Risk prompting system for driving habits of students

Abstract

The invention discloses a student driving habit risk prompting system, and relates to the technical field of intelligent traffic, and the system comprises the steps: collecting the driving habit data of a student during driving, and constructing a driving behavior feature model; comparing the driving behavior characteristic model with a safe driving standard model to obtain risk behavior characteristic data; performing risk level classification on the risk behavior feature data to obtain risk levels; and generating risk prompt information according to the risk level, and prompting the student. According to the invention, the technical problems of insufficient driving behavior evaluation precision and real-time performance in the prior art are solved, and the technical effects of monitoring and evaluating the driving behavior of the student in real time and improving the driving safety are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a risk warning system for students' driving habits. Background Art

[0002] With the rapid development of the transportation industry and the continuous increase in the number of cars, the monitoring and evaluation of students' driving behavior are also facing increasing challenges. Traditional driving behavior evaluation methods usually rely on manual supervision and subsequent feedback, and have problems such as inaccurate evaluation, slow response, and lack of personalized suggestions, which are difficult to meet the needs of improving modern driving safety. Especially during driving training, the lack of real-time, dynamic monitoring and immediate feedback of students' driving behavior has led to many potential safety hazards not being discovered and corrected in time. Summary of the invention

[0003] The present application provides a risk warning system for students' driving habits, which is used to solve the technical problems of the existing technology in terms of the accuracy and real-time performance of driving behavior assessment.

[0004] In view of the above problems, this application provides a risk warning system for students' driving habits.

[0005] The present application provides a risk warning system for students' driving habits, the system comprising:

[0006] A feature model building module, the feature model building module collects the driving habit data of the trainee when driving and builds a driving behavior feature model; a feature comparison module, the feature comparison module compares the driving behavior feature model with the safe driving standard model to obtain risk behavior feature data; a risk classification module, the risk classification module classifies the risk behavior feature data into risk levels to obtain risk levels; a risk warning module, the risk warning module generates risk warning information according to the risk level and prompts the trainee.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] This application collects the driving habit data of trainees when driving, and constructs a driving behavior characteristic model; compares the driving behavior characteristic model with the safe driving standard model to obtain risk behavior characteristic data; classifies the risk behavior characteristic data into risk levels to obtain risk levels; generates risk warning information based on the risk levels to prompt the trainees. The present invention solves the technical problems of the shortcomings of the prior art in terms of driving behavior assessment accuracy and real-time performance, by collecting the driving habit data of trainees, constructing a driving behavior characteristic model, and comparing it with the safe driving standard model to obtain risk behavior characteristic data, then classifying the risk levels, assessing driving risks, and generating corresponding risk warning information based on the assessment results, thereby achieving real-time monitoring and assessment of trainees' driving behaviors and improving the technical effects of driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of the structure of a risk warning system for students' driving habits provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of the process of constructing a driving behavior characteristic model for the risk warning system of the student's driving habits provided in an embodiment of the present application.

[0012] Explanation of the accompanying drawings: feature model building module 11, feature comparison module 12, risk classification module 13, risk warning module 14. DETAILED DESCRIPTION

[0013] The present application aims to solve the technical problems of the existing technology in terms of the accuracy and real-time performance of driving behavior assessment by providing a risk warning system for students' driving habits. By collecting students' driving habit data, a driving behavior characteristic model is constructed, and the model is compared with the safe driving standard model to obtain risk behavior characteristic data. Subsequently, risk level classification is performed, driving risks are assessed, and corresponding risk warning information is generated based on the assessment results, thereby achieving real-time monitoring and assessment of students' driving behavior and improving the technical effect of driving safety.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0016] Examples, such as Figure 1 As shown, the embodiment of the present application provides a risk warning system for students' driving habits, the system comprising:

[0017] The feature model building module 11 collects the driving habit data of the trainee during driving and builds a driving behavior feature model.

[0018] In an embodiment of the present application, the feature model building module first collects key data of the student's driving process in real time at multiple times through on-board sensors or smart devices. These data include steering angle, acceleration, braking force, vehicle speed, and road driving position.

[0019] Next, the collected driving habit data is cleaned and preprocessed. Finally, based on the preprocessed data, the key features of driving behavior are extracted, and a driving behavior feature model is constructed based on these features.

[0020] Further, such as Figure 2 As shown, in the system provided by the embodiment of the application, the feature model building module 11 is also used for:

[0021] The steering data, acceleration data, braking data, speed data and road position data of the trainee at multiple moments of driving are collected as driving habit data; the driving habit data are preprocessed by data cleaning to obtain preprocessed driving habit data; and a driving behavior characteristic model is constructed based on the preprocessed driving habit data.

[0022] In the embodiment of the present application, firstly, driving habit data of the trainee at multiple moments are collected. The multiple moments are multiple moments in a very short time while driving, for example, multiple moments within five seconds. To collect driving habit data, the trainee's steering data (such as steering angle and steering frequency), acceleration data (such as acceleration, acceleration duration, etc.), braking data (such as braking distance, braking acceleration, etc.), speed data (such as vehicle speed, speed fluctuation, etc.) and road driving position data are recorded at multiple moments by the vehicle-mounted sensors, and the collected data are used as driving habit data.

[0023] Next, the collected driving habit data is preprocessed by data cleaning. Specifically, the collected driving habit data is feature extracted to obtain steering angular velocity, acceleration, braking acceleration, driving speed, and road center deviation distance. Then the data at each moment is averaged. By performing mean cleaning on features such as steering angular velocity, acceleration, braking acceleration, driving speed, and road center deviation distance, the average value of each feature is obtained, and these average values ​​are used as preprocessed driving habit data.

[0024] Finally, based on the pre-processed driving habit data, these data are integrated to construct a driving behavior characteristic model.

[0025] Furthermore, in the system provided by the embodiment of the application, the feature model building module 11 is also used for:

[0026] Feature extraction is performed on the driving habit data to obtain steering angular velocity, acceleration, braking acceleration, driving speed and road center deviation distance at multiple time points as driving habit feature data; mean cleaning preprocessing is performed on the steering angular velocity, acceleration, braking acceleration, driving speed and road center deviation distance at multiple time points in the driving habit feature data to obtain average steering angular velocity, average acceleration, average braking acceleration, average driving speed and average road center deviation distance as preprocessed driving habit data.

[0027] In the embodiment of the present application, the driving habit data is firstly feature extracted. Specifically, when performing steering angular velocity extraction, the steering angular velocity at each moment is extracted from the collected steering data. When performing acceleration extraction, the acceleration at each moment is extracted from the acceleration data. When performing braking acceleration extraction, the driving speed at each moment is extracted from the speed data. When performing road center deviation distance extraction, the road center deviation distance at each moment, that is, the degree of deviation between the vehicle and the center line of the road, is extracted from the road driving position data. The extracted data is used as driving habit feature data.

[0028] Subsequently, the steering angular velocity, acceleration, braking acceleration, driving speed and road center deviation distance at multiple moments in the driving habit characteristic data are averaged and cleaned for preprocessing. That is, the steering angular velocity, acceleration, braking acceleration, driving speed and road center deviation distance at multiple moments are averaged to obtain the average steering angular velocity, average acceleration, average braking acceleration, average driving speed and average road center deviation distance, and the calculated average value is used as the preprocessed driving habit data.

[0029] The feature comparison module 12 compares the driving behavior feature model with the safe driving standard model to obtain risk behavior feature data.

[0030] In an embodiment of the present application, the feature comparison module compares the driving behavior feature model obtained above with the pre-built safe driving standard model, determines the difference between the driving habit data of the trainee and the standard driving data, and obtains the risk behavior feature data.

[0031] Furthermore, in the system provided in the embodiment of the application, the feature comparison module 12 is also used for:

[0032] Acquire standard driving data and construct a safe driving standard model; compare the driving behavior characteristic model with the safe driving standard model, calculate and obtain the steering angular velocity difference, acceleration difference, braking acceleration difference, driving speed difference and road center deviation distance difference as risk behavior characteristic data.

[0033] In the embodiment of the present application, standard driving data are first obtained from a preset database, including standard steering angular velocity, standard acceleration, standard braking acceleration, standard driving speed and standard road center deviation distance. Next, a safe driving standard model is constructed based on the obtained standard driving data.

[0034] The driving behavior characteristic model is then compared with the safe driving standard model, that is, the preprocessed driving habit data in the driving behavior characteristic model is compared with the standard driving data in the safe driving standard model, and the difference between the average steering angular velocity and the standard steering angular velocity is calculated, the difference between the average acceleration and the standard acceleration is calculated, the difference between the average braking acceleration and the standard braking acceleration is calculated, the difference between the average driving speed and the standard driving speed is calculated, and the difference between the average road center deviation distance and the standard road center deviation distance is calculated. The calculated steering angular velocity difference, acceleration difference, braking acceleration difference, driving speed difference and road center deviation distance difference are used as risk behavior characteristic data.

[0035] Furthermore, in the system provided in the embodiment of the application, the feature comparison module 12 is also used for:

[0036] The standard steering angular velocity, standard acceleration, standard braking acceleration, standard driving speed and standard road center deviation distance are obtained as standard driving data; and a safe driving standard model is constructed based on the standard driving data.

[0037] In an embodiment of the present application, the standard steering angular velocity, standard acceleration, standard braking acceleration, standard driving speed and standard road center deviation distance are first obtained from a preset database, and these data are used as standard driving data.

[0038] Next, based on the standard driving data, the standard value of each driving feature is directly incorporated into the safe driving standard model. Through this process, the construction of the safe driving standard model is completed.

[0039] The risk classification module 13 classifies the risk behavior characteristic data into risk levels to obtain risk levels.

[0040] In an embodiment of the present application, the risk classification module inputs the risk behavior characteristic data into a pre-built driving habit risk classification model, processes it through the model, and obtains the risk level.

[0041] Furthermore, in the system provided in the embodiment of the application, the risk classification module 13 is also used for:

[0042] According to the driving risk record data of multiple students in historical time, a driving habit risk classification model is constructed based on machine learning; the risk behavior feature data is input into the driving habit risk classification model, and the risk level is obtained by classification output.

[0043] In the embodiment of the present application, the driving risk record data of multiple students are first obtained from the historical data, and then the driving behavior characteristic data of the students are collected, and the technical experts provide the corresponding risk level for each driving risk record data. These data and the annotated risk level together constitute the supervised training data. Then, based on these supervised training data, a machine learning method such as a support vector machine is used to train the model to learn the relationship between driving behavior characteristics and risk levels until the model converges to obtain supervised training data.

[0044] Finally, the risk behavior characteristic data is input into the driving habit risk classification model, and the classification output is used to obtain the risk level.

[0045] Furthermore, in the system provided in the embodiment of the application, the risk classification module 13 is also used for:

[0046] According to the driving risk record data of multiple students in the historical time, a sample risk behavior feature data set is collected, and the risk level is identified for each sample risk behavior feature data to obtain a sample risk level set, wherein the sample risk level includes C risk levels; the sample risk behavior feature data set and the sample risk level set are used as supervised training data, and supervised training of the driving habit risk classification model is performed based on machine learning until convergence, and a constructed driving habit risk classification model is obtained, wherein the loss function in the supervised training of the driving habit risk classification model is:

[0047]

[0048] Among them, LOSS is the loss, N is the number of sample risk behavior characteristic data, C is the number of risk level categories, y i , c is the true label of the risk behavior characteristic data of the i-th sample belonging to the j-th risk level, y i ,c is 1 or 0, Predict the probability that the i-sample risk behavior feature data belongs to the j-th risk level for the driving habit risk classification model being trained.

[0049] In the embodiment of the present application, firstly, the driving risk record data of multiple trainees in the historical time are obtained from the historical database, including the behaviors of multiple trainees in the actual driving process, including steering angular velocity, acceleration, braking acceleration, driving speed and road center deviation distance at multiple moments. Then, the average of these data is calculated respectively, and then compared with the safe driving standard model to determine the sample steering angular velocity difference, sample acceleration difference, sample braking acceleration difference, sample driving speed difference and sample road center deviation distance difference, and these data are used as the sample risk behavior feature data set.

[0050] Afterwards, technical experts identify the risk level of each sample risk behavior characteristic data to obtain a sample risk level set, where the sample risk level includes C risk levels, for example, the risk level is divided into low risk, medium risk and high risk.

[0051] Next, we use the sample risk behavior feature data set and the sample risk level set as supervised training data, and conduct supervised training of the driving habit risk classification model based on machine learning, such as support vector machine. During the training process, the relationship between driving behavior feature data and risk level is learned by repeatedly adjusting parameters. In order to optimize the training effect, a loss function is used in the training to measure the difference between the predicted risk level and the actual risk level. The loss function is Among them, LOSS is the loss, N is the number of sample risk behavior characteristic data, C is the number of risk level categories, y i, c is the true label of the risk behavior characteristic data of the i-th sample belonging to the j-th risk level, y i ,c is 1 or 0, Predict the probability that the i-sample risk behavior feature data belongs to the j-th risk level for the driving habit risk classification model being trained.

[0052] Through the above process, the model continuously adjusts its internal parameters. When the loss value calculated by the loss function is less than the preset threshold, the model is considered to have converged. At this time, the training is stopped and the driving habit risk classification model is output.

[0053] The risk warning module 14 generates risk warning information according to the risk level to warn the trainees.

[0054] In the embodiment of the present application, the risk warning module generates corresponding risk warning information according to the risk level obtained above, and warns the trainee. For example, for a low risk level, a text risk warning message is generated and a notification is given after the driving is completed. For a high risk level, a voice broadcast is performed in real time, and the risk warning message is output to warn the trainee.

[0055] Furthermore, in the system provided in the application embodiment, the risk warning module 14 is also used for:

[0056] Generate risk warning information based on the risk level, wherein different risk levels include different risk warning information; and display the risk warning information to the trainees in at least one of the forms of voice, text or image.

[0057] In the embodiment of the present application, the risk warning module generates different risk warning information according to the risk level of the trainee. Different risk levels include different prompt contents. For example, for a low risk level, the prompt information may be "Your driving behavior is good, please continue to maintain it"; for a medium risk level, the prompt information may be "Please pay attention to the speed and avoid sudden braking"; for a high risk level, a more severe prompt may be issued, such as "Please slow down immediately and stay in the lane." The content of these risk warning messages varies depending on whether the trainee's driving behavior meets the safe driving standards.

[0058] Next, these risk warning information are presented to the trainees in at least one of the following ways: voice, text, or images. For low risk levels, the vehicle display screen will be used to remind the trainees in text form after driving. For high risk levels, the risk warning information will be broadcast in real time in voice form to ensure that the trainees can hear and respond in time. In addition, the warning effect can be further enhanced through images (such as color changes or icons), so that the trainees can perceive the risk level more intuitively and take corresponding safety measures during driving.

[0059] Furthermore, the system provided in the application embodiment also includes:

[0060] A behavior database construction module records the risk behavior characteristic data and risk level of the trainee and constructs a driving behavior database.

[0061] In the embodiment of the present application, the process of constructing the driving behavior database includes recording the risk behavior characteristic data and the corresponding risk level of the trainee. Specifically, the behavior database construction module integrates and summarizes the risk behavior characteristic data and risk level of the trainee obtained in the above process to complete the construction of the driving behavior database.

[0062] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0063] This application collects the driving habit data of trainees when driving, and constructs a driving behavior characteristic model; compares the driving behavior characteristic model with the safe driving standard model to obtain risk behavior characteristic data; classifies the risk behavior characteristic data into risk levels to obtain risk levels; generates risk warning information based on the risk levels to prompt the trainees. The present invention solves the technical problems of the shortcomings of the prior art in terms of driving behavior assessment accuracy and real-time performance, by collecting the driving habit data of trainees, constructing a driving behavior characteristic model, and comparing it with the safe driving standard model to obtain risk behavior characteristic data, then classifying the risk levels, assessing driving risks, and generating corresponding risk warning information based on the assessment results, thereby achieving real-time monitoring and assessment of trainees' driving behaviors and improving the technical effects of driving safety.

[0064] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0066] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. The risk warning system for students’ driving habits is characterized by: The system comprises: A feature model building module, wherein the feature model building module collects driving habit data of the trainee during driving and builds a driving behavior feature model; A feature comparison module, wherein the feature comparison module compares the driving behavior feature model with a safe driving standard model to obtain risk behavior feature data; A risk classification module, wherein the risk classification module classifies the risk behavior characteristic data into risk levels to obtain risk levels; A risk warning module generates risk warning information according to the risk level to warn the students.

2. The risk warning system for students' driving habits according to claim 1 is characterized in that: The feature model building module is used to: Collect the student's steering data, acceleration data, braking data, speed data and road position data at multiple times while driving as driving habit data; Performing data cleaning preprocessing on the driving habit data to obtain preprocessed driving habit data; A driving behavior characteristic model is constructed based on the pre-processed driving habit data.

3. The risk warning system for students' driving habits according to claim 2 is characterized in that: The feature model building module is used to: Extracting features from the driving habit data to obtain steering angular velocity, acceleration, braking acceleration, driving speed, and road center deviation distance at multiple moments as driving habit feature data; The steering angular velocity, acceleration, braking acceleration, driving speed and road center deviation distance in the multiple moments in the driving habit characteristic data are preprocessed by mean cleaning to obtain the average steering angular velocity, average acceleration, average braking acceleration, average driving speed and average road center deviation distance as preprocessed driving habit data.

4. The risk warning system for students' driving habits according to claim 3 is characterized in that: The feature comparison module is used for: Obtain standard driving data and build a standard model for safe driving; The driving behavior characteristic model is compared with the safe driving standard model to calculate the steering angular velocity difference, acceleration difference, braking acceleration difference, driving speed difference and road center deviation distance difference as risk behavior characteristic data.

5. The risk warning system for students' driving habits according to claim 4 is characterized in that: The feature comparison module is used for: Obtaining standard steering angle velocity, standard acceleration, standard braking acceleration, standard driving speed and standard road center deviation distance as standard driving data; A safe driving standard model is constructed based on the standard driving data.

6. The risk warning system for students' driving habits according to claim 1, characterized in that: The risk classification module is used to: Based on the driving risk record data of multiple students in the historical period, a driving habit risk classification model is constructed based on machine learning; The risk behavior characteristic data is input into the driving habit risk classification model, and the classification output obtains the risk level.

7. The risk warning system for students' driving habits according to claim 6, characterized in that: The risk classification module is used to: According to the driving risk record data of multiple students in the historical period, a sample risk behavior characteristic data set is collected, and the risk level is marked for each sample risk behavior characteristic data to obtain a sample risk level set, wherein the sample risk level includes C risk levels; The sample risk behavior feature data set and the sample risk level set are used as supervised training data, and supervised training of the driving habit risk classification model is performed based on machine learning until convergence, thereby obtaining a constructed driving habit risk classification model, wherein the loss function in the supervised training of the driving habit risk classification model is: Among them, LOSS is the loss, N is the number of sample risk behavior characteristic data, C is the number of risk level categories, y i , c is the true label of the risk behavior characteristic data of the i-th sample belonging to the j-th risk level, y i ,c is 1 or 0, Py i j is the probability that the i-th sample risk behavior characteristic data belongs to the j-th risk level predicted by the driving habit risk classification model in training.

8. The risk warning system for students' driving habits according to claim 1 is characterized in that: The risk warning module is used to: Generate risk warning information according to the risk level, wherein different risk levels include different risk warning information; The risk warning information is presented to the trainees in at least one of the forms of voice, text or image.

9. The risk warning system for students' driving habits according to claim 1, characterized in that: The system further comprises: A behavior database construction module records the risk behavior characteristic data and risk level of the trainee and constructs a driving behavior database.

Images