pNN 20 The calculation formula is:

[0027]

[0028] In the formula, NN 20 The number of times the difference between two adjacent RR intervals in the sampling period is greater than 20ms;

[0029] The formula for calculating the power in the low frequency band is:

[0030]

[0031] The formula for calculating the power in the high frequency band is:

[0032]

[0033] Where PSD(f) is the power spectral density of the signal;

[0034] The calculation formula of vehicle distance ξ(t) is:

[0035]

[0036] D min (t) = v i (t)τ+d0

[0037] In the formula, v i is the speed of the vehicle before the braking process begins; τ is the sum of the driver's reaction time and the braking effect time; d0 is the minimum distance between vehicles after the vehicle stops completely; D min(t) is the minimum vehicle spacing; D(t) is the vehicle spacing between the current vehicle and the preceding vehicle at time t;

[0038] Steering wheel turning entropy H p The calculation formula is:

[0039]

[0040] In the formula, p i is the probability of each interval.

[0041] Furthermore, the preprocessing of the EEG signal includes:

[0042] Perform fast Fourier transform on the EEG signal to obtain the power spectrum density feature in the frequency domain of the EEG signal. All EEG signals are divided into five frequency bands. The power spectrum density feature calculation formula is:

[0043]

[0044] In the formula, x n is the nth sample of the EEG signal, N is the length of the EEG signal, ω is the frequency, j is the imaginary unit, and E is the expected value.

[0045] Furthermore, the anger intensity level of each indicator is determined according to the maximum or minimum value of each indicator data and the anger level threshold range corresponding to each indicator. The anger intensity level of each indicator is divided as follows:

[0046] MeanRR >720 is normal, 695 < Mean RR ≤720 is low strength, 685<Mean RR ≤695 is medium strength, Mean RR ≤685 is high strength;

[0047] SDNN<150 is normal state, 150≤SDNN<200 is low intensity, 200≤SDNN<300 is medium intensity, and 300≤SDNN is high intensity;

[0048] pNN20>38.4 is normal, 37.4<pNN2038.4 is low intensity, 36.9<pNN20≤37.4 is medium intensity, and pNN20≤36.9 is high intensity;

[0049] LF\HF<1.52 is normal state, 1.52≤LF\HF<1.82 is low intensity, 1.82≤LF\HF<2.3 is medium intensity, 2.3≤LF\HF is high intensity;

[0050] 0<β%<0.2586 is normal state, 0.2586≤β%<0.3269 is low strength, 0.3269≤β%<0.374 is medium strength, 0.3674≤β% is high strength;

[0051] 0.2183<θ%≤1 is normal state, 0.1539<θ%≤0.2183 is low strength, 0.1216<θ%≤0.1539 is medium strength, 0<θ%≤0.1216 is high strength;

[0052] ξ(t)<0.8 is the normal state, 0.8≤ξ(t)<0.85 is low intensity, 0.85≤ξ(t)<1 is medium intensity, and 1≤ξ(t) is high intensity;

[0053] |a|<0.3 is normal state, 0.3≤|a|<0.5 is low intensity, 0.5≤|a|<1.3 is medium intensity, 1.3≤|a| is high intensity;

[0054] H p <0.5 is normal state, 0.5≤H p <0.6 is low strength, 0.6≤H p <0.7 is medium strength, 0.7≤H p For high strength;

[0055] 0<LA<0.13 is normal state, 0.13≤LA<0.15 is low intensity, 0.15≤LA<0.17 is medium intensity, and 0.17≤LA is high intensity;

[0056] 0<SGP<0.3 is the normal state, 0.3≤SGP<0.5 is low intensity, 0.5≤SGP<0.6 is medium intensity, and 0.6≤SGP is high intensity.

[0057] Furthermore, in the driving anger assessment model, the driving anger intensity is divided into four levels: normal, low intensity, medium intensity, and high intensity, and the corresponding probability ranges are [0, 0.25), [0.25, 0.5), [0.5, 0.75), and [0.75, 1], respectively.

[0058] Furthermore, the driving anger scenarios include five types of scenarios: slow driving, traffic obstruction, illegal driving, rude behavior, and bad road and weather conditions. The anger degree of each VR road traffic driving anger scenario is determined according to the scenario characteristics and is divided into five anger levels. Each type of VR traffic driving anger scenario library contains scenes of five anger levels. The driving anger evaluation module randomly selects one scene from each level of the VR road traffic driving anger scenario library for evaluation, and the selected scenes cannot be repeated. The five scenes are sequentially spliced ​​to form an evaluation driving scenario combination.

[0059] Further, the intervention measures include music therapy, bibliography therapy, mindfulness-based cognitive therapy, A, B, and C therapy, cognitive behavioral affective therapy, healthy lifestyle therapy, and narrative therapy;

[0060] The corresponding principles for anger intensity levels and intervention measures are as follows:

[0061] Normal: No intervention measures;

[0062] Low intensity: book therapy, music therapy;

[0063] Moderate intensity: book therapy, music therapy, cognitive behavioral therapy, healthy lifestyle therapy, narrative therapy;

[0064] High intensity: book therapy, music therapy, cognitive behavioral therapy, healthy lifestyle therapy, A, B, C therapy, mindfulness cognitive therapy, narrative therapy.

[0065] Furthermore, the VR driving simulator includes a VR head display device, a multi-function steering wheel, an accelerator pedal, a brake pedal, a clutch pedal, a shift handle, a seat, an EEG acquisition device, an ECG acquisition device and a driving operation data acquisition device;

[0066] The information platform is an intelligent computer with a touch screen, which synchronously displays the driver's simulated operation scene and outputs the driver's anger intensity level assessment results and corresponding intervention measures.

[0067] The second technical solution adopted by the present invention is:

[0068] A method for assessing and intervening in driver anger, comprising the following steps:

[0069] Obtain personal information entered by the driver;

[0070] Obtain VR road driving anger scenes and stitch them in sequence, and send them to the VR head display device of the VR driving simulator for display;

[0071] Obtain the driver's driving behavior data, ECG data, and EEG data on the VR driving simulator;

[0072] Extract features from the acquired driving behavior data, ECG data, and EEG data;

[0073] The driving anger intensity of the driver is evaluated according to the extracted features to obtain the driving anger intensity level;

[0074] Determining intervention measures for driver anger based on driving anger intensity levels.

[0075] The beneficial effects of the present invention are as follows: the present invention provides a driver's driving anger assessment and intervention system and method based on VR virtual driving, wherein the system uses a VR simulated driver to perform driving tests, which can more comprehensively simulate driving anger situations encountered on actual roads, and has the advantages of good stability, low interference, and safety. In addition, at the beginning of the evaluation, the VR road driving anger scene library automatically selects 5 levels of driving scenes and sequentially splices and sends them to the VR head display device. The information platform receives the driver's electrocardiogram, electroencephalogram and operation data sent by the VR simulated driver in real time, and calculates the driver's driving anger intensity level through the driving anger assessment model. The intervention module outputs corresponding intervention measures according to the driving anger intensity level. The present invention has the advantages of being comprehensive, concise, and effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the embodiments of the present invention or the drawings of related technical solutions in the prior art are introduced below. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0077] Figure 1 is a structural block diagram of a driver driving anger assessment and intervention system in an embodiment of the present invention;

[0078] Figure 2 is a schematic diagram of the connection of various modules of the driver anger assessment and intervention system in an embodiment of the present invention;

[0079] Figure 3is a flowchart of the steps of driver anger assessment and intervention in an embodiment of the present invention;

[0080] Figure 4 is a flowchart of driver driving anger intensity evaluation in an embodiment of the present invention;

[0081] Figure 5 1 is a Bayesian network structure diagram in an embodiment of the present invention. DETAILED DESCRIPTION

[0082] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0083] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0084] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0085] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0086] Terminology explanation:

[0087] HRV: Heart Rate Variability.

[0088] like Figure 1As shown, an embodiment of the present invention provides a driver anger assessment and intervention system based on VR simulated driving, comprising: a VR driving simulator and an information platform, wherein the VR simulated driving device and the information platform are connected via a network to achieve real-time communication;

[0089] The VR driving simulator serves as an evaluation platform, including a VR head display device and a driving control device, and the driving control device is configured with a data acquisition module and a data processing module.

[0090] The information platform is an intelligent computer with a touch screen, which is used for driver login, operation and generation of driving anger scenario information, calculation of the driver's driving anger level and output of intervention methods, synchronous display of the driver's simulated operation scenario and output of the driver's anger intensity level assessment results and corresponding intervention methods.

[0091] In some embodiments, the VR driving simulator includes a data acquisition module and a data analysis module, and the information platform includes a personal information module, a simulated driving module, a driving anger assessment module and an intervention module.

[0092] Specifically, the personal information module includes the driver's name, age, gender, driving experience, driving mileage, and number of violations, and is used to collect and store the driver's personal information, evaluation results, and intervention measures.

[0093] The simulation driving module is based on the historical classic driving rage scenes. Through three-dimensional modeling, an electronic information database of VR road traffic driving rage scenes is established, and the scene information is displayed on the display screen of the VR head display device;

[0094] The data acquisition module includes an electrocardiogram acquisition device, an electroencephalogram acquisition device, and multiple sensors arranged on the VR driving control device, which are used to collect the driver's electrocardiogram, electroencephalogram and driving behavior data in real time;

[0095] The data analysis module is a small intelligent computer that pre-processes the collected driver's electrocardiogram, electroencephalogram and driving behavior data and extracts characteristic indicators;

[0096] The driving anger assessment module is a pre-trained driving anger assessment model based on a Bayesian network. The characteristic indicators of the EEG, ECG and driver behavior data are input into the driving anger assessment model to obtain the driver's driving anger intensity level;

[0097] The intervention method module determines the driver's anger intervention measures according to the anger intensity level output by the driving anger assessment module and the correspondence between the preset anger intensity level and the emotion intervention method.

[0098] The connection diagram of each module of the system is as follows Figure 2As shown, the personal information module is connected to the simulation driving module, the simulation driving module is connected to the data acquisition module, the data acquisition module is connected to the data analysis module, the data analysis module is connected to the driving anger assessment module and to the personal information module, and the driving anger assessment module is connected to the intervention module and to the personal information module.

[0099] When conducting a driver's driving anger assessment, the driver's personal information, selected road scene information, driving anger level and intervention measures are all fed back to the memory of the personal information module of the information platform for backup and preservation.

[0100] The driver anger assessment and intervention system based on VR simulated driving provided by the embodiment of the present invention is constructed by adopting a VR simulated driver and an information platform, can provide a good evaluation environment for the driver's driving anger level, can immersively simulate road traffic anger scenes, and form a complete system of testing - assessment - intervention, which has the advantages of good stability, low interference, and a complete system.

[0101] Based on the above system, the embodiment of the present invention also provides a method for evaluating and intervening in driver anger based on VR simulated driving. The specific steps are as follows: Figure 3 As shown, including:

[0102] Step S1: When the driver starts the assessment, the driver fills in personal information on the information platform, including: the driver's name, age, gender, driving experience, driving mileage, and number of violations;

[0103] Step S2: After the driver inputs personal information, the driving anger module selects driving anger scenes, stitches them in sequence, and sends them to the VR head display device.

[0104] Driving anger scenarios include five types of scenarios: slow driving, traffic obstruction, illegal driving, rude behavior, and bad road and weather conditions. The anger degree of each VR road traffic driving anger scenario is determined based on the characteristics of the scenario and is divided into five anger levels. Each type of VR traffic driving anger scenario library contains scenes of five anger levels. The driving anger module randomly selects one scene from each level of the VR road traffic driving anger scenario library for evaluation, and the selected scenes cannot be repeated. The five scenes are sequentially spliced ​​to form an evaluation driving scenario combination.

[0105] The contents of the VR road traffic driving anger scene library are shown in Table 1:

[0106] Table 1

[0107]

[0108] Step S3: When the driver wears the VR head display device to simulate the operation of the driving device, the data acquisition device collects the driver's electrocardiogram information, electroencephalogram information, driver's operating behavior information, and vehicle status information in real time and sends them to the data processing module.

[0109] In actual applications, when the driver is driving in the simulator, the VR simulator collects the driver's steering wheel angle, accelerator pedal speed, vehicle distance information, vehicle longitudinal acceleration, vehicle lateral acceleration and other control information in real time. The ECG acquisition device collects the driver's ECG information in real time, and the EEG acquisition device collects the driver's EEG information in real time.

[0110] Step S4: After the data processing module pre-processes the collected data and extracts data features, it is sent to the driving anger assessment module. The specific process is as follows: Figure 4 As shown,

[0111] Specifically, when the human body is in different emotional states, the time domain parameter Mean of HRV RR , pNN 20 ,SDNN RR , the frequency domain parameters LF and HF are significantly different, so Mean is selected RR , pNN 20 ,SDNN RR , LF\HF indicators are used to build a driver's driving anger intensity assessment model.

[0112] The ECG processing unit uses wavelet transform to reduce the noise of ECG signals:

[0113]

[0114] Where WT(ε,τ) is the wavelet transform of f(t); f(t) is the ECG signal to be processed; is the wavelet function; ε is the scaling variable of the wavelet function, corresponding to the frequency; τ is the translation variable of the wavelet function, corresponding to the time; t is the time.

[0115] The ECG processing unit then obtains the RR interval sequence of the driver's ECG signal through the built-in R wave detection algorithm, and extracts the time domain and frequency domain features of the HRV signal through a characteristic sliding time window. The calculation formula is as follows:

[0116]

[0117] Mean RR represents the average value of the RR interval in the sampling period, N is the number of RR intervals in the sampling period, RR i represents the i-th RR interval.

[0118]

[0119] Among them, SDNN RR represents the standard deviation of the RR interval within the sampling period, N is the number of RR intervals within the sampling period, Represents the average value of the RR interval within the sampling period, i = 1, 2, 3, ..., N

[0120]

[0121] where pNN 20 It indicates the percentage of the number of adjacent RR intervals with a difference greater than 20ms in the sampling period to the total number of RR intervals;

[0122]

[0123] Where LF represents the power in the low frequency band, HF represents the power in the high frequency band, and PSD(f) is the power spectral density of the signal.

[0124] According to the anger level threshold range corresponding to each ECG indicator, the anger intensity level of each indicator is judged. The anger intensity levels of each ECG indicator are divided as follows:

[0125] Mean RR >720 is normal, 695 < Mean RR ≤720 is low strength, 685<Mean RR ≤695 is medium strength, Mean RR ≤685 is high strength;

[0126] SDNN RR <150 is normal, 150≤SDNN RR <200 means low intensity, 200≤SDNN RR <300 is medium strength, 300≤SDNN RR For high strength;

[0127] pNN 20 >38.4 is normal, 37.4 < pNN 20 ≤38.4 is low intensity, 36.9<pNN 20 ≤37.4 is medium strength, pNN 20 ≤36.9 is high strength;

[0128] LF\HF<1.52 is normal state, 1.52≤LF\HF<1.82 is low intensity, 1.82≤LF\HF<2.3 is medium intensity, 2.3≤LF\HF is high intensity;

[0129] Specifically, the EEG processing unit performs bandpass filtering, artifact removal and resampling on the original EEG signal to obtain preprocessed EEG data;

[0130] The preprocessed EEG data is subjected to fast Fourier transform to obtain the power spectral density feature in the frequency domain of the EEG signal. All EEG signals are divided into five frequency bands, namely δ (1-4 Hz), θ (4-8 Hz), α (8-14 Hz), and β (14-30 Hz). The power spectral density feature calculation formula is:

[0131]

[0132] where x n is the nth sample of the EEG signal, N is the length of the EEG signal, w is the frequency, j is the imaginary unit, E is the expected value, e=2.7.

[0133] According to the anger level threshold range corresponding to each EEG indicator, the anger intensity level of each indicator is judged. The anger intensity levels of each EEG indicator are divided as follows:

[0134] 0<β%<0.2586 is normal state, 0.2586≤β%<0.3269 is low strength, 0.3269≤β%<0.374 is medium strength, 0.3674≤β% is high strength;

[0135] 0.2183<θ%≤1 is normal state, 0.1539<θ%≤0.2183 is low strength, 0.1216<θ%≤0.1539 is medium strength, 0<θ%≤0.1216 is high strength;

[0136] Specifically, when the driver operates the simulated driver, the simulated driver directly outputs the driver's steering wheel angle, accelerator pedal position, longitudinal acceleration, lateral acceleration and vehicle distance, and selects steering wheel steering entropy SE, vehicle distance ξ(t), vehicle longitudinal acceleration |a|, vehicle lateral acceleration LA, and accelerator pedal speed SGP as driver behavior operation data features.

[0137] The driving behavior analysis unit calculates the vehicle spacing characteristic data ξ(t):

[0138]

[0139] D min (t) = v i (t)τ+d0

[0140] Among them, v i is the speed of the vehicle before the braking process begins; τ is the sum of the driver's reaction time and the braking effect time; d0 is the minimum distance between vehicles after the vehicle stops completely; D min(t) is the minimum vehicle distance; D(t) is the vehicle distance between the current vehicle and the preceding vehicle at time t.

[0141] The driving behavior analysis unit calculates the steering wheel turning entropy using the following formula:

[0142]

[0143] Among them, p i is the probability of each interval.

[0144] According to the anger level threshold range corresponding to each driving behavior indicator, the anger intensity level of each indicator is determined. The anger intensity level of each driving behavior indicator is divided as follows:

[0145] ξ(t)<0.8 is the normal state, 0.8≤ξ(t)<0.85 is low intensity, 0.85≤ξ(t)<1 is medium intensity, and 1≤ξ(t) is high intensity;

[0146] |a|<0.3 is normal state, 0.3≤|a|<0.5 is low intensity, 0.5≤|a|<1.3 is medium intensity, 1.3≤|a| is high intensity;

[0147] H p <0.5 is normal state, 0.5≤H p <0.6 is low strength, 0.6≤H p <0.7 is medium strength, 0.7≤H p For high strength;

[0148] 0<LA<0.13 is normal state, 0.13≤LA<0.15 is low intensity, 0.15≤LA<0.17 is medium intensity, and 0.17≤LA is high intensity;

[0149] 0<SGP<0.3 is normal state, 0.3≤SGP<0.5 is low strength, 0.5≤SGP<0.6 is medium strength, and 0.6≤SGP is high strength;

[0150] Step S5: The driving anger assessment module assesses the driving anger intensity of the driver and outputs the anger level to the intervention module.

[0151] Specifically, the driving anger assessment module is a driving anger assessment model constructed based on a Bayesian network. According to step S4,

[0152] Mean value of RR interval within the sampling period RR Index data, the percentage of the number of adjacent RR intervals with a difference greater than 20ms in the sampling period to the total number of RR intervals pNN 20 Index data, RR interval standard deviation SDNN within the sampling period RRIndex data, the ratio of low frequency power to high frequency power LF\HF index data, β wave relative power spectrum β% index data, θ wave relative power spectrum θ% index data, steering wheel turning entropy H P The Bayesian network is constructed based on the indicator data, vehicle distance ξ(t) indicator data, vehicle longitudinal acceleration |a| indicator data, vehicle lateral acceleration LA indicator data, accelerator pedal speed SGP indicator data and their corresponding anger levels, and the driver's driving anger intensity is obtained by learning and calculating the conditional probability.

[0153] Furthermore, if Figure 5 As shown in the figure, driving anger intensity R, ECG index CR, EEG index ER, and driving behavior index DR are hidden nodes, represented by H1, H2, H3, and H4 respectively. Driving anger intensity is the parent node of ECG index, EEG index, and driving behavior index.

[0154] RR interval mean RR 、The percentage of the number of adjacent RR intervals with a difference greater than 20ms to the total number of RR intervals pNN 20 , RR interval standard deviation SDNN RR , the ratio of low-frequency power to high-frequency power LF\HF, the relative power spectrum of β wave β%, the relative power spectrum of θ wave θ%, the vehicle distance ξ(t), the steering wheel turning entropy H P , vehicle longitudinal acceleration |a|, vehicle lateral acceleration LA, accelerator pedal speed SGP are observation nodes, respectively, with Y1-Y 11 express;

[0155] The state set of each node variable is shown in Table 2:

[0156] Table 2

[0157]

[0158]

[0159] Parameter learning is performed, and the conditional probability table between data input factors is obtained based on expert experience and input into the model. The probabilities of the implicit nodes H1, H2, H3, and H4 are calculated:

[0160]

[0161] Where i = [1,4]; j = [1,11], m∈[2,3]

[0162] Through the Bayesian method, the fuzzy concept of driving anger intensity was quantified and divided into four levels: normal, low intensity, medium intensity, and high intensity, with corresponding probability ranges of [0, 0.25), [0.25, 0.5), [0.5, 0.75), and [0.75, 1].

[0163] Step S6: The driving anger assessment module outputs the driving anger intensity of the driver to the intervention module, and the intervention module outputs intervention measures according to the preset corresponding rules.

[0164] Specific intervention methods include music therapy, book therapy, mindfulness-based cognitive therapy, A, B, and C therapy, cognitive behavioral emotional therapy, healthy lifestyle therapy, and narrative therapy;

[0165] The corresponding rules between anger intensity levels and intervention measures are shown in Table 3:

[0166] Table 3

[0167]

[0168] The present invention provides a driver driving anger assessment and intervention system and method based on VR simulated driving of an embodiment, which comprehensively adopts multimodal data, adopts a method combining electrocardiogram, electroencephalogram and driving behavior data, and can analyze the driver's anger level more comprehensively and accurately than a single mode of emotion recognition and evaluation technology; the driving test is carried out in a VR simulated driver, which can more comprehensively simulate the driving anger situation encountered on the actual road, and is safer and more comprehensive than the scene of the real car test, and can reduce the interference of other emotions such as tension on the test results, and can avoid the emergencies and personal injuries encountered in the real car test; output targeted intervention strategies according to the driving anger intensity level of the driver, and after the intervention cycle, the driver can be tested again to achieve long-term intervention for the driver. The present invention can be used as a driving anger detection and intervention tool for novice drivers, professional drivers and private car drivers, and can effectively reduce the "road rage" killers on the road.

[0169] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0170] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A driver anger assessment and intervention system, characterized in that: It includes a VR driving simulator and an information platform, wherein the VR driving simulator and the information platform are connected to the network for real-time communication; The VR driving simulator serves as an operating platform for the driver, and is used to display preset scene information to the driver through a VR head display device, collect driving behavior data, electrocardiogram data, and electroencephalogram data of the driver during simulated driving, and perform feature extraction on the collected driving behavior data, electrocardiogram data, and electroencephalogram data; The information platform is used to obtain the driver's driving anger intensity level according to the extracted features, and determine the driver's anger intervention measures according to the driving anger intensity level.

2. The driver anger assessment and intervention system according to claim 1, characterized in that: The VR driving simulator includes a data acquisition module and a data analysis module. The data acquisition module includes an electrocardiogram acquisition device, an electroencephalogram acquisition device, and a plurality of sensors arranged on the driving control device, which are used to collect the driver's electrocardiogram, electroencephalogram and driving behavior data in real time; The data analysis module is used to pre-process and extract features from the collected driver's electrocardiogram, electroencephalogram and driving behavior data; The information platform includes a personal information module, a simulated driving module, a driving anger assessment module and an intervention module; The personal information module is used to collect and store the driver's personal information, assessment results and intervention measures; The simulated driving module is used to establish an electronic information database of VR road traffic driving anger scenes through three-dimensional modeling based on historical classic driving anger scenes, and display the scene information through a VR head display device; The driving anger assessment module is used to input the extracted features into a pre-trained driving anger assessment model based on a Bayesian network to obtain the driving anger intensity level of the driver; The intervention module is used to determine intervention measures for driver anger according to the anger intensity level output by the driving anger assessment module and the corresponding relationship between the anger intensity level and the emotion intervention method set in advance.

3. The driver anger assessment and intervention system according to claim 2, characterized in that: The data analysis module includes an electrocardiogram analysis unit, an electroencephalogram analysis unit, and a driving behavior analysis unit; The ECG analysis unit is used to pre-process the ECG signal and extract the RR interval mean value in the ECG signal. RR , the percentage of the number of adjacent RR intervals whose difference is greater than the preset value to the total number of RR intervals pNN20, the standard deviation of RR intervals SDNN RR , the ratio of low frequency power to high frequency power LF\HF data characteristics; The EEG analysis unit is used to pre-process the EEG signal and extract data features of the relative power spectrum β% of the β wave and the relative power spectrum θ% of the θ wave in the EEG signal; The driving behavior analysis unit is used to pre-process the driving operation data and extract the vehicle distance ξ(t), steering wheel turning entropy H p , vehicle longitudinal acceleration |a|, vehicle lateral acceleration LA, and accelerator pedal speed SGP data characteristics.

4. The driver anger assessment and intervention system according to claim 3, characterized in that: Mean value of the RR interval RR The calculation formula is: Where N is the number of RR intervals in the sampling period, RR i represents the i-th RR interval; Percent pNN 20 The calculation formula is: In the formula, NN 20 The number of times the difference between two adjacent RR intervals in the sampling period is greater than the preset value; The formula for calculating the power in the low frequency band is: The formula for calculating the power in the high frequency band is: Where PSD(f) is the power spectral density of the signal; The calculation formula of vehicle distance ξ(t) is: D min (t)=v i (t)τ+d0 In the formula, v i is the speed of the vehicle before the braking process begins; τ is the sum of the driver's reaction time and the braking effect time; d0 is the minimum distance between vehicles after the vehicle stops completely; D min (t) is the minimum vehicle spacing; D(t) is the vehicle spacing between the current vehicle and the preceding vehicle at time t; Steering wheel turning entropy H p The calculation formula is: In the formula, p i is the probability of each interval.

5. The driver anger assessment and intervention system according to claim 3, characterized in that: The preprocessing of the EEG signal includes: Perform fast Fourier transform on the EEG signal to obtain the power spectrum density feature in the frequency domain of the EEG signal. All EEG signals are divided into five frequency bands. The power spectrum density feature calculation formula is: In the formula, x n is the nth sample of the EEG signal, N is the length of the EEG signal, ω is the frequency, j is the imaginary unit, and E is the expected value.

6. The driver anger assessment and intervention system according to claim 3, characterized in that: According to the maximum or minimum value of each indicator data and the anger level threshold range corresponding to each indicator, the anger intensity level of each indicator is judged. The anger intensity level of each indicator is divided as follows: Mean RR >720 is normal, 695 <Mean RR ≤720 is low strength, 685 <Mean RR ≤695 is medium strength, Mean RR ≤685 is high strength; SDNN<150 is normal, 150≤SDNN<200 is low intensity, 200≤SDNN<300 is medium intensity, and 300≤SDNN is high intensity; pNN 20 >38.4 is normal, 37.4 < pNN 20 38.4 is low intensity, 36.9<pNN 20 ≤37.4 is medium strength, pNN 20 ≤36.9 is high strength; LF / HF < 1.52 represents the normal state, 1.52 ≤ LF / HF < 1.82 represents low intensity, 1.82 ≤ LF / HF < 2.3 represents medium intensity, 2.3 ≤ LF / HF represents high intensity; 0 < β% < 0.2586 represents the normal state, 0.2586 ≤ β% < 0.3269 represents low intensity, 0.3269 ≤ β% < 0.374 represents medium intensity, 0.3674 ≤ β% represents high intensity; 0.2183 < θ% ≤ 1 represents the normal state, 0.1539 < θ% ≤ 0.2183 represents low intensity, 0.1216 < θ% ≤ 0.1539 represents medium intensity, 0 < θ% ≤ 0.1216 represents high intensity; ξ(t) < 0.8 represents the normal state, 0.8 ≤ ξ(t) < 0.85 represents low intensity, 0.85 ≤ ξ(t) < 1 represents medium intensity, 1 ≤ ξ(t) represents high intensity; |a| < 0.3 represents the normal state, 0.3 ≤ |a| < 0.5 represents low intensity, 0.5 ≤ |a| < 1.3 represents medium intensity, 1.3 ≤ |a| represents high intensity; H p <0.5 is normal state, 0.5≤H p <0.6 is low strength, 0.6≤H p <0.7 is medium strength, 0.7≤H p For high strength; 0 < LA < 0.13 represents the normal state, 0.13 ≤ LA < 0.15 represents low intensity, 0.15 ≤ LA < 0.17 represents medium intensity, 0.17 ≤ LA represents high intensity; 0 < SGP < 0.3 represents the normal state, 0.3 ≤ SGP < 0.5 represents low intensity, 0.5 ≤ SGP < 0.6 represents medium intensity, 0.6 ≤ SGP represents high intensity.

7. The driver anger assessment and intervention system according to claim 2, characterized in that: In the driving anger assessment model, the driving anger intensity is divided into 4 levels: normal, low intensity, medium intensity, and high intensity, and the corresponding probability ranges are [0, 0.25), [0.25, 0.5), [0.5, 0.75), [0.75, 1] respectively.

8. The driver anger assessment and intervention system according to claim 2, characterized in that: Driving anger scenarios include: slow driving, traffic obstruction, illegal driving, rude behavior, and poor road and weather conditions. There are 5 types of scenarios. The anger level of each VR road traffic driving anger scenario is determined according to the scenario characteristics and is divided into 5 anger levels. Each VR traffic driving anger scenario library contains scenarios of 5 anger levels; the driving anger assessment module randomly selects 1 scenario from each level of the VR road traffic driving anger scenario library for evaluation, and the selected scenarios cannot be repeated. The 5 scenarios are sequentially spliced to form an evaluation driving scenario combination.

9. The driver anger assessment and intervention system according to claim 1, characterized in that: The intervention measures include music therapy, book therapy, mindfulness-based cognitive therapy, A, B, C therapy, cognitive-behavioral-emotional therapy, healthy lifestyle therapy, and narrative therapy; The corresponding principles for the anger intensity level and intervention measures are as follows: Normal: No intervention measures; Low intensity: Book therapy, music therapy; Medium intensity: Book therapy, music therapy, cognitive-behavioral therapy, healthy lifestyle therapy, narrative therapy; High intensity: Book therapy, music therapy, cognitive-behavioral therapy, healthy lifestyle therapy, A, B, C therapy, mindfulness-based cognitive therapy, narrative therapy.

10. A method for assessing and intervening in driver anger, characterized in that: It includes the following steps: Obtain the personal information input by the driver; Obtain VR road driving anger scenarios and splice them sequentially, and send them to the VR headset device of the VR driving simulator for display; Obtain the driving behavior data, electrocardiogram data, and electroencephalogram data of the driver on the VR driving simulator; Extract features from the acquired driving behavior data, ECG data, and EEG data; The driving anger intensity of the driver is evaluated according to the extracted features to obtain the driving anger intensity level; Determining intervention measures for driver anger based on driving anger intensity levels.

Images