CN110544373B - Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles - Google Patents
Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles Download PDFInfo
- Publication number
- CN110544373B CN110544373B CN201910773932.2A CN201910773932A CN110544373B CN 110544373 B CN110544373 B CN 110544373B CN 201910773932 A CN201910773932 A CN 201910773932A CN 110544373 B CN110544373 B CN 110544373B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- early warning
- time
- mileage
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/20—Monitoring the location of vehicles belonging to a group, e.g. fleet of vehicles, countable or determined number of vehicles
- G08G1/205—Indicating the location of the monitored vehicles as destination, e.g. accidents, stolen, rental
Abstract
The invention relates to a method for extracting early warning information of a truck and identifying risks based on Beidou Internet of vehicles, which comprises the following steps: step 1, obtaining original data related to vehicle early warning through a vehicle-mounted terminal of a vehicle networking system with a Beidou positioning system, step 2, preprocessing the original data, step 3, extracting key variables of early warning information of a truck, step 4, clustering vehicle safety risks, step 5, judging and analyzing, step 6, and identifying risks. The invention can give the early warning frequency of unit travel mileage and the early warning frequency of unit travel time to a certain vehicle/driver, and can give the judgment function of discriminant analysis based on the historical data of the Internet of vehicles to realize the identification or prediction of the risk.
Description
Technical Field
The invention belongs to the field of traffic safety application, is applied to the traffic transportation industry, and particularly relates to a truck early warning information extraction and risk identification method based on the Beidou Internet of vehicles.
Background
The traffic and transportation department of 12 months 4 and 2019 publishes a statistics bulletin for the development of the traffic and transportation industry of 2018, and data shows that China in 2018 completes 395.69 hundred million tons of road freight transportation all the year round, the road freight transportation amount is increased by 7.3%, and the road freight turnover amount is 71249.21 hundred million tons of kilometers, and the road freight transportation amount is increased by 6.7%. In recent years, the road freight volume and the freight turnover volume of China still gradually rise, but the problem of road traffic safety is increasingly highlighted, and the road traffic safety accident is still a great hidden danger in the development of the traffic industry of China. According to the statistics of the department of transportation and management of the ministry of public security, 1.94 hundred million cars are kept in the country at the end of 2016, wherein 1351.77 ten thousand cars of a cargo vehicle (for short, a truck) account for 7.0 percent of the total number of the cars; 5.04 thousands of truck responsibility road traffic accidents occur in 2016, which account for 30.5% of the total amount of the truck responsibility accidents in the country, and the proportion is far higher than the proportion of the reserved amount of the trucks to the total amount of the automobiles. Therefore, the traffic safety problem of trucks is particularly important, and the related research on how to reduce truck traffic accidents is urgent.
The traditional road traffic safety research mostly takes traffic accident data as a basis, an accident frequency or accident severity model is constructed, and traffic accident characteristics and key influence factors thereof are disclosed from four aspects of people, vehicles, roads and environment. With the rapid development of the internet and computer technology, traffic data acquisition means are more diversified and intelligent, and the application of the car networking technology in the aspect of traffic safety is correspondingly developed. Most of the existing traffic safety research based on the Internet of vehicles relates to a model method, a device and a system for vehicle early warning. For example, patent CN109615879A is a vehicle speed abnormality early warning model based on the internet of vehicles, patent CN109584630A is a vehicle lane change early warning device and early warning method based on the internet of vehicles, CN109147279A is a fatigue driving monitoring early warning method and system based on the internet of vehicles, and patents CN105869439B, CN108986544A, CN109559559A and CN109584631A are anti-collision early warning methods and systems based on the internet of vehicles. However, the existing literature is rarely related to how to improve road traffic safety by using early warning information of truck-mounted devices based on the internet of vehicles.
At present, the application prospect of the internet of vehicles technology in the traffic field is wide, the transportation department stipulates that intercity freight vehicles must be networked, and the operation of trucks is monitored and controlled in real time through the internet of vehicles, so that the road traffic safety is improved. At present, a plurality of provincial and municipal freight vehicles use a Beidou Internet of vehicles system and require trucks to be provided with corresponding safety early warning devices. The Beidou Internet of vehicles system constructs an Internet of vehicles system with vehicles as nodes by adopting advanced Beidou positioning navigation, sensing, control and other technologies, monitors and stores the running track, speed, time, mileage, alarm and early warning information in the running state and the like of the vehicles in real time, wherein the vehicle early warning information mainly comprises overspeed early warning, fatigue driving early warning, collision early warning, rollover early warning and the like. However, there still is no relevant technology for extracting and utilizing the early warning information to improve the road traffic safety. The early warning is frequently generated before the vehicle breaks rules and regulations, and indicates that certain potential risks exist, and the early warning is used for prompting a driver to standardize the driving behavior of the driver and reducing the occurrence of traffic accidents and violation events. The traffic accident or the violation event of the driver belongs to deterministic dangerous behaviors, the occurrence probability is relatively low, and the randomness whether the event occurs is strong; the early warning belongs to potential dangerous behaviors, the occurrence probability is higher, the data volume is larger, and the early warning information can comprehensively and deeply reflect the driving risk of a driver. Therefore, the invention provides a truck early warning information extraction and risk identification method based on Beidou Internet of vehicles, which is characterized in that original data related to vehicle early warning is obtained based on the Internet of vehicles technology, the data is processed, key variables of the early warning information are extracted, and on the basis, clustering and discriminant analysis are carried out on the risk degree of vehicles/drivers, so that high-risk vehicles/drivers can be accurately identified, the drivers can be supervised to develop good driving habits, and the road traffic safety is improved.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a truck early warning information extraction and risk identification method based on the Beidou Internet of vehicles.
In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:
a method for extracting early warning information of a truck and identifying risks based on Beidou Internet of vehicles comprises the following steps:
step 1, obtaining original data related to vehicle early warning through a vehicle-mounted terminal of a vehicle networking provided with a Beidou positioning system, wherein the original data comprises: mileage information, safety early warning information and state data, wherein the state data comprises vehicle ID, ACC state, uploading time and the like;
Preprocessing and screening original data by using a Python programming technology; before analyzing the original data, the data needs to be cleaned and sorted, so that the data quality is improved; the data cleaning comprises the following steps: filling missing values in the data, and identifying abnormal values and redundant data in the data; the method mainly comprises the following steps of preprocessing original data by combining the characteristics of data stored by a vehicle-mounted terminal, wherein the missing value is mainly expressed as a lack attribute value, and the abnormal value is mainly expressed as an overlarge or undersize value of a single attribute:
step 2.1, data missing value operation: the method comprises the steps of performing programming operation by using Python, introducing an os module and a numpy module in the Python, defining a required function, executing a main function, operating a text file stored in a vehicle-mounted terminal, deleting the text file lacking an attribute value, and ensuring the integrity of the attribute;
step 2.2, data abnormal value operation: data outliers include: the mileage is abnormal, the early warning state is abnormal, and the uploading time is abnormal;
1) and (3) mileage exception handling: firstly, traversing all data files, calculating the mileage of the vehicle on the day, secondly, making an accumulated distribution map of the mileage of the vehicle on the day, and determining an overlarge value point and an undersize value point of the mileage of the vehicle on the day; finally, removing travel records with overlarge or undersize numerical values in the driving mileage of the vehicle on the day;
2) and (4) abnormal processing of the early warning state: counting the early warning duration of each early warning bit, solving the duration of single early warning of the early warning bit, and deleting the early warning state with obvious errors;
3) and (3) processing an uploading time exception: calculating the time difference between adjacent uploading points, and eliminating the recording points with the difference value of the adjacent uploading time unchanged or less than zero;
step 2.3, redundant data operation: traversing all records of the vehicle trip on the same day, and deleting the repeatedly uploaded records and the records with smaller data scale on the same day, wherein the records specifically comprise: comparing the uploaded records, deleting the repeatedly uploaded records, and repeatedly executing until all data files are traversed; counting the trip time of the vehicle on the day according to the record with smaller data scale, and deleting the trip record less than 15 min;
step 3, extracting key variables of vehicle early warning information
According to historical travel data of the vehicle, two key variables in the vehicle running early warning information are extracted: the early warning frequency of the unit driving mileage of the vehicle and the early warning frequency of the unit driving time of the vehicle; firstly, counting the total early warning frequency of a specific early warning position of each vehicle in a time period of T days, wherein T is a positive integer; secondly, counting the total driving mileage of each vehicle within T days; thirdly, counting the total driving time of each vehicle within the time period T days; then, calculating the early warning frequency of each vehicle in unit driving mileage and the early warning frequency of each vehicle in unit driving time; taking a vehicle ID as a unique identification code, and counting and summarizing travel record information of the same ID vehicle in different time periods; the method comprises the following specific steps:
step 3.1, counting the early warning frequency of each vehicle at a specific early warning position in a time period T days
Taking travel early warning records of vehicles as objects, firstly taking a vehicle ID as a unique identification code, counting the early warning frequency of each early warning position of each vehicle in one day, then accumulating selected specific early warning positions to obtain the early warning total frequency of each vehicle in the specific early warning position in one day, and finally accumulating the early warning frequency of the same ID vehicle in the specific early warning position every day in a time period T to obtain the early warning total frequency of each vehicle in the specific early warning position in the time period T;
step 3.2, counting the total driving mileage of each vehicle in the period T
The driving mileage is to record the mileage change of the instrument panel of the vehicle and reflect the driving distance of the vehicle; accumulating the driving mileage of the same ID vehicle by taking the vehicle ID number as a unique identification code, and finally obtaining the total driving mileage of each vehicle in a time period T days;
step 3.3, counting the total running time of each vehicle in the time period T days
The vehicle running time does not include the time lost by the vehicle due to waiting or delay, and the principle of vehicle running time extraction is as follows: firstly, calculating the total travel time of the vehicle, and then calculating the parking time, wherein the time difference between the total travel time and the parking time is the total travel time of the vehicle; then, accumulating the running time of the same ID vehicle in the time period T by taking the vehicle ID as a unique identification code to obtain the total running time of each vehicle in the time period T;
step 3.4, dividing the early warning frequency of each vehicle at the specific early warning position in the time period T day obtained in the step 3.1 and the total driving range of each vehicle in the time period T day obtained in the step 3.2 to obtain the early warning frequency of the unit driving range of the vehicle; dividing the early warning frequency of each vehicle at the specific early warning position in the time period T day obtained in the step 3.1 and the total running time of each vehicle in the time period T day obtained in the step 3.3 to obtain the early warning frequency of the vehicle in unit running time;
step 4, clustering of vehicle safety risks
Taking the early warning frequency of the unit driving mileage of the vehicle and the early warning frequency of the unit driving time of the vehicle as clustering objects, and dividing risk grades; clustering the two-dimensional data based on an AGNES hierarchical clustering algorithm;
the method specifically comprises the following steps:
1) determining an input sample set O { (WFM)1，WFT1),(WFM2，WFT2),...,(WFMn，WFTn) And the number of clusters Z value, where WFMiAnd WFTiRespectively representing the early warning frequency of i unit driving mileage and the early warning frequency of unit driving time of the vehicle, wherein i is 1,2, …, n is the number of samples, namely the total number of vehicles or drivers;
2) adopting a bottom-up clustering strategy to sample each object O in the set OiAs a cluster of samples phiiCalculating any two sample clusters phicAnd phihComparing the distances, wherein c is not equal to h, and searching two sample clusters phi with the shortest distanceh、ΦcAs a new set of sample clusters, phiv＝Φh∪ΦcWherein c, v and h are positive integers, and the values are all less than or equal to n;
3) cluster distance metric function
The proximity degree between the two clusters is determined by the two clusters together, the average distance is adopted to calculate the aggregation degree between any two sample clusters, and the aggregation degree is used for representing the similarity degree of the two sample clusters;
G＝(WFMg,WFTg),Q＝(WFMq,WFTq) (6)
in the formula: phih,ΦcRespectively represent a certain sample cluster, | phih|、|Φc| represents a cluster of samples Φ respectivelyh,ΦcThe number of middle elements, G and Q respectively represent a sample cluster phih,ΦcA certain sample in (1), WFMgWFT (weighted average) representing the warning frequency of g unit mileage of a vehiclegIndicating the warning frequency of the vehicle in g units of travel time, WFMqWFT (weighted average) representing the warning frequency of the vehicle for q units of mileageqThe early warning frequency of the vehicle Q unit running time is represented, dist (G, Q) represents the Euclidean distance between two samples G and Q;
4) comparing the average distance between each sample cluster obtained by calculation in the step 3), merging two clusters with the closest distance based on a clustering merging principle, continuously updating and merging to form a new cluster, and performing cluster division again;
5) determination of termination condition
According to the set value of the clustering number Z, if the clustering number is equal to the value Z, clustering is not needed, and clustering is terminated to obtain a Z-type risk level;
Step 5.1, determining category variable and discriminant variable
Carrying out risk grade division according to the risk degree of the vehicle, dividing the risk grade into 1 grade, 2 grade and … Z grade according to the clustering result in the step 4, taking the early warning frequency of the unit driving mileage of the vehicle and the early warning frequency of the unit driving time of the vehicle as discrimination variables of the discrimination analysis, and taking the risk grade of the vehicle as a category variable of the discrimination analysis;
step 5.2, establishing a discrimination function, namely determining the quantitative relation between the category variable and the discrimination variable according to the sample data, and establishing the Fisher discrimination function by using a Fisher discrimination criterion;
step 6, risk identification
Classifying and distinguishing the new samples according to the established Fisher distinguishing function
1) Calculating the center of the category to which the sample point belongs in the Y space; (2) for the new sample, calculating Fisher discriminant function value Y0Construction of Y0Distance function W (Y) from the center of each class0) And calculate Y0Distance from the center of each category; (3) the category to which the image belongs is determined by a distance discrimination method.
On the basis of the scheme, the over-small value of the mileage in the step 2.2 is determined by the cumulative distribution of daily mileage, and is determined according to less than 2% of the total sample; the judgment method for the excessive value of the mileage is determined according to the principle of 3 sigma of basic statistical knowledge.
On the basis of the scheme, in step 3.2, the calculation formula of the total driving range of each vehicle in the time period T is as follows:
wherein M isiIs the total driving range of vehicle i over a period of T days; m isijIs the mileage of vehicle i on day j, i ═ 1,2 … n; j is 1,2 … T.
On the basis of the scheme, in step 3.3, the total time of the vehicle trip is as follows: the time difference between the starting time of the vehicle on day recording and the ending time of the stop recording is the total time of the vehicle trip; parking time: and finding a continuous invariant point of the mileage record, and counting the duration of the continuous invariant point to obtain the stop time of the vehicle.
On the basis of the above scheme, in step 5.2, the Fisher discriminant function is:
y＝b1x1+b2x2+…+bpxp(8)
in the formula, bαFor the discrimination coefficient, α is 1,2, …, and p, Y is a dimension of the sample in the low-dimensional Y space.
On the basis of the above-mentioned scheme,in step 6, the distance function W (Y)0) As follows:
in the formula (I), the compound is shown in the specification,
respectively representing the center points of the e-th and f-th class samples in the Y space,
representing the center points of all samples of class e and class f in Y space,
∑-1an inverse matrix representing the e and f type covariance matrixes; wherein e is 1,2, …, Z, f is 1,2, …, Z, e ≠ f;
when W (Y)0)>At 0, the new sample point belongs to class e.
Drawings
The invention has the following drawings:
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a flow chart of data outlier screening;
FIG. 3 is a graph of the clustering results of vehicle risks;
FIG. 4 is a spatial distribution diagram of sample points in the discriminant function.
Detailed Description
The invention is described in further detail below with reference to figures 1-4.
And step 1, acquiring data.
The vehicle ID, the uploading time, the vehicle safety early warning condition and the driving mileage record are obtained through the vehicle networking vehicle-mounted terminal of the Beidou positioning system.
Step 2.1, traversing all data files, deleting travel records which do not meet requirements, deleting text files lacking attribute values, ensuring the integrity of attributes, and deleting records repeatedly uploaded by the vehicle-mounted terminal;
and 2.2, traversing all the data files, and calculating the driving mileage of the vehicle on the current day, the accumulated early warning frequency of each early warning position on the current day and the time difference of the beginning and the end of the current day. Determining an oversize value and an undersize value point of the driving mileage of the vehicle according to the driving mileage accumulation distribution map of all vehicles; rejecting records with too large or too small mileage (the point with too small value is determined according to 2% of the total sample with insufficient mileage on the same day, and the point with too large value is defined by using 3 sigma principle according to basic statistical knowledge), and rejecting travel records with the time difference of the origin-destination point of the same day being less than 0;
step 3, extracting truck trip key variables
And 3.1, counting the early warning frequency of each vehicle in the time period T. Taking the trip log of the vehicle every day as a unit, respectively counting the early warning frequency according to early warning types (overspeed early warning, fatigue driving early warning, collision early warning and the like), obtaining the early warning frequency of each early warning position of the vehicle every day, and accumulating the selected early warning positions according to the requirement to calculate the total early warning frequency of the selected early warning positions on the day. Suppose that the internet-of-vehicles data records S pieces of early warning information with the length of L for each vehicle in one day, wherein L is the total digit of the early warning, namely the total quantity of the early warning types of the vehicle-mounted equipment, the early warning information can be expressed as:
wherein the content of the first and second substances,
s＝1,2,…,S,l＝1,2,…,L，
if counting the early warning frequency a of the first early warning position in one day of a vehiclelThen, there are:
namely recording the early warning frequency of the ith position when the vehicle travels the same day. And accumulating the early warning frequencies of the selected specific early warning positions to obtain the total early warning frequency of each vehicle at the selected early warning positions on the same day.If the first three warning positions are selected, the total warning frequency of the vehicle in the selected warning positions in one day is
And finally, accumulating the early warning frequency of the vehicle (at the selected early warning position) in the same ID vehicle time period T days every day to obtain the early warning total frequency of the vehicle (at the selected early warning position) in the T days.
And 3.2, counting the total driving range of each vehicle in the time period T. The change of the driving mileage records the mileage change of a vehicle instrument panel and reflects the driving distance of the vehicle. Specifically, the current-day driving range of the same vehicle ID is accumulated by taking the vehicle ID number as a unique identification code, and the total driving range of each vehicle in a time period T is finally obtained; the mileage on the day can be calculated by a 'mileage' attribute field of the vehicle, namely, the difference value of mileage of the mileage record starting point and the mileage record ending point on the day.
And 3.3, counting the total running time of each vehicle in the time period T. Vehicle travel time does not include the time lost by the vehicle due to waiting or delays. The principle of vehicle travel time extraction is as follows: the total travel time of the vehicle is calculated firstly, then the parking time is calculated, and the time difference between the total travel time and the parking time is the travel time of the vehicle. 1) Total time of vehicle trip: the total time of the vehicle trip is the duration from the beginning to the end of the recording of the vehicle-mounted instrument, and the time difference between the starting time of the recording of the starting point and the ending time of the stopping of the recording of the current day of the vehicle is the total time of the vehicle trip; 2) parking time: the position where the mileage is continuously unchanged is the static position of the vehicle, the continuous unchanged point of the mileage record is searched, the duration time of the continuous unchanged point is counted, and the stop time of the vehicle is obtained; 3) vehicle time of day: subtracting the parking time from the obtained total travel time to obtain the current-day driving time of the vehicle; 4) total travel time of vehicle in T days: similar to the vehicle total mileage statistics, the vehicle ID is used as a main key, and the vehicle running time with the same ID is accumulated to obtain the total running time of the vehicle.
Step 3.4, calculating the early Warning Frequency (WFM) of the unit driving mileage and the early Warning Frequency (WFT) of the unit driving time according to the early warning frequency of the unit driving mileage and the early warning frequency of the selected early warning position;
WFM＝TW/TM (3)
WFT＝TW/TT (4)
wherein, WFM (waiting Frequency per mileage) -the prewarning Frequency of unit driving range, WFT (waiting Frequency per Travel time) -the prewarning Frequency of unit driving time, TW (total waiting Frequency) -the prewarning total Frequency, TM (total mileage) -the total driving range, and TT (total Travel time) -the total driving time.
And 4, clustering the vehicle safety risks, and classifying the early warning levels of the sample vehicles based on an AGNES (AGglometric NESTING) hierarchical clustering algorithm. The method specifically comprises the following steps: 1) determining an input sample set O { (WFM)1，WFT1),(WFM2，WFT2),...,(WFMn，WFTn) And the number of clusters Z, Where (WFM)i，WFTi) And (i is 1,2, …, n) represents the early warning frequency per unit mileage and the early warning frequency per unit driving time of the vehicle respectively. 2) Adopting a bottom-up clustering strategy to sample each object O in the set OiAs a cluster of samples phiiCalculating any two sample clusters phicAnd phihThe distance (c ≠ h) between the two clusters of samples phi is compared to find the two closest clustersh、ΦcAs a new set of sample clusters, phiv＝Φh∪Φc. 3) Cluster distance metric function. The proximity between two clusters is determined by the two clusters together, and the aggregation between any two sample clusters is calculated by using the average distance (also called average-linking method) at this time to represent the similarity measurement mode of the two sample clusters.
G＝(WFMg,WFTg),Q＝(WFMq,WFTq) (6)
In the formula: phih,ΦcRespectively represent a certain sample cluster, | phih|、|Φc| represents a cluster of samples Φ respectivelyh,ΦcThe number of middle elements, G and Q respectively represent a sample cluster phih,ΦcDist (G, Q) denotes the euclidean distance between two samples G, Q;
4) comparing the average distance between each sample cluster obtained by calculation in the step 3), merging two clusters with the shortest distance based on a clustering merging principle, continuously updating and merging to form a new cluster, and performing cluster division again. E.g. cluster phi1Cluster of phi2The distance between the clusters is the minimum, then phi1And phi2Are merged to form a new cluster. 5) And (5) judging the termination condition. And according to the set value Z of the clustering number, if the clustering number is equal to the value Z, clustering is not needed, and clustering is terminated to obtain the risk level of the Z class.
Step 5.1, mean value inspection and covariance array homogeneity inspection, in order to ensure that the effect of discriminant analysis is ideal, the mean values of the discriminant variables under a plurality of categories are obviously different, otherwise, the probability of giving a wrong discriminant result is higher, and generally, the mean value inspection of the populations is firstly carried out, namely, whether the difference between groups of the discriminant variables under each category is obvious is discriminated.
The basic idea of Fisher discriminant analysis is to project and then discriminate, wherein the projection in discriminant analysis is the key of discriminant analysis, and high-dimensional data points are projected to low-dimensional data points according to the principle of maximizing inter-class dispersion and minimizing intra-class dispersion, so that the maximum inter-class separation of the sample is achieved. P-dimensional X-space sample points are projected into r (r < ═ p) -dimensional Y-space. The functional form of the Fisher discriminant function is as follows:
y＝b1x1+b2x2+…+bpxp(8)
wherein the coefficient bαCalled the discriminant coefficient, isThe influence of each input variable on the discriminant function can be determined by the principle of maximum inter-group dispersion and minimum intra-group dispersion. Y is a dimension of the sample in the low-dimensional Y space.
And transforming the sample points in the high-dimensional space into the low-dimensional space through transforming the original data coordinate system. Separating the total sample points as much as possible by coordinate transformation, firstly calculating the center of the class to which the sample points belong in Y space during discrimination, and calculating the Fisher discrimination function value Y of a new sample0And Y in Y space0The distance from the center of each class is determined by the distance determination method (mahalanobis distance) to determine the class to which the class belongs, and Y is constructed0Distance function W (Y) from the center of each class0)，
Wherein the content of the first and second substances,
respectively representing the center points of the e-th and f-th class samples in the Y space,
representing the center points of all samples of class e and class f in Y space,
∑-1and the inverse matrix of the e and f type covariance matrixes is shown.
When W (Y)0)>At 0, the new sample point belongs to class e.
And 5.2, determining a discrimination factor. And (4) according to the clustering result in the step (4), the risk levels of the vehicles are classified into 1 level, 2 level and … Z level, two key variables (the early warning frequency of unit driving mileage and the early warning frequency of unit driving time of the vehicles) of the early warning information are selected as discrimination factors or discrimination variables, and the risk levels of the vehicles are used as category variables for discrimination analysis.
And 5.3, establishing a discriminant function. Taking the risk grade determined in the step 4 as a category variable of discriminant analysis, taking the early warning frequency of unit travel mileage of the vehicle and the early warning frequency of unit travel time as discriminant variables, determining the quantitative relation between the category variable and the discriminant variables according to the existing sample data, and establishing a discriminant criterion by using a Fisher discriminant function;
step 5.3 the Fisher discriminant analysis described above can be operated by discriminant analysis in SPSS software to find the discriminant function.
And 5.4, checking the model result, and judging the interpretation degree of the model according to the frequency percentage of the confusion matrix or the distribution and position conditions of each sample point in the Fisher discriminant function space.
And 6, risk identification. And judging and predicting unknown classes of the new data through a discriminant function. For the sample points of new data, risk identification of the new sample (vehicle/driver) is achieved based on the discriminant function.
(1) Calculating the center of the category to which the sample point belongs in the Y space; (2) calculating Fisher discrimination function values of the new samples and distances between the new samples and the centers of all the classes in Y space; (3) and judging the category of the object by utilizing the distance judgment.
The invention can give the early warning frequency of unit travel mileage and the early warning frequency of unit travel time to a certain vehicle/driver, and can give the judgment function of discriminant analysis based on historical data to realize the identification or prediction of the risk.
1 case data introduction
The data used by the invention is from the internet of vehicles data from 2017 in 9 months to 2017 in 10 months provided by a certain enterprise, and comprises dynamic data items: early warning flag, mileage, update upload time, and static data item: vehicle ID (terminal number), etc. The raw data is first preprocessed and screened. Finally, 11139 original records are screened and processed to obtain 10039 effective records. And combining data of vehicles with the same ID on different days, taking one ID as a sample to obtain 862 vehicle samples, and evaluating the risk of the vehicle/driver. The resulting data pattern from the final collation is shown in table 1 below:
the interpretation of each variable is defined as follows:
column 1 ID represents the code of the vehicle, which is the unique code that identifies the vehicle. Different vehicles have different vehicle IDs, one vehicle ID for the same vehicle.
Column 3 WFT represents the frequency of warnings per unit of travel time (per hour) by the vehicle.
Table 1 arrangement of basic format (part) of sample data
2 clustering analysis based on case data
Processing the data into a format shown in table 1, clustering based on the early warning frequency of the second row of unit travel mileage and the early warning frequency of the third row of unit travel time in case data, firstly importing the data shown in table 1, inputting a preset Z value (Z is 3), calculating the distance between sample points by combining a formula (5) and a formula (6), and clustering according to the average distance between the sample points to obtain the clustering result. Partial results obtained by the clustering analysis are shown in table 2, the meaning of the first 3 columns of data is consistent with that in table 1, and the last column rl (risk level) is the clustering result of the vehicle and marks the risk level of the vehicle.
TABLE 2 clustering results of sample data
Fig. 3 shows the cluster analysis result of the sample data. As shown, the sample points may be classified into three categories, and the larger the values of WFM and WFT, the higher the warning frequency of the vehicle, i.e. the greater the risk level of the vehicle, so when the values of WFM and WFT are grouped into three categories, their risks may be defined as: 1-safe, 2-general, 3-dangerous. The WFM and WFT values of the sample points marked by circles in the graph are small, so that the sample with the lowest risk level is in a safe state; the diamonds represent sample points with larger WFM and WFT values, and the sample points have the highest risk level and are in a dangerous state; the triangles represent sample points with risk levels in the general state.
2 establishing a discriminant function based on the clustering result
And determining the quantity relation between the category variable and the discrimination variable according to the existing data by using the risk grade of the vehicle determined in the last step as a category variable of discrimination analysis and the early warning frequency of the unit mileage and the early warning frequency of the unit running time of the vehicle as the discrimination variable, and establishing a discrimination function by using a Fisher discrimination criterion.
(1) Determining a discrimination factor
Two key variables of the early warning information are selected as discrimination factors, namely the early warning frequency of unit driving mileage of the vehicle and the early warning frequency of unit driving time are used as discrimination variables.
(2) Establishing a discriminant function
And establishing a discrimination function to realize the identification and judgment of the risk vehicles, and establishing a related discrimination function according to the driving risk indexes and the risk early warning grades of all vehicles. The risk level of the vehicle is used as a category variable of discriminant analysis, the early warning frequency of unit travel mileage of the vehicle and the early warning frequency of unit travel time of the vehicle are used as discriminant variables, and a Fisher discriminant function is established to analyze the risk intensity of each vehicle, so that the risk assessment of the vehicle is realized. The risk discrimination function is established according to a Fisher discrimination method, and the risk grade discrimination is realized.
1. Established discriminant function
TABLE 3 coefficient of discriminant function
And obtaining a discriminant function:
in the formula: x is the number of1WFM (Warning frequency per mileage traveled); x is the number of2WFT (Unit Driving time)Inter-warning frequency).
2. Calculating a category center position
The category center positions of the three types of sample points are calculated as follows:
TABLE 4 function at Category center
3. Test of discrimination ability
In order to test whether the projection of the Fisher discriminant function can well separate various samples, and further judge which discriminant function is more important to the interpretation degree of the discriminant result, two characteristic values, the percentage of the interpreted variance and the cumulative percentage of the interpreted variance need to be calculated.
TABLE 5 summary of the calculated results
It can be seen that the ability of the first discrimination function to interpret the variance is 100% and the ability of the second discrimination function to interpret the variance is 0%, so the second discrimination function can be omitted. The final discriminant function obtained is:
y＝-1.019-2.619x1+5.426x2(10)
and substituting the Fisher discriminant function into the new sample point, calculating the distance between the new sample point and the center of each category, and discriminating the category to which the new sample point belongs by using distance discrimination.
If a new sample (new travel behavior data of a vehicle or a driver, or new travel behavior data of the latest month as it is) is obtained through data acquisition, the data of the new sample is as follows: the early warning frequency WFM of unit driving mileage is 0.735 times/10 km, and the early warning frequency WFT of unit driving time is 3.101 times/h.
Taking the new samples (0.735,3.101) as an example, the new samples are first substituted into the discriminant function (10) and the value is calculated to be 13.882, i.e., 13.882 for the new samples mapped into the one-dimensional spatial sample set Y. Then, since the final discriminant function is a single function of equation (10), it can be seen from function 1 of table 4 that the center points of each class are mapped to values of-0.700, 10.893, and 27.826, respectively, in the one-dimensional space Y. And calculating the distance between the new sample point and each category center point in the mapping Y, and judging the distance by using the formula (8), wherein the distance can be directly calculated as 14.582, 2.988 and 13.946 respectively because the judgment function is in a single function form. It can be known that the new sample point after mapping is closest to the central point of the second class, and finally the risk class of the new sample is determined to belong to the second class, that is, the risk class is general.
(4) Analysis of results
The distribution and position of each sample point in the Fisher discriminant function space are shown in fig. 4. From the distribution and the position of the sample points in the Fisher discriminant function space, the distribution of each category is concentrated, so that the discriminant effect is ideal.
Those not described in detail in this specification are within the skill of the art.
Claims (6)
1. The utility model provides a method of extraction and risk identification of freight train early warning information based on big dipper car networking which characterized in that includes several following steps:
step 1, obtaining original data related to vehicle early warning through a vehicle-mounted terminal of a vehicle networking provided with a Beidou positioning system, wherein the original data comprises: mileage information, safety early warning information and state data, wherein the state data comprises a vehicle ID, an ACC state and uploading time;
step 2, preprocessing the original data;
preprocessing and screening original data by using a Python programming technology; before analyzing the original data, the original data needs to be cleaned and sorted, so that the data quality is improved; the data cleaning comprises the following steps: filling missing values in the data, and identifying abnormal values and redundant data in the data; the method comprises the following specific steps of preprocessing original data by combining the characteristic of data storage of the vehicle-mounted terminal:
step 2.1, data missing value operation: the method comprises the steps of performing programming operation by using Python, introducing an os module and a numpy module in the Python, defining a required function, executing a main function, operating a text file stored in a vehicle-mounted terminal, deleting the text file lacking an attribute value, and ensuring the integrity of the attribute;
step 2.2, data abnormal value operation: the data abnormal value comprises abnormal mileage, abnormal early warning state and abnormal uploading time;
1) and (3) mileage exception handling: firstly, traversing all data files, calculating the mileage of the vehicle on the day, secondly, making an accumulated distribution map of the mileage of the vehicle on the day, and determining an overlarge value point and an undersize value point of the mileage of the vehicle on the day; finally, removing travel records with overlarge or undersize numerical values in the driving mileage of the vehicle on the day;
2) and (4) abnormal processing of the early warning state: counting the early warning duration of each early warning bit, solving the duration of single early warning of the early warning bit, and deleting the early warning state with obvious errors;
3) and (3) processing an uploading time exception: calculating the time difference between adjacent uploading points, and eliminating the recording points with the difference value of the adjacent uploading time unchanged or less than zero;
step 2.3, redundant data operation: traversing all records of the vehicle trip on the same day, and deleting the repeatedly uploaded records and the records with smaller data scale on the same day, wherein the records specifically comprise: comparing the uploaded records, deleting the repeatedly uploaded records, and repeatedly executing until all data files are traversed; counting the trip time of the vehicle on the day according to the record with smaller data scale, and deleting the trip record less than 15 min;
step 3, extracting key variables of the vehicle early warning information;
according to the historical travel data of the vehicle, two key variables in the vehicle running early warning information are extracted: the early warning frequency of the unit driving mileage of the vehicle and the early warning frequency of the unit driving time of the vehicle; firstly, counting the total early warning frequency of a specific early warning position of each vehicle in a time period of T days, wherein T is a positive integer; secondly, counting the total driving mileage of each vehicle within T days; thirdly, counting the total driving time of each vehicle within the time period T days; then, calculating the early warning frequency of each vehicle in unit driving mileage and the early warning frequency of each vehicle in unit driving time; taking a vehicle ID as a unique identification code, and counting and summarizing travel record information of the same ID vehicle in different time periods; the method comprises the following specific steps:
step 3.1, counting the early warning frequency of each vehicle at a specific early warning position within a time period T days;
taking travel early warning records of vehicles as objects, firstly taking a vehicle ID as a unique identification code, counting the early warning frequency of each early warning position of each vehicle in one day, then accumulating selected specific early warning positions to obtain the early warning total frequency of each vehicle in the specific early warning position in one day, and finally accumulating the early warning frequency of the same ID vehicle in the specific early warning position every day in a time period T to obtain the early warning total frequency of each vehicle in the specific early warning position in the time period T;
step 3.2, counting the total driving mileage of each vehicle within the time period T days;
the driving mileage is to record the mileage change of the instrument panel of the vehicle and reflect the driving distance of the vehicle; accumulating the driving mileage of the same ID vehicle by taking the vehicle ID number as a unique identification code, and finally obtaining the total driving mileage of each vehicle in a time period T days;
step 3.3, counting the total driving time of each vehicle in the period T days;
the vehicle running time does not include the time lost by the vehicle due to waiting or delay, and the principle of vehicle running time extraction is as follows: firstly, calculating the total travel time of the vehicle, and then calculating the parking time, wherein the time difference between the total travel time and the parking time is the total travel time of the vehicle; then, accumulating the running time of the same ID vehicle in the time period T by taking the vehicle ID as a unique identification code to obtain the total running time of each vehicle in the time period T;
step 3.4, dividing the early warning frequency of each vehicle at the specific early warning position in the time period T day obtained in the step 3.1 and the total driving range of each vehicle in the time period T day obtained in the step 3.2 to obtain the early warning frequency of the unit driving range of the vehicle; dividing the early warning frequency of each vehicle at the specific early warning position in the time period T day obtained in the step 3.1 and the total running time of each vehicle in the time period T day obtained in the step 3.3 to obtain the early warning frequency of the vehicle in unit running time;
step 4, clustering the vehicle safety risks;
taking the early warning frequency of the unit driving mileage of the vehicle and the early warning frequency of the unit driving time of the vehicle as clustering objects, and dividing risk grades; clustering the two-dimensional data based on an AGNES hierarchical clustering algorithm;
the method specifically comprises the following steps:
1) determining an input sample set O { (WFM)1，WFT1),(WFM2，WFT2),...,(WFMn，WFTn) And the number of clusters Z value, where WFMiAnd WFTiRespectively representing the early warning frequency of i unit driving mileage and the early warning frequency of unit driving time of the vehicle, wherein i is 1,2, …, n is the number of samples, namely the total number of vehicles or drivers;
2) adopting a bottom-up clustering strategy to sample each object O in the set OiAs a cluster of samples phiiCalculating any two sample clusters phicAnd phihComparing the distances, wherein c is not equal to h, and searching two sample clusters phi with the shortest distanceh、ΦcAs a new set of sample clusters, phiv＝Φh∪ΦcWherein c, v and h are positive integers, and the values are all less than or equal to n;
3) clustering cluster distance metric function;
the proximity degree between the two clusters is determined by the two clusters together, the average distance is adopted to calculate the aggregation degree between any two sample clusters, and the aggregation degree is used for representing the similarity degree of the two sample clusters;
G＝(WFMg,WFTg),Q＝(WFMq,WFTq) (6)
in the formula: phih,ΦcRespectively represent a certain sample cluster, | phih|、|Φc| represents a cluster of samples Φ respectivelyh,ΦcThe number of middle elements, G and Q respectively represent a sample cluster phih,ΦcA certain sample in (1), WFMgWFT (weighted average) representing the warning frequency of g unit mileage of a vehiclegIndicating the warning frequency of the vehicle in g units of travel time, WFMqWFT (weighted average) representing the warning frequency of the vehicle for q units of mileageqThe early warning frequency of the vehicle Q unit running time is represented, dist (G, Q) represents the Euclidean distance between two samples G and Q;
4) comparing the average distance between each sample cluster obtained by calculation in the step 3), merging two clusters with the closest distance based on a clustering merging principle, continuously updating and merging to form a new cluster, and performing cluster division again;
5) judging a termination condition;
according to the set value of the clustering number Z, if the clustering number is equal to the value Z, clustering is not needed, and clustering is terminated to obtain a Z-type risk level;
step 5, judging and analyzing;
step 5.1, determining a category variable and a discrimination variable;
carrying out risk grade division according to the risk degree of the vehicle, dividing the risk grade into 1 grade, 2 grade and … Z grade according to the clustering result in the step 4, taking the early warning frequency of the unit driving mileage of the vehicle and the early warning frequency of the unit driving time of the vehicle as discrimination variables of the discrimination analysis, and taking the risk grade of the vehicle as a category variable of the discrimination analysis;
step 5.2, establishing a discrimination function, namely determining the quantitative relation between the category variable and the discrimination variable according to the sample data, and establishing the Fisher discrimination function by using a Fisher discrimination criterion;
step 6, risk identification;
classifying and distinguishing the new samples according to the established Fisher distinguishing function;
(1) calculating the center of the category to which the sample point belongs in the Y space; (2) for the new sample, calculating Fisher discriminant function value Y0Construction of Y0Distance function W (Y) from the center of each class0) And calculate Y0Distance from the center of each category; (3) the category to which the image belongs is determined by a distance discrimination method.
2. The method for extracting early warning information and identifying risks of trucks based on the Beidou Internet of vehicles as claimed in claim 1, wherein the over-small points of mileage in step 2.2 are determined by cumulative daily mileage distribution, and are determined according to less than 2% of the total sample; the judgment method of the excessive value points of the mileage is determined according to the principle of 3 sigma of basic statistical knowledge.
3. The method for extracting early warning information and identifying risks of trucks based on the Beidou Internet of vehicles as claimed in claim 1, wherein in step 3.2, the calculation formula of the total driving range of each truck in the time period T is as follows:
Wherein M isiIs the total driving range of vehicle i over a period of T days; m isijIs the mileage of vehicle i on day j, i ═ 1,2 … n; j is 1,2 … T.
4. The method for extracting early warning information and identifying risks of trucks based on the Beidou Internet of vehicles as claimed in claim 1, wherein in step 3.3, the total time of vehicle traveling: the time difference between the starting time of the vehicle on day recording and the ending time of the stop recording is the total time of the vehicle trip; parking time: and finding a continuous invariant point of the mileage record, and counting the duration of the continuous invariant point to obtain the stop time of the vehicle.
5. The method for extracting early warning information and identifying risks of trucks based on the Beidou Internet of vehicles as claimed in claim 1, wherein in step 5.2, the Fisher discriminant function is as follows:
y＝b1x1+b2x2+…+bpxp(8)
in the formula, x1,x2,...,xpTo discriminate variables, bαFor the discrimination coefficient, α is 1,2, …, and p, Y is a dimension of the sample in the low-dimensional Y space.
6. The method for extracting early warning information and identifying risks of trucks based on Beidou Internet of vehicles according to claim 5, wherein in step 6, the distance function W (Y) is0) As follows:
in the formula (I), the compound is shown in the specification,
respectively representing the center points of the e-th and f-th class samples in the Y space,
representing the center points of all samples of class e and class f in Y space,
∑-1an inverse matrix representing the e and f type covariance matrixes; wherein e is 1,2, …, Z, f is 1,2, …, Z, e ≠ f;
when W (Y)0)>At 0, the new sample point belongs to class e.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910773932.2A CN110544373B (en) | 2019-08-21 | 2019-08-21 | Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910773932.2A CN110544373B (en) | 2019-08-21 | 2019-08-21 | Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110544373A CN110544373A (en) | 2019-12-06 |
| CN110544373B true CN110544373B (en) | 2020-11-03 |
Family
ID=68711676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910773932.2A Expired - Fee Related CN110544373B (en) | 2019-08-21 | 2019-08-21 | Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110544373B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111401399B (en) * | 2019-12-24 | 2023-11-10 | 中国国家铁路集团有限公司 | Accident early warning and classifying method and device for railway freight |
| CN111612372A (en) * | 2020-05-29 | 2020-09-01 | 上海万位科技有限公司 | Commercial vehicle default risk early warning method based on operation condition |
| CN112245728B (en) * | 2020-06-03 | 2022-11-29 | 北京化工大学 | Respirator false positive alarm signal identification method and system based on integrated tree |
| CN111881952A (en) * | 2020-07-14 | 2020-11-03 | 扬州大学 | Driver tendency analysis method based on early warning big data |
| CN112270460B (en) * | 2020-09-30 | 2023-10-27 | 交通运输部规划研究院 | Overweight truck cargo source site identification method based on multi-source data |
| CN112258014A (en) * | 2020-10-17 | 2021-01-22 | 中国石油化工股份有限公司 | Clustering and grouping-based risk discrimination analysis method for heat exchangers |
| CN112183899A (en) * | 2020-11-04 | 2021-01-05 | 北京嘀嘀无限科技发展有限公司 | Method, device, equipment and storage medium for determining safety degree prediction model |
| CN113256981B (en) * | 2021-06-09 | 2021-09-21 | 天津所托瑞安汽车科技有限公司 | Alarm analysis method, device, equipment and medium based on vehicle driving data |
| CN113657432A (en) * | 2021-06-30 | 2021-11-16 | 桂林电子科技大学 | Commercial vehicle driving behavior risk level identification method based on Internet of vehicles data |
| CN114132339A (en) * | 2021-12-29 | 2022-03-04 | 阿维塔科技(重庆)有限公司 | Automobile display method and system, vehicle and computer storage medium |
| CN114162136A (en) * | 2021-12-29 | 2022-03-11 | 阿维塔科技(重庆)有限公司 | Automobile driving route display method and system, vehicle and computer storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7859392B2 (en) * | 2006-05-22 | 2010-12-28 | Iwi, Inc. | System and method for monitoring and updating speed-by-street data |
| CN102881179A (en) * | 2012-06-15 | 2013-01-16 | 浙江吉利汽车研究院有限公司杭州分公司 | Active safety information collecting method and information service system for automobile |
| US9849887B2 (en) * | 2015-08-27 | 2017-12-26 | Intelligent Imaging Systems, Inc. | Contextual driver behavior monitoring |
| CN107742193A (en) * | 2017-11-28 | 2018-02-27 | 江苏大学 | A kind of driving Risk Forecast Method based on time-varying state transition probability Markov chain |
| CN108074396A (en) * | 2016-11-10 | 2018-05-25 | 关晓芙 | The evaluation method that drives safely and system |
| CN108417091A (en) * | 2018-05-10 | 2018-08-17 | 武汉理工大学 | Driving risk section identification based on net connection vehicle and early warning system and method |
-
2019
- 2019-08-21 CN CN201910773932.2A patent/CN110544373B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7859392B2 (en) * | 2006-05-22 | 2010-12-28 | Iwi, Inc. | System and method for monitoring and updating speed-by-street data |
| CN102881179A (en) * | 2012-06-15 | 2013-01-16 | 浙江吉利汽车研究院有限公司杭州分公司 | Active safety information collecting method and information service system for automobile |
| US9849887B2 (en) * | 2015-08-27 | 2017-12-26 | Intelligent Imaging Systems, Inc. | Contextual driver behavior monitoring |
| CN108074396A (en) * | 2016-11-10 | 2018-05-25 | 关晓芙 | The evaluation method that drives safely and system |
| CN107742193A (en) * | 2017-11-28 | 2018-02-27 | 江苏大学 | A kind of driving Risk Forecast Method based on time-varying state transition probability Markov chain |
| CN108417091A (en) * | 2018-05-10 | 2018-08-17 | 武汉理工大学 | Driving risk section identification based on net connection vehicle and early warning system and method |
Non-Patent Citations (1)
| Title |
|---|
| 基于聚类分析的营运驾驶人危险驾驶行为研究;牛增良 等;《山东交通学院学报》;20140331;第22卷(第1期);第19-23页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110544373A (en) | 2019-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110544373B (en) | Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles | |
| CN112508392B (en) | Dynamic evaluation method for traffic conflict risk of hidden danger road section of mountain area double-lane highway | |
| CN110197588B (en) | Method and device for evaluating driving behavior of large truck based on GPS track data | |
| CN106846801B (en) | Method for detecting abnormal behavior of regional loitering based on vehicle track | |
| CN103562978B (en) | Vehicle data analysis method and vehicle data analysis system | |
| CN111340427B (en) | Freight car running state identification method based on track data | |
| CN109408557B (en) | Traffic accident cause analysis method based on multiple correspondences and K-means clustering | |
| WO2018099574A1 (en) | System and method for determining a vehicle classification from gps tracks. | |
| CN108932255B (en) | Vehicle comprehensive capacity analysis method and device | |
| CN110562261B (en) | Method for detecting risk level of driver based on Markov model | |
| CN110570655B (en) | Vehicle feature evaluation method based on hierarchical clustering and decision tree | |
| Saha et al. | Application of principal component analysis for outlier detection in heterogeneous traffic data | |
| Wu et al. | Clustering of several typical behavioral characteristics of commercial vehicle drivers based on GPS data mining: Case study of highways in China | |
| CN113095387B (en) | Road risk identification method based on networking vehicle-mounted ADAS | |
| US11761858B2 (en) | Hierarchical data structure and method for prediction of tire wear | |
| CN111985749A (en) | Road risk assessment method and device | |
| CN113263993A (en) | Fault early warning method and device, communication equipment and storage medium | |
| CN110705628B (en) | Method for detecting risk level of driver based on hidden Markov model | |
| CN116753938A (en) | Vehicle test scene generation method, device, storage medium and equipment | |
| Simoncini et al. | Vehicle classification from low frequency GPS data | |
| Watanabe et al. | A method for the estimation of coexisting risk-inducing factors in traffic scenarios | |
| Chen et al. | The impact of truck proportion on traffic safety using surrogate safety measures in China | |
| CN114999181B (en) | Highway vehicle speed abnormality identification method based on ETC system data | |
| CN115688003A (en) | Driver identification method and device, computer equipment and readable storage medium | |
| CN114741428A (en) | AI behavior analysis early warning system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201103 Termination date: 20210821 |