Auxiliary driving risk prediction method based on driver state monitoring and related device

Abstract

The invention discloses an auxiliary driving risk prediction method based on driver state monitoring and a related device, and relates to the technical field of auxiliary driving risk prediction. In the scheme of the invention, unsupervised clustering is carried out on historical driver physiological data and historical vehicle driving data by using a clustering algorithm; preliminarily dividing collision risk grades of automobile operation; constructing a training data set and a test data set according to a plurality of groups of historical driver physiological data, historical vehicle driving data and corresponding collision risk levels, and obtaining a trained collision risk prediction model; and finally, based on the actually measured vehicle driving data and the actually measured driver physiological data, risk prediction is carried out by using the collision risk prediction model so as to obtain an auxiliary driving risk prediction result. In addition, the auxiliary driving risk prediction result can also be used for updating division of collision risk levels in combination with actual conditions. The problem of insufficient prediction accuracy caused by the fact that a traditional risk prediction method is only based on vehicle operation data is solved, and the prediction accuracy of the driving risk is improved.

Description

Technical Field

[0001] The present application relates to the technical field of assisted driving risk prediction, and in particular to an assisted driving risk prediction method and related devices based on driver status monitoring. Background Art

[0002] With the continuous development of intelligent driving technology, the role of assisted driving systems in improving driving safety and reducing traffic accidents is becoming increasingly significant. However, existing assisted driving systems still have obvious shortcomings when dealing with complex road conditions and emergencies. For example, their real-time monitoring of the driver's status and risk prediction capabilities are relatively weak. Traditional risk prediction methods are usually based only on vehicle operating data, but ignore the key factor of the driver's physiological state, resulting in low prediction accuracy when dealing with emergencies. Moreover, during long driving or complex road conditions, factors such as driver fatigue, distraction, and emotional fluctuations will greatly increase the risk of accidents. Therefore, there is an urgent need for a risk prediction system that can comprehensively integrate the driver's physiological state and vehicle operating data to more accurately judge the risk level during driving and improve the intelligence level and safety performance of assisted driving systems. Summary of the Invention

[0003] The purpose of this application is to provide an assisted driving risk prediction method and related devices based on driver status monitoring, which can accurately determine the risk level in the assisted driving process.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides an assisted driving risk prediction method based on driver status monitoring, comprising the following steps:

[0006] Several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before a collision during simulated vehicle driving are collected.

[0007] A clustering algorithm is used to perform unsupervised clustering on several groups of historical driver physiological data and historical vehicle driving data to preliminarily divide the collision risk level of vehicle operation.

[0008] Based on several sets of historical driver physiological data and historical vehicle driving data and the corresponding collision risk levels, a training dataset and a test dataset are constructed; the training dataset is used to train the collision risk prediction model; the test dataset is used to test the performance of the collision risk prediction model.

[0009] Based on the measured vehicle driving data and the measured driver physiological data, a collision risk prediction model is used to perform risk prediction to obtain assisted driving risk prediction results; the assisted driving risk prediction results are also used to update the classification of collision risk levels based on actual conditions.

[0010] Optionally, collecting several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before a collision occurs during simulated vehicle driving, specifically includes the following steps:

[0011] The driver's pulse wave signal data is collected within 5 seconds before the collision occurs during simulated vehicle driving as historical driver physiological data.

[0012] For any historical driver physiological data, the corresponding distance to the preceding vehicle, vehicle driving time and vehicle driving speed are collected as historical vehicle driving data.

[0013] Optionally, constructing a training dataset and a test dataset based on several sets of historical driver physiological data and historical vehicle driving data and corresponding collision risk levels specifically includes the following steps:

[0014] For any historical driver physiological data, the driver's heart rate is calculated based on the historical driver physiological data.

[0015] For any historical vehicle driving data, the maximum deceleration and the distance to the preceding vehicle before the collision are determined according to the historical vehicle driving data.

[0016] The heart rate of any group of drivers and the corresponding maximum deceleration, distance to the vehicle in front before collision, and vehicle speed are used as model inputs, and the corresponding collision risk level is used as a label to construct a sample data.

[0017] Divide several sample data into training data sets and test data sets according to the preset ratio.

[0018] Optionally, the driver's heart rate is calculated according to the following formula:

[0019]

[0020] Where HR is the driver's heart rate, and t2-t1 represents the time difference between two consecutive peaks of the pulse wave data.

[0021] The maximum deceleration is determined according to the following formula:

[0022]

[0023] Among them, a max is the maximum deceleration, v f is the final velocity 5s before the collision, v iis the initial velocity, and T is the time from the initial moment to 5s before the collision.

[0024] Determine the distance to the vehicle in front before the collision using the following formula:

[0025] D i =D a -D b .

[0026] Among them, D i D is the distance to the vehicle in front 5 seconds before the collision. a is the distance to the preceding vehicle, D b The distance to the main vehicle.

[0027] Optionally, a clustering algorithm is used to perform unsupervised clustering on several groups of historical driver physiological data and historical vehicle driving data to preliminarily classify the collision risk level of vehicle operation, specifically including the following steps:

[0028] Several sets of historical driver physiological data and historical vehicle driving data are input into the k-means model for training, and unsupervised learning is performed. Training is performed under different cluster numbers k, and the cluster centers and family distribution under each cluster number k are obtained.

[0029] According to the cluster distribution under each cluster number k, the target cluster number with the best clustering effect is selected, and each class center obtained by dividing the target cluster number corresponds to a different collision risk level.

[0030] Optionally, the collision risk prediction model is a model constructed and trained using the XGBoost algorithm; the training set data is input into the XGBoost algorithm model for training, and multiple decision trees are iteratively constructed to continuously optimize the loss function so that the overall prediction accuracy is gradually improved; after obtaining the trained collision risk prediction model, the test data set is input into the collision risk prediction model to calculate the mean square error loss, and when the mean square error loss is lower than a preset threshold, the collision risk prediction model is applied to the risk prediction of measured vehicle driving data and measured driver physiological data.

[0031] Secondly, this application provides an assisted driving risk prediction system based on driver status monitoring, including the following functional modules:

[0032] The historical simulated driving data acquisition module is used to collect several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before the collision occurs in simulated vehicle driving.

[0033] The module for preliminary classification of collision risk levels is used to perform unsupervised clustering of several groups of historical driver physiological data and historical vehicle driving data using a clustering algorithm to preliminarily classify the collision risk level of vehicle operation.

[0034] The training and test dataset construction module is used to construct a training dataset and a test dataset based on several sets of historical driver physiological data and historical vehicle driving data and corresponding collision risk levels; the training dataset is used to train the collision risk prediction model; the test dataset is used to test the performance of the collision risk prediction model.

[0035] The assisted driving collision risk prediction module is used to perform risk prediction based on the measured vehicle driving data and the measured driver physiological data using a collision risk prediction model to obtain assisted driving risk prediction results; the assisted driving risk prediction results are also used to update the classification of collision risk levels based on actual conditions.

[0036] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the aforementioned method for assisted driving risk prediction based on driver status monitoring.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the assisted driving risk prediction method based on driver status monitoring described above are implemented.

[0038] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the assisted driving risk prediction method based on driver status monitoring described above.

[0039] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0040] The present application provides an assisted driving risk prediction method and related device based on driver status monitoring. In this method, first, several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before a collision occurs during simulated vehicle driving are collected. Then, a clustering algorithm is used to perform unsupervised clustering on these several sets of historical driver physiological data and historical vehicle driving data to preliminarily divide the collision risk level of the vehicle operation. Then, based on the several sets of historical driver physiological data, historical vehicle driving data and corresponding collision risk levels, a training data set and a test data set are constructed, and a trained collision risk prediction model is obtained; finally, based on the measured vehicle driving data and the measured driver physiological data, the collision risk prediction model is used to perform risk prediction, thereby obtaining an assisted driving risk prediction result. In addition, the assisted driving risk prediction result will also be used to update the classification of collision risk levels based on actual conditions. This application collects driver physiological data and vehicle driving data at the same time, realizing the effective integration of multi-dimensional data, solving the problem of insufficient prediction accuracy caused by traditional risk prediction methods based only on vehicle operation data, and improving the accuracy of driving risk prediction. After obtaining the assisted driving risk prediction results, it also combines the actual situation to feed back to the classification of collision risk levels, realizes real-time update of collision risk levels, optimizes the adaptability of the model, and improves the intelligence level and safety performance of the assisted driving system. In summary, this application accurately judges the collision risk in the assisted driving process, guides the driver to execute driving strategies under different traffic conditions, effectively improves the safety of following vehicle driving, and improves traffic efficiency and urban road traffic safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 A flowchart of a method for predicting risk in assisted driving based on driver status monitoring is provided in one embodiment of the present application.

[0043] Figure 2 This is a flowchart of step S1 in an assisted driving risk prediction method based on driver status monitoring provided in one embodiment of the present application.

[0044] Figure 3 This is a flowchart of step S2 in an assisted driving risk prediction method based on driver status monitoring provided in one embodiment of the present application.

[0045] Figure 4This is a flowchart of step S3 in an assisted driving risk prediction method based on driver status monitoring provided in one embodiment of the present application.

[0046] Figure 5 A schematic diagram of the functional modules of an assisted driving risk prediction system based on driver status monitoring provided in one embodiment of the present application.

[0047] Figure 6 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0049] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0050] The embodiment of the present application provides an assisted driving risk prediction method based on driver status monitoring. In an exemplary embodiment, Figure 1 As shown, the following steps are included:

[0051] S1. Collect several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before the collision occurs in simulated vehicle driving. Use the driver physiological signal intelligent measurement device to collect driver physiological data, and the vehicle driving track recorder to collect vehicle driving data. In this embodiment, Figure 2 As shown, step S1 specifically includes the following steps:

[0052] S11. Collect the driver's pulse wave signal data within 5 seconds before the collision occurs during simulated vehicle driving as historical driver physiological data.

[0053] S12. For any historical driver physiological data, collect the corresponding distance to the preceding vehicle, vehicle driving time, and vehicle driving speed as historical vehicle driving data.

[0054] S2, using clustering algorithms to perform unsupervised clustering on several groups of historical driver physiological data and historical vehicle driving data, and preliminarily classify the collision risk level of the vehicle operation. Figure 3 As shown, step S2 specifically includes the following steps:

[0055] S21. Inputting several sets of historical driver physiological data and historical vehicle driving data into the k-means model for training, performing unsupervised learning, training under different cluster numbers k, and obtaining the cluster center and family distribution under each cluster number k.

[0056] S22. According to the cluster distribution under each cluster number k, a target cluster number with the best clustering effect is selected, and each class center obtained by dividing the target cluster number corresponds to a different collision risk level.

[0057] As an exemplary embodiment, the number of clusters k is selected as 3. At this time, the family distribution is optimal, and the class centers are: cluster0 represents a stable heart rate, a large distance, and a moderate speed, and is marked as "low risk"; cluster1 represents a medium level, and is marked as "medium risk"; cluster2 represents a high heart rate, a fast speed, and a short distance, and is marked as "high risk".

[0058] S3, based on several sets of historical driver physiological data and historical vehicle driving data and corresponding collision risk levels, construct a training data set and a test data set; the training data set is used to train the collision risk prediction model; the test data set is used to test the performance of the collision risk prediction model. In this embodiment, Figure 4 As shown, step S3 specifically includes the following steps:

[0059] S31. For any historical driver physiological data, calculate the driver's heart rate based on the historical driver physiological data. Specifically, calculate the driver's heart rate according to the following formula:

[0060]

[0061] Where HR is the driver's heart rate, and t2-t1 represents the time difference between two consecutive peaks of the pulse wave data.

[0062] S32. For any historical vehicle driving data, determine the maximum deceleration and the distance to the preceding vehicle before the collision based on the historical vehicle driving data. In this embodiment, the maximum deceleration is determined according to the following formula:

[0063]

[0064] Among them, a max is the maximum deceleration, v f is the final velocity 5s before the collision, v i is the initial velocity, and T is the time from the initial moment to 5s before the collision.

[0065] The distance to the vehicle in front before the collision can be determined according to the following formula:

[0066] Di =D a -D b .

[0067] Among them, D i D is the distance to the vehicle in front 5 seconds before the collision. a is the distance to the preceding vehicle, D b The distance to the main vehicle.

[0068] S33. Using any group of driver's heart rates and the corresponding maximum deceleration, distance to the preceding vehicle before collision, and vehicle speed as model inputs, and the corresponding collision risk level as a label, a sample data set is constructed.

[0069] Specifically, the risk level judgment criteria are:

[0070] High risk: driver's heart rate ≥100 bpm, vehicle speed ≥80 km / h, distance to the vehicle in front before collision ≤10 m, maximum deceleration ≥3.0 m / s 2 .

[0071] Medium risk: The driver's heart rate is between 80-100 bpm, the vehicle's speed is between 40-80 km / h, the distance to the vehicle in front before the collision is 10-20 meters, and the maximum deceleration is 1.5-3.0 m / s. 2 .

[0072] Low risk: driver's heart rate ≤80bpm, vehicle speed ≤40km / h, distance from the vehicle in front before collision ≥20m, maximum deceleration ≤1.5m / s 2 .

[0073] S34: Divide the plurality of sample data into a training data set and a test data set according to a preset ratio. In this embodiment, the data set is divided into a training data set and a test data set according to a ratio of 4:1.

[0074] The training dataset is used to train the collision risk prediction model. Specifically, at the beginning of model training, the initial prediction value is set. (such as all sample labels y i The mean of ), as the baseline prediction result before the first decision tree training, the prediction result of the overall model is The prediction result of the t-th tree is the sum of the cumulative prediction results of the previous t-1 trees and the prediction output of the t-th tree for sample i, as shown in the following formula:

[0075]

[0076] in, Represents the prediction results of the first t trees for sample i, represents the cumulative prediction results of the first t-1 trees, ft (X i ) represents the predicted output of the t-th tree for sample i.

[0077] Next, we construct the objective function. According to the XGBoost algorithm structure defined in the solution, the objective function includes a loss function and a regularization term, specifically:

[0078]

[0079] Among them, Obj(θ) is the objective function, θ is the model parameter, I represents the loss function, y i is the true value of sample i, Ω(f i ) is a regularization term used to control the complexity of the t-th decision tree (such as the depth of the tree and the number of leaf nodes) to prevent the model from overfitting and ensure reliability in practical applications. Constant is a constant term in the XGBoost model.

[0080] Each round of iteration (round t) builds a new decision tree f t , by optimizing the objective function to reduce the prediction error, the specific process is as follows:

[0081] T1. Calculate the negative gradient (residual) and determine the learning target for the new tree. Specifically, based on the current cumulative prediction value and the loss function, calculate the negative gradient of each sample as the target value to be fitted by the t-th tree. This negative gradient reflects the current model's prediction deviation for sample i, and the training goal of the t-th tree is to minimize this deviation.

[0082] T2, generate candidate split points of the decision tree and build the tree structure. Specifically, for the input feature vector X i =[H i ,a i ,D i ] (i.e., heart rate, maximum deceleration, and distance between two vehicles), sort the values ​​of each feature and generate candidate split points (such as quantile division based on the feature value). Traverse all candidate split points, calculate the residual sum of squares of the left and right subtrees after the split, select the split point that makes the "residual sum of squares after the split the largest", and recursively construct the node structure of the decision tree (such as splitting the "distance to the preceding vehicle" feature first, then splitting the "heart rate" feature, etc.). In this embodiment, the regularization term Ω(f i ) Limit the depth of the tree and the number of leaf nodes (such as controlling the maximum depth and the minimum number of split samples) to avoid overfitting caused by excessively complex tree structures.

[0083] T3. Calculate the leaf node weights and optimize the output of the current tree. Specifically, after the decision tree structure is determined (i.e., the leaf node division is completed), calculate the optimal weight w for each leaf node. j(j is the leaf node number), to minimize the objective function. The tth tree f t (x i The output of ) is the weight w of the leaf node to which sample i belongs. j .

[0084] T4, update the cumulative prediction value to reduce the overall loss. Specifically, the output f of the tth tree t (x i ) is added to the previous cumulative forecast value Get the new predicted value At this time, the overall loss function value (mean square error) decreases due to the reduction of residuals, and the prediction accuracy is improved.

[0085] Repeat steps T1 to T4, continuously generating new decision trees (t = 1, 2, ..., T) until the iteration termination condition is met (such as reaching the preset maximum number of trees T, or the loss function decreases by less than a threshold). During the training process, model performance is optimized by adjusting hyperparameters (such as the maximum tree depth, learning rate, regularization parameter, etc.). The solution uses the training set to find the optimal hyperparameters to ensure the model's generalization ability on the test set and avoid overfitting.

[0086] After multiple rounds of iteration, an XGBoost model consisting of T decision trees is obtained. Its final prediction value for sample i corresponds to the numerical label of the risk level (0, 1, 2), combined with the risk level interval obtained by k-means clustering (such as Compared with the cluster center), when For low risk, When the risk is medium, When the risk is high, the prediction results of low, medium and high risks can be output to achieve accurate prediction of driving risks.

[0087] The test data set is used to determine the accuracy of the model by calculating the mean square error function according to the following formula:

[0088]

[0089] when , it indicates that the prediction effect of the trained collision risk prediction model is accurate and can be applied.

[0090] S4. Based on the measured vehicle driving data and the measured driver physiological data, a collision risk prediction model is used to perform risk prediction and obtain an assisted driving risk prediction result. The assisted driving risk prediction result is also used to update the collision risk level classification based on actual conditions. The assisted driving risk prediction result is the collision risk level to be predicted. Based on the risk level predicted by the collision risk model, the relevant indicators "driver's heart rate, distance between two vehicles, vehicle and driving speed" are then calculated. The rationality of the predicted result is determined based on actual conditions. If it is not rational, the relevant indicators can be adjusted. If it is reasonable, no changes are required. In actual application, the range of relevant indicators can also be determined based on local traffic conditions.

[0091] Compared with the prior art, the method provided by the above embodiment of the present application has the following beneficial effects:

[0092] 1. In this embodiment, the data acquisition module simultaneously collects the driver's physiological data and vehicle driving data, realizing the effective fusion of multi-dimensional data, solving the problem of insufficient prediction accuracy caused by traditional risk prediction methods based only on vehicle operation data, and improving the accuracy of driving risk prediction.

[0093] 2. The k-means unsupervised learning algorithm is used to classify collision risk levels, which can scientifically divide them into low, medium and high risk levels, achieve accurate classification of different risk levels, and better balance the impact of different types of data on the overall model during the classification process.

[0094] 3. Select the XGBoost algorithm model for risk prediction. This algorithm iteratively constructs multiple decision trees and continuously optimizes the loss function, which helps solve the overfitting problem that may occur during model training and ensures the reliability of the model in practical applications.

[0095] 4. By feeding back the prediction results of the risk prediction module to the risk level judgment module, real-time updating of the risk level is achieved, the adaptability of the model is optimized, and the intelligence level and safety performance of the assisted driving system are improved.

[0096] 5. The method provided in this application can more accurately judge the risk level during driving, guide drivers to implement driving strategies under different traffic conditions, effectively improve the safety of following driving, and improve traffic efficiency and urban road traffic safety.

[0097] Based on the same inventive concept, the embodiment of the present application also provides a system for implementing the above-mentioned assisted driving risk prediction method based on driver status monitoring. The solution provided by the system is similar to the solution described in the above-mentioned method. In an exemplary embodiment, Figure 5As shown, an assisted driving risk prediction system based on driver status monitoring is provided, comprising:

[0098] The historical simulated driving data acquisition module is used to collect several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before the collision occurs in simulated vehicle driving.

[0099] The module for preliminary classification of collision risk levels is used to perform unsupervised clustering of several groups of historical driver physiological data and historical vehicle driving data using a clustering algorithm to preliminarily classify the collision risk level of vehicle operation.

[0100] The training and test dataset construction module is used to construct a training dataset and a test dataset based on several sets of historical driver physiological data and historical vehicle driving data and corresponding collision risk levels; the training dataset is used to train the collision risk prediction model; the test dataset is used to test the performance of the collision risk prediction model.

[0101] The assisted driving collision risk prediction module is used to perform risk prediction based on the measured vehicle driving data and the measured driver physiological data using a collision risk prediction model to obtain assisted driving risk prediction results; the assisted driving risk prediction results are also used to update the classification of collision risk levels based on actual conditions.

[0102] certainly, Figure 5 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different functions. Figure 5 One or at least two components of the system shown.

[0103] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it can implement an assisted driving risk prediction method based on driver status monitoring provided in the above embodiment.

[0104] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0105] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0106] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0107] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0108] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0109] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0110] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0111] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0112] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. An assisted driving risk prediction method based on driver status monitoring, characterized in that: include: Collect several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before the collision during simulated vehicle driving; Using a clustering algorithm to perform unsupervised clustering on several groups of the historical driver physiological data and the historical vehicle driving data, and preliminarily classify the collision risk level of the vehicle operation; constructing a training dataset and a test dataset based on several sets of the historical driver physiological data and the historical vehicle driving data and corresponding collision risk levels; The training data set is used to train the collision risk prediction model; The test data set is used to test the performance of the collision risk prediction model; Based on the measured vehicle driving data and the measured driver physiological data, the collision risk prediction model is used to perform risk prediction to obtain an assisted driving risk prediction result; The assisted driving risk prediction result is also used to update the classification of the collision risk level in combination with actual conditions.

2. The assisted driving risk prediction method based on driver status monitoring according to claim 1 is characterized in that: Collect several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before the collision during simulated vehicle driving, including: Collect the driver's pulse wave signal data within 5 seconds before the collision during simulated vehicle driving as historical driver physiological data; For any historical driver physiological data, the corresponding distance to the preceding vehicle, vehicle driving time and vehicle driving speed are collected as historical vehicle driving data.

3. The assisted driving risk prediction method based on driver status monitoring according to claim 2 is characterized in that: Based on several sets of the historical driver physiological data and the historical vehicle driving data and the corresponding collision risk levels, a training data set and a test data set are constructed, specifically including: For any of the historical driver physiological data, calculate the driver's heart rate according to the historical driver physiological data; For any of the historical vehicle driving data, determining a maximum deceleration and a distance to a preceding vehicle before a collision based on the historical vehicle driving data; Using any set of the driver's heart rate and the corresponding maximum deceleration, distance to the preceding vehicle before collision, and vehicle speed as model inputs, and the corresponding collision risk level as a label, a sample data set is constructed; Divide several sample data into training data sets and test data sets according to the preset ratio.

4. The assisted driving risk prediction method based on driver status monitoring according to claim 3 is characterized in that: The driver's heart rate is calculated according to the following formula: Where HR is the driver's heart rate, t2-t1 represents the time difference between two consecutive peaks of the pulse wave data; The maximum deceleration is determined according to the following formula: Among them, a max is the maximum deceleration, v f is the final velocity 5s before the collision, v i is the initial velocity, T represents the time from the initial moment to 5s before the collision; Determine the distance to the vehicle in front before the collision using the following formula: D i =D a -D b ; Among them, D i D is the distance to the vehicle in front 5 seconds before the collision. a is the distance to the preceding vehicle, D b The distance to the main vehicle.

5. The assisted driving risk prediction method based on driver status monitoring according to claim 1 is characterized in that: A clustering algorithm is used to perform unsupervised clustering on several groups of the historical driver physiological data and the historical vehicle driving data, and a preliminary classification of the collision risk level of the vehicle operation is performed, specifically including: Inputting several sets of the historical driver physiological data and the historical vehicle driving data into a k-means model for training, performing unsupervised learning, training under different cluster numbers k, and obtaining cluster centers and family distribution under each cluster number k; According to the cluster distribution under each cluster number k, the target cluster number with the best clustering effect is selected, and each class center obtained by dividing according to the target cluster number corresponds to a different collision risk level.

6. The assisted driving risk prediction method based on driver status monitoring according to claim 1 is characterized in that: The collision risk prediction model is a model constructed and trained using the XGBoost algorithm; the training set data is input into the XGBoost algorithm model for training, and multiple decision trees are iteratively constructed to continuously optimize the loss function, so that the overall prediction accuracy is gradually improved; After obtaining the trained collision risk prediction model, the test data set is input into the collision risk prediction model to calculate the mean square error loss. When the mean square error loss is lower than a preset threshold, the collision risk prediction model is applied to the risk prediction of the measured vehicle driving data and the measured driver physiological data.

7. An assisted driving risk prediction system based on driver status monitoring, characterized in that: include: A historical simulated driving data collection module is used to collect several sets of historical driver physiological data and corresponding historical vehicle driving data within 5 seconds before the collision occurs during simulated vehicle driving; a collision risk level preliminary classification module, configured to perform unsupervised clustering on a plurality of groups of the historical driver physiological data and the historical vehicle driving data using a clustering algorithm, and to preliminarily classify the collision risk level of the vehicle operation; A training and testing data set construction module, configured to construct a training data set and a testing data set based on several sets of the historical driver physiological data and the historical vehicle driving data and corresponding collision risk levels; The training data set is used to train the collision risk prediction model; The test data set is used to test the performance of the collision risk prediction model; An assisted driving collision risk prediction module is used to perform risk prediction based on measured vehicle driving data and measured driver physiological data using the collision risk prediction model to obtain an assisted driving risk prediction result; The assisted driving risk prediction result is also used to update the classification of the collision risk level in combination with actual conditions.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and runnable on the processor, characterized in that the processor executes the computer program to implement the assisted driving risk prediction method based on driver status monitoring according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the assisted driving risk prediction method based on driver status monitoring according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the assisted driving risk prediction method based on driver status monitoring according to any one of claims 1 to 6 is implemented.

Images