CN-116861212-B - Heavy engineering vehicle driving performance evaluation method based on natural driving data

CN116861212BCN 116861212 BCN116861212 BCN 116861212BCN-116861212-B

Abstract

The invention discloses a heavy engineering vehicle driving performance evaluation method based on natural driving data, which comprises the steps of firstly, collecting vehicle motion data in a natural driving environment and preprocessing the data; then, corresponding statistical characteristics are extracted for driving behaviors under different road types, driving performance evaluation is conducted based on the driving behavior statistical indexes, finally, driving scores are calculated based on driving performance evaluation results, and finally, a driving performance evaluation system is obtained. The invention can be used for driver performance assessment and safety education management, enhancing driver road traffic safety consciousness and reducing road traffic accidents.

Inventors

MA YONGFENG
YU LINGHUA
ZHANG CHENXIAO
CHEN SHUYAN
LU JIAN
HU XIAOJIAN

Assignees

东南大学

Dates

Publication Date: 20260512
Application Date: 20230629

Claims (5)

1. The heavy engineering vehicle driving performance evaluation method based on the natural driving data is characterized by comprising the following steps of: step 1, acquiring vehicle motion data of a heavy engineering vehicle in a natural driving environment, and preprocessing the vehicle motion data to obtain driving behavior data; step 2, dividing the driving behavior data obtained in the step 1 into driving behavior data under urban road types and driving behavior data under rural road types, and respectively extracting statistical characteristics from the driving behavior data under different road types to obtain a heavy engineering vehicle driving performance evaluation index; and 3, based on the evaluation index obtained in the step 2, carrying out correlation analysis and principal component analysis on the evaluation index, clustering the obtained principal components, and establishing a driving performance evaluation model according to a clustering result, wherein the specific process is as follows: Step 3.1, carrying out pearson correlation analysis on the heavy engineering vehicle driving performance evaluation index obtained in the step 2, carrying out KMO (KMO) test and Bartlett test, and carrying out dimension reduction on the evaluation index by a principal component analysis method; Step 3.2, the principal components obtained through the principal component analysis method are clustered into three categories by adopting a K-means clustering method, characteristic parameters of central points of the three categories are compared, the category with the highest characteristic parameters is marked as aggressive driving performance, the category with the lowest characteristic parameters is marked as cautious driving performance, and the category with the characteristic parameters at the middle level is marked as normal driving performance; step 4, calculating a driving score according to the driving performance evaluation model established in the step 3, and constructing a heavy engineering vehicle driving performance evaluation system; In the step 4, three types of driving performances are digitally represented by 123, the driving performance score is 1 when the driving performance is aggressive, the driving performance score is 2 when the driving performance is normal, the driving performance score is 3 when the driving performance is careful, and the driving gain rate is calculated: , Wherein, the Representing the driver In driving section Is provided with a driving performance score of (a), Representing the driver In driving section Is provided with a driving performance score of (a), Representing the driver In driving section Representing fluctuations in driving performance variation of the driver during continuous running; calculating a driving volatility score based on driving gain rate I.e. the sample standard deviation of the driving gain rate of the driver over a journey: , Wherein, the An average value of the driving gain rate is represented, Indicating the total estimated driving fragment number of the driver over a journey.
2. The method for evaluating the driving performance of the heavy engineering vehicle based on the natural driving data according to claim 1, wherein in the step 1, the vehicle motion data comprises the following variables of satellite number, UTC time, longitude and latitude, speed, course angle, height, vertical speed, transverse acceleration and longitudinal acceleration, the acquired vehicle motion data is placed in a matrix according to time sequence, one column of the matrix corresponds to one variable, one row of the matrix is all variables obtained by acquiring one-time vehicle motion data, and the vehicle motion data acquisition frequency is 10Hz; Preprocessing vehicle motion data to obtain driving behavior data, wherein the driving behavior data comprises the following specific steps: 1.1, performing invalid value processing on vehicle motion data, deleting invalid column variables and invalid sample rows, defining the column variables with empty data as invalid column variables, and defining the sample rows with empty data, sample rows with satellite numbers less than 5 and sample rows with invalid longitude and latitude, speed and height of 0; 1.2, deleting sample rows with the speed, the course angle, the vertical speed, the transverse acceleration and the longitudinal acceleration being 0 at the same time on the basis of 1.1, filling the missing data by using a moving average method, and filling an average value of 10 values before and after the column of the missing data; 1.3, converting the vehicle motion data with the acquisition frequency of 10Hz into 1Hz, namely taking the average value of 10 data acquired in the current second as the statistic of the current second.
3. The method for evaluating driving performance of heavy engineering vehicle based on natural driving data according to claim 1, wherein in the step 2, the range of the rural road is defined on the topographic map, the driving behavior data with the longitude and latitude within the defined range is marked as the driving behavior data under the rural road type, and the driving behavior data with the longitude and latitude outside the defined range is marked as the driving behavior data under the urban road type; For driving behavior data under each road type, dividing the driving behavior data into a plurality of driving fragments by taking 1s as a step length and 60s as a sliding time window length, and taking a type corresponding to a part with longer driving time as the type of the driving fragment when the driving behavior data under the urban road type and the rural road type exist in a certain driving fragment at the same time; And respectively extracting statistical characteristics of longitudinal driving behavior and transverse driving behavior data under different road types, wherein the longitudinal driving behavior comprises speed and longitudinal acceleration, and the transverse driving behavior comprises transverse acceleration, so that a heavy engineering vehicle driving performance evaluation index is obtained, and the evaluation index comprises average absolute deviation of speed, variation coefficient of speed, quartile divergence coefficient of speed, maximum value of longitudinal acceleration, minimum value of longitudinal acceleration, average value of longitudinal acceleration, standard deviation of longitudinal acceleration, average absolute deviation of longitudinal acceleration and average absolute deviation of transverse acceleration.
4. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the steps of the heavy engineering vehicle driving performance assessment method based on natural driving data as claimed in any one of claims 1 to 3.
5. A computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the heavy engineering vehicle driving performance assessment method based on natural driving data as claimed in any one of claims 1 to 3.

Description

Heavy engineering vehicle driving performance evaluation method based on natural driving data Technical Field The invention relates to a heavy engineering vehicle driving performance evaluation method based on natural driving data, and belongs to the technical field of driving performance evaluation. Background Along with the acceleration of the urban process in China, heavy engineering transport vehicles including concrete mixing transport vehicles, sand transporting vehicles, dregs and soil transporting vehicles and the like become important carriers for material transportation between construction sites and resource sites. The heavy engineering vehicle has high heavy center position, large cargo capacity, difficult monitoring of driving behavior and high driving risk. The driving performance is an important index for measuring the driving safety, and the evaluation of the driving performance of the driver of the heavy engineering vehicle is helpful for the management of the fleet, so that safer and more economical driving conditions are brought to the driving fleet. At present, the research on driving behaviors has achieved abundant results, but the defect still exists that the related research on the national driving performance evaluation is still in an initial stage. Most of the existing methods for evaluating the driving performance of engineering vehicle drivers are simple weighted scoring, but the comprehensive evaluation model of the driving performance at home and abroad cannot be directly applied to domestic heavy engineering vehicles due to the differences in various aspects such as home and abroad conditions, vehicle types, driver group characteristics and the like. Disclosure of Invention The technical problem to be solved by the invention is to provide a heavy engineering vehicle driving performance evaluation method based on natural driving data, which is provided for heavy engineering vehicle drivers by combining with natural driving environment. The invention adopts the following technical scheme for solving the technical problems: a heavy engineering vehicle driving performance evaluation method based on natural driving data comprises the following steps: step 1, acquiring vehicle motion data of a heavy engineering vehicle in a natural driving environment, and preprocessing the vehicle motion data to obtain driving behavior data; step 2, dividing the driving behavior data obtained in the step 1 into driving behavior data under urban road types and driving behavior data under rural road types, and respectively extracting statistical characteristics from the driving behavior data under different road types to obtain a heavy engineering vehicle driving performance evaluation index; Step 3, based on the evaluation index obtained in the step 2, carrying out correlation analysis and principal component analysis on the evaluation index successively, clustering the obtained principal components, and establishing a driving performance evaluation model according to a clustering result; And 4, calculating a driving score according to the driving performance evaluation model established in the step 3, and constructing a heavy engineering vehicle driving performance evaluation system. In the step 1, the vehicle motion data comprises the following variables of satellite number, UTC time, longitude and latitude, speed, course angle, altitude, vertical speed, transverse acceleration and longitudinal acceleration, wherein the acquired vehicle motion data are placed in a matrix according to the time sequence, one column of the matrix corresponds to one variable, one row of the matrix is all variables obtained by acquiring one-time vehicle motion data, and the acquisition frequency of the vehicle motion data is 10Hz; Preprocessing vehicle motion data to obtain driving behavior data, wherein the driving behavior data comprises the following specific steps: 1.1, carrying out invalid value processing on vehicle motion data, deleting invalid column variables and invalid sample rows, defining the column variables with empty data as invalid column variables, and defining the sample rows with empty data, sample rows with satellite numbers less than 5 and sample rows with invalid longitude and latitude, speed and height of 0; 1.2 on the basis of 1.1, deleting sample rows with the speed, the course angle, the vertical speed, the transverse acceleration and the longitudinal acceleration being 0 at the same time, filling the missing data by using a moving average method, and filling an average value of 10 values before and after the column of the missing data; 1.3 converting vehicle motion data with the acquisition frequency of 10Hz into 1Hz, namely taking the average value of 10 data acquired in the current second as the statistic of the current second. In the step 2, the range of the rural road is defined on the topographic map, the driving behavior data with the longitude and latitude within the defined range is m