Vehicle control method and device, electronic equipment and storage medium

Abstract

The invention provides a vehicle control method and device, electronic equipment and a storage medium, and the vehicle control method comprises the steps: obtaining real-time physiological information of a driver in a vehicle driving process if a starting signal of an intelligent driving system of a vehicle is detected; based on a preset driver trust degree evaluation model, according to the real-time physiological information, evaluating the real-time trust degree of the driver to the intelligent driving system; and based on a preset adjustment rule, adjusting the intervention frequency and intervention mode of the intelligent driving system according to the real-time credibility. On the basis, the accuracy and adaptability of trust evaluation are improved, and then the driving safety and reliability are effectively improved.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and particularly to a vehicle control method, device, electronic device, and storage medium. Background Art

[0002] With the continuous development of intelligent driving technology, the proportion and duration of the driver's participation in driving tasks are continuously decreasing. However, the current intelligent driving technology has not yet reached a fully mature stage, and driving safety cannot be fully guaranteed in many traffic environments. Therefore, vehicles equipped with intelligent driving functions still require the driver to be able to take over the vehicle control in a timely manner at critical moments.

[0003] The driver's trust in the intelligent driving system plays an important role in the driver's willingness and timing to take over the vehicle. If the trust is too high, the driver is prone to over-rely on the intelligent driving system and cannot intervene and take over the vehicle quickly at critical moments, which may pose a threat to traffic safety. If the trust is too low, the driver may over-monitor the intelligent driving system, frequently generate takeover intentions, increase the driver's cognitive load and fatigue, and reduce the overall driving experience.

[0004] However, current intelligent driving systems mostly monitor the driver's basic state based on the driver's facial features and the physical state of the vehicle. However, for identifying the driver's mental state, especially the degree of trust of the driver in the intelligent driving system, there has been no in-depth exploration, which greatly restricts the overall performance and safety of the intelligent driving system. Summary of the Invention

[0005] In view of the above, it is necessary to provide a vehicle control method, device, electronic device, and storage medium to solve the technical problem that current intelligent driving systems mostly monitor the driver's basic state based on the driver's facial features and the physical state of the vehicle, but for identifying the driver's mental state, especially the degree of trust of the driver in the intelligent driving system, there has been no in-depth exploration, which greatly restricts the overall performance and safety of the intelligent driving system.

[0006] In a first aspect, this application provides a vehicle control method, the method including: if an activation signal of the intelligent driving system of the vehicle is detected, obtaining the real-time physiological information of the driver during the process of driving the vehicle; based on a preset trust degree evaluation model of the driver, evaluating the real-time trust degree of the driver in the intelligent driving system according to the real-time physiological information; based on a preset adjustment rule, adjusting the intervention frequency and intervention method of the intelligent driving system according to the real-time trust degree.

[0007] In the vehicle control method of the present application, first, when the opening signal of the intelligent driving system of the vehicle is detected, the real-time physiological information of the driver during the driving process is obtained, which can accurately reflect the psychological state of the driver. Further, the real-time physiological information is input into the trustworthiness evaluation model exclusive to the driver to obtain the real-time trustworthiness of the driver in the intelligent driving system. Considering the individual differences in physiological performance and trust performance among different drivers, it avoids the deviation that may be brought by a general trustworthiness evaluation model, can more accurately capture the relationship between the physiological characteristics and trust state of the individual driver, improves the accuracy and adaptability of trust evaluation, and meets the personalized evaluation requirements. Finally, the intervention frequency and intervention method of the intelligent driving system are adjusted in real time according to the real-time trustworthiness. This real-time dynamic adjustment method enables the intelligent driving system to better adapt to the psychological state of the driver, thereby effectively improving driving safety and reliability, reducing the probability of traffic accidents, and thus enhancing the driver's satisfaction with the intelligent driving system, which helps to promote the popularization and application of intelligent driving technology.

[0008] In some embodiments of the present application, before the method evaluates the real-time trustworthiness of the driver in the intelligent driving system according to the real-time physiological information based on the preset trustworthiness evaluation model of the driver, it further includes: obtaining the initial physiological information of the driver when driving the vehicle in a variety of preset driving environments, where the initial physiological information is marked with the initial trustworthiness of the driver in the intelligent driving system; based on the initial physiological information and the initial trustworthiness corresponding to the initial physiological information, and using the cross-entropy loss function for training and optimization, the pre-trained trustworthiness evaluation model of the driver is obtained.

[0009] In some embodiments of the present application, after the method obtains the pre-trained trustworthiness evaluation model of the driver, it further includes: obtaining the actual driving environment of the driver during the driving process and the behavior information in the driving environment; based on the pre-trained trustworthiness evaluation model of the driver, according to the real-time physiological information, the actual driving environment, and the behavior information, model adaptive learning is performed to obtain the trustworthiness evaluation model of the driver after adaptive learning.

[0010] In some embodiments of the present application, the model adaptive learning based on the pre-trained trustworthiness evaluation model of the driver according to the real-time physiological information, the actual driving environment, and the behavior information includes: using the pre-trained trustworthiness evaluation model of the driver as the source domain model, and using the real-time physiological information as the target domain data, and adjusting the parameters of the source domain model based on the source domain model and the target domain data.

[0011] In some embodiments of the present application, for the model adaptive learning of the pre-trained driver trust evaluation model based on the real-time physiological information, the actual driving environment, and the behavior information, it further includes: if the actual driving environment is a new driving environment, using the behavior information as the feedback input of the pre-trained driver trust evaluation model, and combining the real-time physiological information to adjust the parameters of the pre-trained driver trust evaluation model.

[0012] In some embodiments of the present application, for the model adaptive learning of the pre-trained driver trust evaluation model based on the real-time physiological information, the actual driving environment, and the behavior information, it further includes: dividing the real-time physiological information based on a preset time period to obtain a plurality of batch data; based on the time period, adjusting the parameters of the pre-trained driver trust evaluation model sequentially according to each batch data in the plurality of batch data.

[0013] In some embodiments of the present application, for the model adaptive learning of the pre-trained driver trust evaluation model based on the real-time physiological information, the actual driving environment, and the behavior information, it further includes: if the acquisition quantity of the real-time physiological information reaches a preset value, adjusting the parameters of the pre-trained driver trust evaluation model.

[0014] In a second aspect, the present application further provides a vehicle control device applied to a vehicle. The device includes: an acquisition module, configured to acquire the real-time physiological information of the driver during the driving of the vehicle if an activation signal of the intelligent driving system of the vehicle is detected; an evaluation module, configured to evaluate the real-time trust of the driver in the intelligent driving system based on the preset driver trust evaluation model according to the real-time physiological information; and an adjustment module, configured to adjust the intervention frequency and intervention method of the intelligent driving system based on a preset adjustment rule according to the real-time trust.

[0015] In a third aspect, the present application further provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the steps of the vehicle control method described in the above embodiments are implemented.

[0016] In a fourth aspect, the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the vehicle control method described in the above embodiments is implemented.

[0017] Understandably, the vehicle control device in the second aspect, the electronic device in the third aspect, and those in the fourth aspect provided above all correspond to the vehicle control method in the first aspect. Therefore, the beneficial effects they can achieve can refer to the beneficial effects in the corresponding vehicle control method provided above, and will not be elaborated here. Description of the Drawings

[0018] Figure 1 It is a schematic flowchart of a vehicle control method provided by an embodiment of the present application.

[0019] Figure 2 It is a schematic flowchart of the acquisition steps of a trust evaluation model provided by another embodiment of the present application.

[0020] Figure 3 It is a schematic diagram of the functional modules of a vehicle control device provided by an embodiment of the present application.

[0021] Figure 4 It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present application.

[0022] Description of Element Symbols Electronic device 10 Memory 11 Processor 12 Vehicle control device 100 Acquisition module 110 Evaluation module 120 Adjustment module 130 The following specific embodiments will further illustrate the present application in conjunction with the above drawings. Specific Embodiments

[0023] In order to make the technical problems, technical solutions, and beneficial effects solved by the present application clearer, the present application will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0024] In order to be able to understand the embodiments of the present invention more clearly, the present invention will be described in detail below in conjunction with the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0025] Research shows that the degree of trust of a driver in an intelligent driving system is closely related to the driver's brain electrical activity. By analyzing specific frequency bands (such as alpha waves, beta waves, etc.) in the brain electrical signals, the degree of trust of the driver in the intelligent driving system can be effectively evaluated.

[0026] However, currently, a general evaluation model is mostly used to evaluate the trust level of drivers, failing to fully consider the individual differences in physiological signals and trust performance among different drivers. As a result, there are significant differences in the accuracy and adaptability of trust level evaluation among different drivers, which cannot meet the personalized needs, affect the adaptability of the intelligent driving system, and limit the actual application effect of intelligent driving technology.

[0027] Based on this, the embodiments of the present application provide a vehicle control method, including: if an opening signal of the intelligent driving system of the vehicle is detected, obtaining the real-time physiological information of the driver during the driving process; based on a preset trust level evaluation model of the driver, evaluating the real-time trust level of the driver in the intelligent driving system according to the real-time physiological information; based on a preset adjustment rule, adjusting the intervention frequency and intervention method of the intelligent driving system according to the real-time trust level.

[0028] In the vehicle control method of the present application, first, when an opening signal of the intelligent driving system of the vehicle is detected, the real-time physiological information of the driver during the driving process is obtained, which can accurately reflect the psychological state of the driver. Further, the real-time physiological information is input into the trust level evaluation model exclusive to the driver to obtain the real-time trust level of the driver in the intelligent driving system, considering the individual differences in physiological performance and trust performance among different drivers, avoiding the deviation that may be brought by the general trust level evaluation model, being able to more accurately capture the relationship between the physiological characteristics and trust state of the individual driver, improving the accuracy and adaptability of trust evaluation, and meeting the personalized evaluation needs. Finally, the intervention frequency and intervention method of the intelligent driving system are adjusted in real time according to the real-time trust level. This real-time dynamic adjustment method enables the intelligent driving system to better adapt to the psychological state of the driver, thereby effectively improving driving safety and reliability, reducing the probability of traffic accidents, and thus enhancing the satisfaction of the driver with the intelligent driving system, contributing to the popularization and application of intelligent driving technology.

[0029] Please refer to Figure 1 , which is a schematic flowchart of the vehicle control method provided by an embodiment of the present application.

[0030] The vehicle control method of the embodiments of the present application can be applied to one or more Figure 4In the electronic device 10 shown, the electronic device 10 is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0031] Among them, the electronic device 10 can be an in-vehicle device of a vehicle, such as a body control module (BCM), a vehicle control unit (VCU), etc.

[0032] Specifically, the vehicle control method of the embodiments of the present application specifically includes the following steps. According to different requirements, the order of some steps in this flowchart can be changed, and some steps can be omitted.

[0033] S10: If an enabling signal of the intelligent driving system of the vehicle is detected, obtain the real-time physiological information of the driver during the driving process.

[0034] In some embodiments of the present application, the vehicle may include an electroencephalogram sensing device. The electronic device 10 is communicatively connected to the electroencephalogram sensing device, and the initial electroencephalogram signal can be collected through the electroencephalogram sensing device. After the electronic device 10 detects the enabling signal of the intelligent driving system, the electroencephalogram sensing device will immediately start collecting the real-time physiological information of the driver during the driving process. Among them, the real-time physiological information includes, but is not limited to, the real-time electroencephalogram signal of the driver, etc.

[0035] Further, the acquisition rate of the electroencephalogram sensing device can be set to 500 Hz to ensure the high precision and real-time performance of the real-time electroencephalogram signal. The real-time electroencephalogram signal needs to cover the main frequency bands such as alpha waves (8 - 13 Hz), beta waves (13 - 30 Hz), etc., to comprehensively reflect the electroencephalogram activity of the driver.

[0036] S20: Based on a pre-set trust degree evaluation model of the driver, evaluate the real-time trust degree of the driver in the intelligent driving system according to the real-time physiological information.

[0037] In some embodiments of the present application, before evaluating the driver's real-time trust in the intelligent driving system, the electronic device can perform preprocessing on the real-time physiological information. For example, a band-pass filter can be used to set the frequency range between 1 - 50 Hz and perform a filtering operation on the real-time physiological information to remove noise and power frequency interference in the real-time physiological information. Then, a fourth-order filter is used to perform a filtering operation on the real-time physiological information again to ensure the smoothness and real-time nature of the real-time physiological information, which can provide more reliable data support for subsequent trust evaluation, and further make the driver trust evaluation more accurate.

[0038] Furthermore, the preprocessed real-time physiological information is used as an input feature vector and input into a pre-constructed and trained trust evaluation model dedicated to the driver. Through the internal operation and analysis of the trust evaluation model, the output result is normalized, that is, a specific mathematical transformation algorithm is used to map the original output value to the standardized interval of [0, 1], so as to ensure the standardization and consistency of the output result. Then, the normalized output result is directly used as the real-time trust, which helps the subsequent quantitative analysis of the real-time trust and can provide a more reliable decision-making basis for the intelligent driving system.

[0039] In some embodiments of the present application, the electronic device 10 can perform online analysis on the collected real-time physiological information according to a preset calculation period to ensure the immediacy and accuracy of trust monitoring, and further ensure that the driver's trust state can be accurately captured and responded to. Among them, the calculation period of the real-time trust can be set to once per second to ensure the real-time update of the trust state, and further ensure that the vehicle can quickly respond to changes in the driver's driving state.

[0040] Furthermore, when the electronic device 10 performs online analysis, the data caching duration can be 5 seconds, which can not only provide a sufficient number of data points for the analysis process to reduce interference caused by data fluctuations, but also not cause response delays, thus ensuring the dynamic tracking and effective response of the intelligent driving system to the driver's trust state and providing a safer and more reliable intelligent driving experience for the driver.

[0041] It should be noted that the specific steps for obtaining the trust evaluation model are described in detail later Figure 2 and will not be repeated here to avoid redundancy.

[0042] S30: Based on a preset adjustment rule, adjust the intervention frequency and intervention method of the intelligent driving system according to the real-time trust.

[0043] In some embodiments of the present application, the electronic device 10 divides the real-time trust level into a high-trust state and a low-trust state. The high-trust threshold can be 0.8. When the real-time trust level is greater than or equal to 0.8, it indicates that the driver is in a high-trust state. The low-trust threshold can be 0.3. When the real-time trust level is less than or equal to 0.3, it indicates that the driver is in a low-trust state.

[0044] Among them, the intervention frequency and the intervention method are two key parameters that the intelligent driving system dynamically adjusts according to the driver's real-time trust level. The intervention frequency refers to the number of times the intelligent driving system intervenes and controls the vehicle driving behavior within a certain time range. The intervention method is the specific form in which the intelligent driving system intervenes and controls the vehicle.

[0045] Furthermore, the adjustment rules include but are not limited to the following rules: (1) Intervention frequency adjustment: When the real-time trust level is in the low-trust state, the intervention frequency is set to 5 times per minute to provide more assisted driving support to improve driving safety. When the real-time trust level is in the high-trust state, the intervention frequency is reduced to 1 time per minute to give the driver more autonomous control and enhance the driving experience. (2) Intervention method adjustment: When the real-time trust level is in the low-trust state, the active intervention method is adopted, such as the active activation of functions such as automatic braking and lane keeping, to enhance the intervention in the driver's operation. When the real-time trust level is in the high-trust state, the passive intervention method is adopted, such as providing prompts or suggestions when necessary, to reduce the intervention in the driver's operation.

[0046] Among them, the delay time for the intelligent driving system to respond to the above adjustment operations shall not exceed 200 milliseconds to ensure the timeliness of the adjustment and driving safety.

[0047] The adjustment rules in the above embodiments reasonably balance driving safety and driving experience, improve the overall performance of the intelligent driving system, enable the intelligent driving system to perform well among different user groups, and broaden the application scope of the intelligent driving system.

[0048] In the vehicle control method of the present application, first, when the opening signal of the intelligent driving system of the vehicle is detected, the real-time physiological information of the driver during the driving process is obtained, which can accurately reflect the psychological state of the driver. Further, the real-time physiological information is input into the trustworthiness evaluation model exclusive to the driver to obtain the real-time trustworthiness of the driver in the intelligent driving system. Considering the individual differences in physiological performance and trust performance among different drivers, it avoids the deviation that may be brought by a general trustworthiness evaluation model, can more accurately capture the relationship between the physiological characteristics and trust state of the individual driver, improves the accuracy and adaptability of trust evaluation, and meets the personalized evaluation needs. Finally, according to the real-time trustworthiness, the intervention frequency and intervention method of the intelligent driving system are adjusted in real time. This real-time dynamic adjustment method enables the intelligent driving system to better adapt to the psychological state of the driver, thereby effectively improving driving safety and reliability, reducing the probability of traffic accidents, and thus enhancing the driver's satisfaction with the intelligent driving system, which helps to promote the popularization and application of intelligent driving technology.

[0049] In some embodiments of the present application, after each adjustment of the intelligent driving system, the adjustment effect is evaluated through the changes in the driver's behavior information and real-time trustworthiness, and is fed back to the adaptive learning mechanism to continuously optimize the trustworthiness evaluation model, thereby improving the accuracy of trustworthiness evaluation.

[0050] Please refer to Figure 2 , which is a schematic flowchart of the acquisition steps of the trustworthiness evaluation model provided by an embodiment of the present application.

[0051] Specifically, the acquisition steps of the trustworthiness evaluation model in the embodiments of the present application specifically include the following steps. According to different requirements, the order of some steps in this flowchart can be changed, and some steps can be omitted.

[0052] S21: Obtain the initial physiological information of the driver when driving the vehicle in a variety of preset driving environments.

[0053] Among them, the initial physiological information is marked with the initial trustworthiness of the driver in the intelligent driving system. In this embodiment, the initial trustworthiness of the driver in the intelligent driving system can be recorded by the multi-dimensional subjective scale method. Specifically, the subjective evaluation of the driver's trustworthiness in the intelligent driving system is collected through a standardized multi-dimensional trust scale, and then the initial trustworthiness is determined based on the subjective evaluation and the preset scoring rules.

[0054] In some embodiments of the present application, the initial physiological information includes but is not limited to the initial electroencephalogram signals generated by the driver when driving the vehicle in different preset driving environments. For example, a variety of preset driving environments include but are not limited to urban driving, highway driving, night driving and other environments, which can ensure the comprehensiveness and diversity of the collected data.

[0055] Specifically, the electronic device 10 pre-controls the electroencephalogram sensing device of the vehicle to collect the initial electroencephalogram signals generated by the driver in a variety of preset driving environments. Among them, the acquisition rate of the electroencephalogram sensing device can be set to 500Hz, which can ensure the high precision and real-time performance of the initial electroencephalogram signals. The initial electroencephalogram signals need to cover the main frequency bands such as alpha waves (8-13Hz) and beta waves (13-30Hz), and can comprehensively reflect the electroencephalogram activities of the driver.

[0056] It should be noted that in the data acquisition stage, at least 30 minutes of initial physiological information and the corresponding initial trust levels need to be collected to ensure the stability and representativeness of the data.

[0057] S22: Based on the initial physiological information and the initial trust levels corresponding to the initial physiological information, and using the cross-entropy loss function for training optimization, a pre-trained trust level evaluation model of the driver is obtained.

[0058] In some embodiments of the present application, the specific steps of obtaining a pre-trained trust level evaluation model of the driver based on the initial physiological information and the initial trust levels corresponding to the initial physiological information, and using the cross-entropy loss function for training optimization include: extracting feature data from the initial physiological information based on a preset feature extraction algorithm; determining key features in the feature data based on a preset feature selection algorithm; training a preset deep neural network based on the key features and the initial trust levels corresponding to the key features, and using the cross-entropy loss function for training optimization to obtain a pre-trained trust level evaluation model of the driver.

[0059] Specifically, through the time-frequency joint analysis algorithm, the initial electroencephalogram signals are deeply analyzed in both the time domain and the frequency domain, and time domain indexes such as average value, standard deviation, kurtosis, and skewness are respectively extracted. And frequency band features such as alpha waves, beta waves, and theta waves are obtained through short-time Fourier transform and wavelet transform to extract feature data that can reflect the trust state. Further, in combination with time-frequency analysis methods such as continuous wavelet transform, functional connection indexes and network features between brain regions are extracted to capture deeper feature data. Further, the feature data can be preprocessed through preprocessing operations such as artifact removal, denoising, and baseline correction to ensure the reliability and stability of the feature data. Further, the extracted feature data is subjected to redundancy removal processing by using the principal component analysis algorithm, regression analysis algorithm, random forest algorithm, etc., and finally key features that can effectively distinguish different trust states are obtained. Finally, based on the key features and the initial trust levels corresponding to the key features, the deep neural network is trained, and the cross-entropy loss function is used for training optimization to obtain a pre-trained trust level evaluation model of the driver, which can effectively distinguish the physiological characteristics of the driver in different trust states, thereby improving the recognition accuracy and adaptability of the intelligent driving system to individual differences.

[0060] In some embodiments of the present application, during the formal training phase of the deep neural network, hyperparameters can also be continuously adjusted by means of grid search, Bayesian optimization, etc. to ensure that the trustworthiness evaluation model can reach an ideal level in terms of stability and generalization. After the training is completed, the electronic device 10 saves a set of corresponding network weights for each driver, forming a trustworthiness evaluation model exclusive to the driver.

[0061] In other embodiments, in the case of a large data scale, a general model can be pre-constructed as the initial basis for model training, and then the general model is fine-tuned according to the personalized needs of different drivers, so as to obtain a trustworthiness evaluation model exclusive to the driver, which can retain useful information on general features, focus on personalized features, and improve the training efficiency and effect.

[0062] Among them, the evaluation of model performance can refer to indicators such as confusion matrix, precision, recall, harmonic mean of precision and recall (F1-score), and area under the receiver operating characteristic curve (Area Under the ROC Curve, AUC), etc., to comprehensively verify the recognition ability under different trust states, and then make targeted optimization adjustments, such as adjusting the model structure, parameters or training data, etc., to improve the performance of the trustworthiness evaluation model in practical applications.

[0063] In some embodiments of the present application, the setting of the training parameters of the deep neural network includes but is not limited to: learning rate: set to 0.001 to ensure the stable convergence of the model; batch size: adopt a batch size of 32 to optimize the efficiency of the training process; number of iterations: during the training process, the number of iterations is not less than 100 times to ensure the sufficient training of the model. Model verification: use the cross-entropy loss function to verify the generalization ability of the model, and the number of verification folds is set to 5 folds.

[0064] In some embodiments of the present application, the dimension of the extracted key features needs to be controlled within 100 dimensions to balance the complexity and computational efficiency of the trustworthiness evaluation model.

[0065] In some embodiments of the present application, the network structure of the deep neural network includes at least three hidden layers, and the number of neurons in each layer is not less than 128.

[0066] S23: Obtain the actual driving environment of the driver during the process of driving the vehicle and the behavior information in the driving environment.

[0067] Among them, the actual driving environment includes but is not limited to the state parameters of the vehicle, and the behavior information includes but is not limited to acceleration, braking, steering, etc.

[0068] S24: Based on a pre-trained driver trustworthiness assessment model, perform model adaptive learning according to real-time physiological information, the actual driving environment, and behavioral information to obtain an adaptive learning-based driver trustworthiness assessment model.

[0069] In some embodiments of the present application, the specific steps of performing model adaptive learning based on a pre-trained driver trustworthiness assessment model according to real-time physiological information, the actual driving environment, and behavioral information include: using the pre-trained driver trustworthiness assessment model as the source domain model and the real-time physiological information as the target domain data, and adjusting the parameters of the source domain model based on the source domain model and the target domain data.

[0070] Specifically, obtain the difference features between the source domain data and the target domain data of the source domain model, and update the parameters of the fully connected classification layer of the source domain model based on the difference features to adapt to the data distribution of the target domain data. Among them, the data distribution difference between the source domain data and the target domain data should not exceed 10%, which helps to ensure that there is sufficient similarity between the source domain data and the target domain data during the update process of the source domain model, thus guaranteeing the effect of adaptive learning, avoiding the degradation of model performance or the failure of optimization caused by excessive data distribution differences, and ensuring the accuracy and reliability of model evaluation. Among them, the adjustment range of the parameters of the source domain model does not exceed ±5% to prevent the source domain model from overfitting the new data.

[0071] In the above embodiments, by obtaining the difference features between the source domain data and the target domain data, the trustworthiness assessment model can better adapt to the data distribution of the target domain data, thereby improving the adaptability of the trustworthiness assessment model to different driving environments and enabling the trustworthiness assessment model to more accurately evaluate the driver's trustworthiness in various situations. Further, updating the parameters of the fully connected classification layer of the source domain model based on the difference features allows the trustworthiness assessment model to quickly adjust and optimize based on a small amount of real-time physiological information on the basis of the initial physiological information, not only reducing the dependence on a large amount of real-time data, but also enabling knowledge transfer between different driving environments, enhancing the generalization ability of the trustworthiness assessment model, so that the trustworthiness assessment model can make a more accurate assessment more quickly when facing a new driving environment without having to retrain the entire model. Further, only updating the parameters of the fully connected classification layer of the trustworthiness assessment model instead of all the parameters of the entire trustworthiness assessment model greatly reduces the computational amount and resource consumption, enabling the trustworthiness assessment model to be updated and optimized more quickly, improving the update efficiency of the trustworthiness assessment model, and being able to more timely reflect the changes in driver trustworthiness, providing a more real-time feedback and adjustment basis for the intelligent driving system.

[0072] In some embodiments of the present application, based on a pre-trained trustworthiness evaluation model of a driver, according to real-time physiological information, the actual driving environment, and behavior information, the specific steps of model adaptive learning further include: If the actual driving environment is a new driving environment, use the behavior information as the feedback input of the pre-trained trustworthiness evaluation model of the driver, and combine the real-time physiological information to adjust the parameters of the pre-trained trustworthiness evaluation model of the driver.

[0073] Specifically, for real-time physiological information lacking a large amount of labeled data, the electronic device 10 can combine unsupervised and semi-supervised methods, use the driver's behavior information as the feedback input of the trustworthiness evaluation model, and achieve the rapid adaptation of the trustworthiness evaluation model. Among them, the training weight allocation ratio of the trustworthiness evaluation model can be 70% for behavior information and 30% for real-time physiological information.

[0074] Furthermore, the present application does not need to perform complex retraining on the entire trustworthiness evaluation model. Only local updates and optimizations are performed on the trustworthiness evaluation model according to the feedback behavior information, which not only reduces the cost of model update, simplifies the update process, improves the efficiency of model update, and enables the model to adapt to new driving situations more quickly.

[0075] In the above embodiments, using the behavior information as the feedback input enables the trustworthiness evaluation model to adjust the evaluation of trustworthiness according to the driver's behavior performance in actual driving. For example, if the driver shows behaviors inconsistent with the expectations of the trustworthiness evaluation model in a certain driving environment, the trustworthiness evaluation model can immediately adjust according to this behavior information to more accurately reflect the driver's true trust state. When the driver's physiological state shows abnormalities, the trustworthiness evaluation model can timely discover these changes through the feedback behavior information and re-evaluate the driver's trustworthiness, thereby enhancing the adaptability of the trustworthiness evaluation model to the dynamic driving environment and ensuring that the trustworthiness evaluation model can accurately evaluate the driver's trustworthiness under different conditions.

[0076] In some embodiments of the present application, based on a pre-trained trustworthiness evaluation model of a driver, according to real-time physiological information, the actual driving environment, and behavior information, the specific steps of model adaptive learning further include: Divide the real-time physiological information based on a preset time period to obtain multiple batch data; Based on the time period, according to each batch data in the multiple batch data, sequentially adjust the parameters of the pre-trained trustworthiness evaluation model of the driver.

[0077] In the above embodiments, the network layer and some weights of the trust degree evaluation model are periodically trained in small batches, and the trust degree evaluation model is optimized at any time, which can effectively reduce the computational complexity of each update, make more efficient use of computing resources, speed up the model update speed, and continuously optimize the performance of the trust degree evaluation model, so that the trust degree evaluation model maintains high sensitivity and accuracy in identifying the trust state of the driver.

[0078] In some embodiments of the present application, based on the pre-trained trust degree evaluation model of the driver, according to the real-time physiological information, actual driving environment, and behavior information, the specific steps of model adaptive learning further include: if the acquisition quantity of the real-time physiological information all reaches a preset value (for example, 1000 pieces), adjust the parameters of the pre-trained trust degree evaluation model of the driver.

[0079] In the above embodiments, taking the acquisition quantity of the real-time physiological information reaching the preset value as a trigger condition to adjust the model parameters can avoid insufficient model update caused by too small data volume, ensure that the model is updated when the data volume reaches a certain level, and improve data processing efficiency.

[0080] In other embodiments, if the difference between the real-time trust degree and the subjective trust degree is less than or equal to the first preset threshold, and / or, when the prediction confidence of the trust degree evaluation model is greater than or equal to the second preset threshold, it indicates that the trust degree evaluation model performs stably and has a relatively high prediction accuracy within a certain period, then the parameter update frequency of the trust degree evaluation model can be reduced to reduce resource consumption and computational burden.

[0081] In the acquisition steps of the trust degree evaluation model of the present application, first, through the initial physiological information of the driver in a variety of preset driving environments and the corresponding initial trust degree of the driver in the intelligent driving system, and using the cross-entropy loss function for training and optimization, a pre-trained trust degree evaluation model of the driver is obtained. Considering the individual differences of different drivers in physiological information and trust performance, it avoids the deviation that may be brought by a general trust degree evaluation model, can more accurately capture the relationship between the physiological characteristics of the driver individual and the trust state, improves the adaptability and recognition accuracy of the trust degree evaluation, and provides a basis for subsequent adaptive learning. Further, based on the pre-trained trust degree evaluation model of the driver, according to the real-time physiological information, actual driving environment, and behavior information, model adaptive learning is carried out to obtain the trust degree evaluation model of the driver after adaptive learning, ensuring that the trust degree evaluation model is continuously optimized according to real-time data, enabling the trust degree evaluation model to adapt to the changes of the driver state and driving environment, further improving the long-term adaptability and recognition accuracy of the trust degree evaluation, and thus enhancing the robustness and stability of the intelligent driving system in various driving environments.

[0082] Please refer to Figure 3 , which is a schematic diagram of the functional modules of the vehicle control device 100 provided in an embodiment of the present application.

[0083] In this embodiment, based on the same concept as the vehicle control method in the above Figure 1 illustrated embodiment, the present application also provides a vehicle control device 100, which can be used to execute the above vehicle control method. For the sake of convenience of description, in the schematic diagram of the composition of the vehicle control device 100 embodiment, only the parts related to the embodiments of the present application are shown. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the vehicle control device 100, and it may include more or fewer components than those illustrated, or combine certain components, or have different component arrangements.

[0084] Specifically, the vehicle control device 100 provided in the embodiment of the present application includes an acquisition module 110, an evaluation module 120, and an adjustment module 130.

[0085] The acquisition module 110 is used to obtain the real-time physiological information of the driver during the driving process of the vehicle if the opening signal of the intelligent driving system of the vehicle is detected.

[0086] In some embodiments of the present application, the vehicle may include an electroencephalogram sensing device. The electronic device 10 is communicatively connected to the electroencephalogram sensing device, and the initial electroencephalogram signal can be collected through the electroencephalogram sensing device. After the electronic device 10 detects the opening signal of the intelligent driving system, the electroencephalogram sensing device will immediately start collecting the real-time physiological information of the driver during the driving process of the vehicle. Among them, the real-time physiological information includes but is not limited to the real-time electroencephalogram signal of the driver, etc.

[0087] Furthermore, the acquisition rate of the electroencephalogram sensing device can be set to 500 Hz to ensure the high precision and real-time performance of the real-time electroencephalogram signal. The real-time electroencephalogram signal needs to cover the main frequency bands such as the alpha wave (8 - 13 Hz) and the beta wave (13 - 30 Hz) to comprehensively reflect the electroencephalogram activity of the driver.

[0088] The evaluation module 120 is used to evaluate the real-time trust level of the driver in the intelligent driving system based on a preset trust level evaluation model of the driver according to the real-time physiological information.

[0089] In some embodiments of the present application, before evaluating the driver's real-time trust in the intelligent driving system, the electronic device can perform preprocessing on the real-time physiological information. For example, a band-pass filter can be used to set the frequency range between 1 - 50 Hz and perform a filtering operation on the real-time physiological information to remove noise and power frequency interference in the real-time physiological information. Then, a fourth-order filter is used to perform a filtering operation on the real-time physiological information again to ensure the smoothness and real-time nature of the real-time physiological information, which can provide more reliable data support for subsequent trust evaluation, thereby making the driver trust evaluation more accurate.

[0090] Further, the preprocessed real-time physiological information is used as an input feature vector and input into a pre-constructed and trained trust evaluation model specific to the driver. Through the internal operation and analysis of the trust evaluation model, the output result is normalized, that is, a specific mathematical transformation algorithm is used to map the original output value to the standardized interval of [0, 1], thereby ensuring the standardization and consistency of the output result. Furthermore, the normalized output result is directly used as the real-time trust, which helps with the subsequent quantitative analysis of the real-time trust and can provide a more reliable decision-making basis for the intelligent driving system.

[0091] In some embodiments of the present application, the electronic device 10 can perform online analysis on the collected real-time physiological information according to a preset calculation period to ensure the immediacy and accuracy of trust monitoring, thereby ensuring that the driver's trust state can be accurately captured and responded to. Among them, the calculation period of the real-time trust can be set to once per second to ensure the real-time update of the trust state, and thus ensure that the vehicle can quickly respond to changes in the driver's driving state.

[0092] Further, when the electronic device 10 performs online analysis, the data caching duration can be 5 seconds, which can not only provide a sufficient number of data points for the analysis process to reduce interference caused by data fluctuations but also not cause response delays, thereby ensuring the dynamic tracking and effective response of the intelligent driving system to the driver's trust state and providing a safer and more reliable intelligent driving experience for the driver.

[0093] It should be noted that the specific steps for obtaining the trust evaluation model are described in detail above Figure 2 and will not be repeated here to avoid redundancy.

[0094] The adjustment module 130 is used to adjust the intervention frequency and intervention method of the intelligent driving system based on a preset adjustment rule according to the real-time trust. <>

[0095] In some embodiments of the present application, the electronic device 10 divides the real-time trust level into a high-trust state and a low-trust state. The high-trust level threshold can be 0.8. When the real-time trust level is greater than or equal to 0.8, it indicates that the driver is in a high-trust state; the low-trust level threshold can be 0.3. When the real-time trust level is less than or equal to 0.3, it indicates that the driver is in a low-trust state.

[0096] Among them, the intervention frequency and the intervention method are two key parameters dynamically adjusted by the intelligent driving system according to the driver's real-time trust level. The intervention frequency refers to the number of times the intelligent driving system intervenes in and controls the vehicle driving behavior within a certain time range. The intervention method is the specific form in which the intelligent driving system intervenes in and controls the vehicle.

[0097] Furthermore, the adjustment rules include but are not limited to the following rules: (1) Intervention frequency adjustment: When the real-time trust level is in the low-trust state, the intervention frequency is set to 5 times per minute to provide more assisted driving support to improve driving safety. When the real-time trust level is in the high-trust state, the intervention frequency is reduced to 1 time per minute to give the driver more autonomous control and enhance the driving experience. (2) Intervention method adjustment: When the real-time trust level is in the low-trust state, the active intervention method is adopted, such as the active enabling of functions such as automatic braking and lane keeping, to enhance the intervention in the driver's operation. When the real-time trust level is in the high-trust state, the passive intervention method is adopted, such as providing prompts or suggestions when necessary, to reduce the intervention in the driver's operation.

[0098] Among them, the delay time for the intelligent driving system to respond to the above adjustment operations shall not exceed 200 milliseconds to ensure the timeliness of the adjustment and driving safety.

[0099] The adjustment rules in the above embodiments reasonably balance driving safety and driving experience, improve the overall performance of the intelligent driving system, enable the intelligent driving system to perform well among different user groups, and broaden the application scope of the intelligent driving system.

[0100] In the vehicle control device 100 of the present application, first, when the opening signal of the intelligent driving system of the vehicle is detected, the real-time physiological information of the driver during the driving process is obtained, which can accurately reflect the psychological state of the driver. Further, the real-time physiological information is input into the trustworthiness evaluation model exclusive to the driver to obtain the real-time trustworthiness of the driver in the intelligent driving system. Considering the individual differences in physiological performance and trust performance among different drivers, it avoids the biases that may be brought by a general trustworthiness evaluation model, can more accurately capture the relationship between the physiological characteristics and trust state of the individual driver, improves the accuracy and adaptability of trust evaluation, and meets the personalized evaluation requirements. Finally, according to the real-time trustworthiness, the intervention frequency and intervention method of the intelligent driving system are adjusted in real time. This real-time dynamic adjustment method enables the intelligent driving system to better adapt to the psychological state of the driver, thereby effectively improving driving safety and reliability, reducing the probability of traffic accidents, and thus enhancing the driver's satisfaction with the intelligent driving system, which helps to promote the popularization and application of intelligent driving technology.

[0101] Please refer to Figure 4 , which is a schematic diagram of the hardware structure of the electronic device 10 provided by an embodiment of the present application.

[0102] The electronic device 10 provided by the embodiment of the present application includes, but is not limited to, a memory 11, a processor 12, and a computer program stored in the memory 11 and executable on the processor 12, such as a driver trustworthiness program. When the computer program is executed by the processor, it implements the vehicle control method as described in the above embodiment.

[0103] Figure 4 Only the electronic device 10 with a memory 11 and a processor 12 is shown here. Those skilled in the art can understand that Figure 4 the shown structure does not limit the electronic device 10, and it may include fewer or more components than shown, or combine some components, or have different component arrangements.

[0104] In some embodiments of the present application, the electronic device 10 can be communicatively connected to devices such as a desktop computer, a notebook, a handheld computer, and a cloud server.

[0105] In some embodiments of the present application, the electronic device 10 can perform human-computer interaction with the user through means such as a keyboard, a mouse, a remote control, a touchpad, or a voice control device.

[0106] In some embodiments of the present application, the electronic device 10 may further include a network device and / or a client device. Among them, the network device includes, but is not limited to, a single network server, a server group composed of multiple network servers, and a cloud server composed of a large number of hosts or network servers based on cloud computing (Cloud Computing).

[0107] In some embodiments of the present application, the network where the electronic device 10 is located includes, but is not limited to, the Internet, wide area network, metropolitan area network, local area network, virtual private network (VPN), etc.

[0108] In some embodiments of the present application, the memory 11 stores multiple computer-readable instructions to implement a vehicle control method. The processor 12 can execute multiple instructions to achieve: if an activation signal of the intelligent driving system of the vehicle is detected, obtain the real-time physiological information of the driver during the driving process; based on a preset trust degree evaluation model of the driver, evaluate the real-time trust degree of the driver in the intelligent driving system according to the real-time physiological information; based on a preset adjustment rule, adjust the intervention frequency and intervention method of the intelligent driving system according to the real-time trust degree.

[0109] Specifically, for the specific implementation method of the above instructions by the processor 12, reference can be made to Figure 1 the description of the relevant steps in the corresponding embodiments, which will not be elaborated here.

[0110] Those skilled in the art can understand that the schematic diagram is only an example of the electronic device 10, and does not constitute a limitation on the electronic device 10. The electronic device 10 can be a bus structure or a star structure. The electronic device 10 can also include more or fewer other hardware or software than shown in the figure, or different component arrangements. For example, the electronic device 10 can also include input / output devices, network access devices, etc.

[0111] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, in Figure 4 it is only represented by one arrow, but it does not mean that there is only one bus or one type of bus. The bus is set to realize the connection and communication between the memory 11 and at least one processor 12, etc.

[0112] It should be noted that the electronic device 10 is only an example. Other existing or future possible electronic products that can be adapted to the present application should also be included in the protection scope of the present application and are included herein by reference.

[0113] In some embodiments of the present application, the processor 12 may be composed of integrated circuits in some embodiments. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same or different functions, including a combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 12 is the control core (Control Unit) of the electronic device 10, connecting various components of the entire electronic device 10 through various interfaces and lines, and by running or executing programs or modules stored in the memory 11 (such as executing a driver trust program, etc.), and calling the data stored in the memory 11 to perform various functions of the electronic device 10 and process data.

[0114] The processor 12 executes the operating system of the electronic device 10 and various installed application programs. The processor 12 executes the application programs to implement the steps in each of the above embodiments of a vehicle control method, such as Figure 1 the steps shown.

[0115] Exemplarily, the computer program can be divided into one or more modules / units. One or more modules / units are stored in the memory 11 and executed by the processor 12 to complete the present application. One or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device 10. For example, the computer program can be divided into an acquisition module 110, an evaluation module 120, and an adjustment module 130.

[0116] The above integrated units implemented in the form of software function modules can be stored in a computer-readable storage medium. The above software function modules are stored in a storage medium, including several instructions for causing a computer device (which can be a personal computer, a computer device, or a network device, etc.) or a processor (Processor) to execute a part of the vehicle control method in each embodiment of the present application.

[0117] If the integrated module / unit of the electronic device 10 is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above embodiment methods of the present application, it can also be completed by a computer program instructing relevant hardware devices. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above method embodiments can be implemented.

[0118] Among them, the computer program includes computer program code, which can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory and other memories, etc.

[0119] An embodiment of the present application further provides a computer-readable storage medium (not shown in the figure). The computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions are executed by a processor in the electronic device 10 to implement a vehicle control method according to any one of the above embodiments.

[0120] Specifically, the computer-readable storage medium can be non-volatile or volatile. The computer-readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD memory, DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 can be an internal storage unit of the electronic device 10 in some embodiments, such as the mobile hard disk of the electronic device 10. The memory 11 can also be an external storage device of the electronic device 10 in other embodiments, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 10. The memory 11 can be used not only to store application software installed in the electronic device 10 and various types of data, such as the code of a driver trust program, etc., but also to temporarily store data that has been output or will be output.

[0121] Furthermore, the computer-readable storage medium mainly includes a storage program area and a storage data area. Among them, the storage program area can store an operating system, application programs required for at least one function, etc.; the storage data area can store data created according to the use of the blockchain node, etc.

[0122] In the embodiments of the present application, it should be noted that unless otherwise clearly specified and limited, the meaning of "a plurality of" refers to two or more.

[0123] In the embodiments of the present application, it should be noted that unless otherwise clearly specified and limited, words such as "for example" are used to give examples, illustrations or explanations. Any embodiment or design solution described as "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, using words such as "for example" aims to present relevant concepts in a specific manner.

[0124] In the description of the present application, it should be noted that unless otherwise clearly specified or limited, the terms "install", "connect", and "couple" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or a connection that allows mutual communication; it can be a direct connection, or an indirect connection through an intermediate medium, and it can be the internal communication between two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0125] In the description of the present application, it should be noted that unless otherwise clearly specified or limited, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more features.

[0126] In the description of the present application, it should be noted that unless otherwise clearly specified or limited, "and / or" is merely a correlative relationship describing associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present application generally represents an "or" relationship between the associated objects before and after.

[0127] If there is no special instruction, all steps of the present application can be carried out in sequence or randomly. For example, the method includes steps A and B, indicating that the method may include steps A and B carried out in sequence, or may also include steps B and A carried out in sequence. For example, it is mentioned that the method may further include step C, indicating that step C can be added to the method in any order. For example, the method may include steps A, B, and C, or may also include steps A, C, and B, or may also include steps C, A, and B, etc.

[0128] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0129] In several embodiments provided by the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0130] In each embodiment of the present application, each functional module may be integrated into a processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware, or in the form of a combination of hardware and software functional modules.

[0131] In addition, it is obvious that the word "including" does not exclude other units or steps, and the singular does not exclude the plural. A plurality of units or devices described in the specification may also be implemented by one unit or device through software or hardware.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit them. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application may be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A vehicle control method, characterized in that, Applied to a vehicle, the method includes: If an activation signal of the intelligent driving system of the vehicle is detected, obtain the real-time physiological information of the driver during the process of driving the vehicle; Based on the preset trustworthiness evaluation model of the driver, evaluate the real-time trustworthiness of the driver in the intelligent driving system according to the real-time physiological information; Based on the preset adjustment rules, adjust the intervention frequency and intervention method of the intelligent driving system according to the real-time trustworthiness.

2. The vehicle control method according to claim 1, wherein Before the method evaluates the real-time trustworthiness of the driver in the intelligent driving system according to the real-time physiological information based on the preset trustworthiness evaluation model of the driver, it further includes: Obtain the initial physiological information of the driver when driving the vehicle in a variety of preset driving environments, where the initial physiological information is marked with the initial trustworthiness of the driver in the intelligent driving system; Based on the initial physiological information and the corresponding initial trustworthiness of the initial physiological information, and using the cross-entropy loss function for training and optimization, obtain the pre-trained trustworthiness evaluation model of the driver.

3. The vehicle control method according to claim 2, wherein After the method obtains the pre-trained trustworthiness evaluation model of the driver, it further includes: Obtain the actual driving environment of the driver during the process of driving the vehicle and the behavior information in the driving environment; Based on the pre-trained trustworthiness evaluation model of the driver, perform model adaptive learning according to the real-time physiological information, the actual driving environment, and the behavior information, and obtain the trustworthiness evaluation model of the driver after adaptive learning.

4. The vehicle control method according to claim 3, characterized in that, The performing model adaptive learning based on the pre-trained trustworthiness evaluation model of the driver according to the real-time physiological information, the actual driving environment, and the behavior information includes: Use the pre-trained trustworthiness evaluation model of the driver as the source domain model, use the real-time physiological information as the target domain data, and based on the source domain model and the target domain data, adjust the parameters of the source domain model.

5. The vehicle control method according to claim 3, characterized in that, The performing model adaptive learning based on the pre-trained trustworthiness evaluation model of the driver according to the real-time physiological information, the actual driving environment, and the behavior information further includes: If the actual driving environment is a new driving environment, use the behavior information as the feedback input of the pre-trained trustworthiness evaluation model of the driver, and combine the real-time physiological information to adjust the parameters of the pre-trained trustworthiness evaluation model of the driver.

6. The vehicle control method according to claim 3, wherein The performing model adaptive learning based on the pre-trained trustworthiness evaluation model of the driver according to the real-time physiological information, the actual driving environment, and the behavior information further includes: Divide the real-time physiological information based on a preset time period to obtain a plurality of batch data; Based on the time period, adjust the parameters of the pre-trained trustworthiness evaluation model of the driver in sequence according to each batch data in the plurality of batch data.

7. The vehicle control method according to claim 3, wherein Based on the pre-trained driver trust evaluation model, the model adaptive learning according to the real-time physiological information, the actual driving environment, and the behavior information further includes: If the acquisition quantities of the real-time physiological information all reach preset values, adjust the parameters of the pre-trained driver trust evaluation model.

8. A vehicle control device, characterized in that, Applied to a vehicle, the device includes: An acquisition module, configured to, if an activation signal of the intelligent driving system of the vehicle is detected, acquire the real-time physiological information of the driver during the driving of the vehicle; An evaluation module, configured to evaluate the real-time trust of the driver in the intelligent driving system based on the preset driver trust evaluation model and according to the real-time physiological information; An adjustment module, configured to adjust the intervention frequency and intervention method of the intelligent driving system based on a preset adjustment rule and according to the real-time trust.

9. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the vehicle control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the vehicle control method according to any one of claims 1 to 7.

Images