Driver road driving ability evaluation method based on big data

Abstract

The invention discloses a big data-based driver road driving ability evaluation method. The method comprises the steps of performing multi-dimensional acquisition of driving behavior data of coaches and trainees; setting a standard event threshold value for identifying and distinguishing normal behaviors and abnormal behaviors; in the interaction process of the coach and the trainee, screening indexes which have significant differences between the coach and the trainee and are superior to the trainee, and carrying out quantitative scoring on the process behavior indexes through a standard event threshold value; determining an optimal weight by taking the coach score superior to the trainee score as an optimization target; and after the optimal weight of each behavior index is determined, the total score of each item is calculated in a weighted manner, so that the road driving ability of the driver can be evaluated. According to the invention, through a scientific, objective and comprehensive evaluation system, an effective tool is provided for driving training, the driving skill and safety awareness of trainees are improved, the quality of training and examination is improved, and finally the purpose of improving road traffic safety is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of motor vehicle driver training and examination, and in particular to a method for evaluating a driver's road driving ability based on big data. Background Art

[0002] In terms of the road driving skills test for drivers, countries generally adopt result-oriented evaluation indicators, such as turn signal (failure to turn off the turn signal after turning), line pressing (wheels rolling over the edge of the road), parking (stopping in the middle of the road), driving route (not driving according to the prescribed route), speed (insufficient duration of maximum speed), gear matching (failure to smoothly increase or decrease gears according to instructions), braking (failure to brake to slow down), etc. In terms of the score evaluation of the subject three test, different points are usually deducted according to the degree of mistakes made by the trainees in each project, and the scores are given according to the standards, ranging from 5 points to 100 points. For example, 5 points will be deducted if the vehicle moves when starting, and 100 points will be deducted if the direction control is unstable. However, less attention is paid to the specific progress in the training process, and there is a lack of scientific and quantitative training effect evaluation based on driving behavior data. The result-oriented evaluation method mainly focuses on the final results of the test, while ignoring the specific progress in the training process and the continuous improvement of driving behavior. This evaluation method may not fully reflect the driver's skill improvement and behavior changes during the training process.

[0003] Therefore, how to provide a method for evaluating a driver's road driving ability based on big data has become a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method for evaluating a driver's road driving ability based on big data. Through a scientific, objective and comprehensive evaluation system, an effective tool is provided for driving training, which helps to improve students' driving skills and safety awareness, improve the quality of training and examinations, and ultimately achieve the goal of improving road traffic safety.

[0005] The present invention solves the technical problem by adopting the following technical solution:

[0006] A method for evaluating a driver's road driving ability based on big data comprises the following steps:

[0007] S1, obtain training behavior data, and collect multi-dimensional driving behavior data of coaches and students;

[0008] S2, determine the standard event threshold, set the standard event threshold to identify and distinguish normal behavior from abnormal behavior;

[0009] S3, determine the behavioral indicator scores, screen the indicators in which the coach has significant differences with the trainee and is better than the trainee during the interaction between the coach and the trainee, and quantify the scores of these process behavioral indicators through standard event thresholds;

[0010] S4, determine the optimal indicator weight, take the coach score better than the student score as the optimization goal, and determine the optimal weight;

[0011] S5, calculate the project scores. After determining the optimal weights of various behavioral indicators, the total score of each project is weighted to achieve the driver's road driving ability assessment.

[0012] Furthermore, in step S2, one or more standard event thresholds are determined by analyzing the distribution of the driving behavior data, and the data are divided into different intervals according to the standard event thresholds.

[0013] Furthermore, the behavioral indicators are evaluated from three dimensions: security, stability, and efficiency.

[0014] Furthermore, the safety includes the maximum steering wheel speed, the number of passive braking times, the speed variation coefficient, the number of sudden accelerations, the number of sudden decelerations, the failure to turn on the turn signal, and the insufficient turn signal duration index.

[0015] Furthermore, the stability includes training trajectory score, acceleration smoothness, starting deceleration position, maximum absolute value of steering wheel angle in the first 100 meters, number of positive and negative changes in steering wheel angle, and speed and gear mismatch duration indicators.

[0016] Furthermore, the efficiency includes an average speed index.

[0017] Furthermore, the calculation method of each behavioral indicator score is as follows:

[0018] Speed ​​variation coefficient: It is an important indicator to measure the stability of vehicle speed changes, which is the ratio of the standard deviation of speed to the average speed;

[0019] Number of positive and negative changes in the steering wheel angle: The smaller value of the steering wheel angle [-10°, 10°] is regarded as 0, and the steering wheel angle is counted every 2 seconds to see whether it has a positive or negative change. If yes, it is counted as 1 time, and the total number of times in all 2-second time windows is counted;

[0020] Acceleration and lateral acceleration smoothness: Acceleration is calculated by Fourier transform and the frequency is obtained. The energy of the frequency component is calculated and the cutoff frequency is defined as 0.05. The total energy and low-frequency energy are calculated. The low-frequency energy ratio is the acceleration smoothness.

[0021] Start deceleration position: Calculate the cumulative distance based on the speed. Continuous deceleration within 0.4s is considered deceleration. Calculate the position where deceleration first occurs. If no deceleration position is found, it is considered as no deceleration.

[0022] Training track score: round the x coordinate to an integer, determine whether the y coordinate is within the selected distance range, and calculate the percentage of points within the range, which is the training track score.

[0023] The present invention discloses a method for evaluating a driver's road driving ability based on big data, which has the following features:

[0024] Beneficial effects:

[0025] 1) Based on the multi-dimensional collection of the driver's driving behavior data, the present invention constructs a set of process evaluation index systems based on data-driven, evaluates the students' driving behavior from three dimensions: safety, stability and efficiency, and constructs evaluation indicators such as the maximum steering wheel speed, training trajectory score, average speed, etc., which comprehensively reflect the students' driving skills and safety level, provide students with instant feedback and personalized improvement suggestions, and help students adjust their driving behavior in time and improve learning efficiency. This method can provide personalized training suggestions for each driver based on the evaluation results to help them improve their driving skills. At the same time, by identifying bad driving behavior, this method helps to improve road safety and reduce the occurrence of traffic accidents.

[0026] 2) Based on the multi-dimensional collection of driver training process data, the present invention constructs a set of standard event threshold evaluation methods. By determining the index threshold, the items are scored according to the threshold, so as to obtain the scores of each index of the item and determine the comprehensive scores of each item in subject three. Through this evaluation method, the student's performance on each evaluation index can be comprehensively considered to provide a comprehensive driving ability score that more fairly reflects the student's actual driving level. The evaluation method not only focuses on passing the test, but also focuses on cultivating students' good driving habits and safety awareness, which helps students maintain safe driving behavior for a long time after obtaining a driver's license. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of the method of the present invention;

[0028] Figure 2 This is a schematic diagram of the initial deceleration position characteristics of the instructor and the student;

[0029] Figure 3 It is a schematic diagram of the trajectory characteristics of the coach and the student of the present invention;

[0030] Figure 4 Schematic diagram of the characteristics of the similarity index of the trajectories of the coach and the trainee of the present invention. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] refer to Figure 1 The present invention discloses a method for evaluating a driver's road driving ability based on big data, comprising the following steps:

[0033] S1, obtain training behavior data, and collect driving behavior data of coaches and students in multiple dimensions; the input data of the algorithm includes the number of sudden accelerations and decelerations of different projects of coaches and students, training trajectories, acceleration smoothness, longitudinal acceleration smoothness, lateral acceleration smoothness, vehicle speed data, speed and gear mismatch duration, number of sudden lane changes, and number of positive and negative changes in steering wheel angles, etc. First, convert the driving behavior time series data into distance coordinate data with a granularity of 1 meter.

[0034] S2, determine the standard event threshold. In data analysis and behavior assessment, setting the standard event threshold is a key method for identifying and distinguishing normal behavior from abnormal behavior. One or more standard event thresholds are determined by analyzing the distribution of data. The standard event threshold divides the data into different intervals. Once a data point exceeds these preset limits, it will be given different scores based on its degree of deviation.

[0035] S3, determine the score of behavioral indicators, screen the indicators that are significantly different from the trainees and better than the trainees during the interaction between the coach and the trainee, and quantify and score these process behavioral indicators through standard event thresholds to more accurately evaluate the coach's teaching effect and the trainee's learning progress. The scoring system can be based on the differences between the coach and the trainee, and whether the coach's performance meets the preset standards.

[0036] S4, determine the optimal indicator weight. In order to optimize the coach's teaching effect, it is necessary to determine the key indicators and assign weights accordingly to reflect the importance of the indicator in the overall teaching effect. The coach's score is better than the student's score and the optimal weight is determined.

[0037] S5, calculate the project scores. After determining the optimal weights of each behavioral indicator, calculate the total score of each project in a weighted manner to reflect its impact on the project. According to the distribution of scores, determine the excellent, good, medium and poor intervals, thereby providing a basis for improving teaching methods and improving learning efficiency.

[0038] The behavioral indicators are evaluated from three dimensions: safety, stability, and efficiency. The safety includes the maximum steering wheel speed, the number of passive brakes, the speed variation coefficient, the number of sudden accelerations, the number of sudden decelerations, the failure to turn on the turn signal, and the insufficient duration of the turn signal. The stability includes the training trajectory score, the acceleration smoothness, the starting deceleration position, the maximum absolute value of the steering wheel angle in the first 100 meters, the number of positive and negative changes in the steering wheel angle, and the duration of speed and gear mismatch. The efficiency includes the average speed indicator. The evaluation indicators of each project are shown in Appendix 1-9. The number of bad driving behaviors is also an important indicator for evaluating driving behavior in Subject III.

[0039] Among them, Table 1 is the straight-line driving evaluation index system.

[0040]

[0041] Table 1 and Table 2 are the evaluation index system for shifting gears and parking within 100 meters.

[0042]

[0043] Table 2 and Table 3 show the lane change evaluation index system.

[0044]

[0045] Table 3 and Table 4 show the evaluation index system for passing the pedestrian crossing.

[0046]

[0047] Table 4 and Table 5 show the evaluation index system for meeting practice.

[0048]

[0049] Table 5 and Table 6 show the overtaking practice evaluation index system.

[0050]

[0051] Table 6 and Table 7 show the evaluation index system for U-turn practice.

[0052]

[0053] Tables 7 and 8 show the evaluation index system for pull-over parking.

[0054]

[0055] Table 8

[0056] Table 9 shows the evaluation index system for left turns at intersections.

[0057]

[0058] Table 9

[0059] To further optimize the technical solution, the calculation method for each behavioral indicator score is as follows:

[0060] Speed ​​variation coefficient: It is an important indicator to measure the stability of vehicle speed changes, which is the ratio of the standard deviation of speed to the average speed;

[0061] Number of positive and negative changes in the steering wheel angle: The smaller value of the steering wheel angle [-10°, 10°] is regarded as 0, and the steering wheel angle is counted every 2 seconds to see whether it has a positive or negative change. If yes, it is counted as 1 time, and the total number of times in all 2-second time windows is counted;

[0062] Acceleration and lateral acceleration smoothness: Acceleration is calculated by Fourier transform and the frequency is obtained. The energy of the frequency component (the square of the amplitude) is calculated, and the cutoff frequency is defined as 0.05. The total energy and low-frequency energy are calculated, and the low-frequency energy ratio is the acceleration smoothness.

[0063] Start deceleration position: Calculate the cumulative distance based on the speed. Continuous deceleration within 0.4s is considered deceleration. Calculate the position where deceleration first occurs. If no deceleration position is found, it is considered as no deceleration.

[0064] Training track score: round the x coordinate to an integer, determine whether the y coordinate is within the selected distance range, and calculate the percentage of points within the range, which is the training track score.

[0065] The present invention is a set of road driving behavior evaluation methods developed based on the subject three road driving skills evaluation test. By collecting vehicle operation and driving behavior data, setting indicator thresholds and determining optimal weights according to the distribution of each indicator, the trainee's driving ability and operating level are evaluated from the dimensions of safety, stability and efficiency. In the road driving skills test training, it is divided into 9 sub-items such as straight-line driving, 100-meter gear changes, lane changes, passing sidewalks, meeting exercises, overtaking exercises, U-turn exercises, pull-over parking, and left turns at intersections, involving a total of 74 indicators. Each sub-item covers multiple small items, each small item involves a total of multiple evaluation items, and each evaluation item covers multiple driving behavior indicators. The present invention aims to construct a set of process-based road driving skills evaluation index system based on the driver's control behavior data collected from the actual vehicle, and develop a driver's road driving behavior evaluation method based on big data based on the constructed index system. In terms of test training, a process-based road driving skill evaluation index system is adopted to comprehensively evaluate the driver's driving operation skills during the training process from the three dimensions of safety, stability and efficiency. The evaluation process of determining the optimal weight, setting the index threshold and determining the weighted total score comprehensively, efficiently and comprehensively reflects the driver's comprehensive driving skills and operation level, better captures every detail and behavior of the driver in the process, and provides more comprehensive, real-time and personalized feedback, which helps the driver to continuously improve and enhance his skills.

[0066] The present invention provides an effective tool for driving training through a scientific, objective and comprehensive evaluation system, which helps to improve students' driving skills and safety awareness, improve the quality of training and examinations, and ultimately achieve the goal of improving road traffic safety.

[0067] Example

[0068] Take the evaluation index of the starting deceleration position in the pedestrian crossing project as an example. Figure 2 , Figure 3 and Figure 4 As shown. The cumulative distance is calculated based on the speed. Continuous deceleration within 0.4s is considered as deceleration, and the location where the first deceleration occurs is calculated; if the deceleration location is not found, it is considered as no deceleration. The data set is divided into meters. After removing outliers, the data set range is 0-100m, as follows:

[0069] 1) Feature analysis and indicator determination: Describe the driving behavior feature distribution of coaches and students, and screen indicators that are significantly different from those of coaches and students and better than those of students.

[0070] 2) Determine the threshold: Set the indicator threshold according to the distribution, and give different scores to different metrics when they exceed the limit.

[0071] The initial deceleration position is -1m-6.5m, excellent, 100

[0072] The initial deceleration position is 6.5m-11.8m, good, 70

[0073] The first deceleration position is between 11.8m and 15.6m, medium, 50

[0074] The first deceleration position is beyond 15.6m, difference, 0

[0075] 3) Weight calculation: Calculate the weight of each indicator from the dimensions of safety, stability and efficiency. The weight calculation of each indicator of the pedestrian crossing project is shown in Table 10:

[0076]

[0077] Table 10

[0078] 4) Comprehensive score calculation: Weighted total score is calculated, and the excellent, good, fair, and poor intervals are determined based on the distribution. The total score of the project is calculated according to the following formula:

[0079] Total project score = Σ indicator score * weight.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating a driver's road driving ability based on big data, characterized in that: Includes the following steps: S1, obtain training behavior data, and collect multi-dimensional driving behavior data of coaches and students; S2, determine the standard event threshold, set the standard event threshold to identify and distinguish normal behavior from abnormal behavior; S3, determine the behavioral indicator scores, screen the indicators in which the coach has significant differences with the trainee and is better than the trainee during the interaction between the coach and the trainee, and quantify the scores of these process behavioral indicators through standard event thresholds; S4, determine the optimal indicator weight, take the coach score better than the student score as the optimization goal, and determine the optimal weight; S5, calculate the project scores. After determining the optimal weights of various behavioral indicators, the total score of each project is weighted to achieve the driver's road driving ability assessment.

2. The method for evaluating a driver's road driving ability based on big data according to claim 1, characterized in that: In step S2, one or more standard event thresholds are determined by analyzing the distribution of the driving behavior data, and the data are divided into different intervals according to the standard event thresholds.

3. The method for evaluating a driver's road driving ability based on big data according to claim 2 is characterized in that: Behavioral indicators are evaluated from three dimensions: security, stability and efficiency.

4. The method for evaluating a driver's road driving ability based on big data according to claim 3 is characterized in that: The safety includes the maximum steering wheel speed, number of passive braking times, speed variation coefficient, number of sudden accelerations, number of sudden decelerations, indicators of not turning on the turn signal and insufficient turn signal duration.

5. The method for evaluating a driver's road driving ability based on big data according to claim 4 is characterized in that: The stability includes training track score, acceleration smoothness, starting deceleration position, maximum absolute value of steering wheel angle in the first 100 meters, number of positive and negative changes in steering wheel angle, and speed and gear mismatch duration indicators.

6. The method for evaluating a driver's road driving ability based on big data according to claim 5 is characterized in that: The efficiency includes an average speed indicator.

7. The method for evaluating a driver's road driving ability based on big data according to claim 6 is characterized in that: The scores for each behavioral indicator are calculated as follows: Speed ​​variation coefficient: It is an important indicator to measure the stability of vehicle speed changes, which is the ratio of the standard deviation of speed to the average speed; Number of positive and negative changes in the steering wheel angle: The smaller value of the steering wheel angle [-10°, 10°] is regarded as 0, and the steering wheel angle is counted every 2 seconds to see whether it has a positive or negative change. If yes, it is counted as 1 time, and the total number of times in all 2-second time windows is counted; Acceleration and lateral acceleration smoothness: Acceleration is calculated by Fourier transform and the frequency is obtained. The energy of the frequency component is calculated and the cutoff frequency is defined as 0.

05. The total energy and low-frequency energy are calculated. The low-frequency energy ratio is the acceleration smoothness. Start deceleration position: Calculate the cumulative distance based on the speed. Continuous deceleration within 0.4s is considered deceleration. Calculate the position where deceleration first occurs. If no deceleration position is found, it is considered as no deceleration; Training track score: round the x coordinate to an integer, determine whether the y coordinate is within the selected distance range, and calculate the percentage of points within the range, which is the training track score.

Images