CN113095713A - Driver space risk early warning method based on public transportation historical alarm data - Google Patents

Driver space risk early warning method based on public transportation historical alarm data Download PDF

Info

Publication number
CN113095713A
CN113095713A CN202110468206.7A CN202110468206A CN113095713A CN 113095713 A CN113095713 A CN 113095713A CN 202110468206 A CN202110468206 A CN 202110468206A CN 113095713 A CN113095713 A CN 113095713A
Authority
CN
China
Prior art keywords
alarm
risk
driver
data
warning
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.)
Granted
Application number
CN202110468206.7A
Other languages
Chinese (zh)
Other versions
CN113095713B (en
Inventor
于世军
周鹏
邓社军
卞张蕾
彭浪
刘根基
杨孝清
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yangzhou University
Original Assignee
Yangzhou University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yangzhou University filed Critical Yangzhou University
Priority to CN202110468206.7A priority Critical patent/CN113095713B/en
Publication of CN113095713A publication Critical patent/CN113095713A/en
Application granted granted Critical
Publication of CN113095713B publication Critical patent/CN113095713B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/088—Non-supervised learning, e.g. competitive learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06313—Resource planning in a project environment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a driver space risk early warning method based on public traffic historical warning data, which comprises the steps of obtaining public traffic driver driving warning data, transmitting the public traffic driver driving warning data into a database, and dividing a road analysis area according to a city where the warning data is located; carrying out risk classification on the bus drivers, and carrying out spatial connection on the alarm data and the divided risk analysis areas; connecting alarm data of different types of high-risk drivers with the divided risk analysis areas in space; comparing the individual high-risk alarm analysis area with the common high-risk alarm analysis area, and predicting the risk probability of the early-warning driver based on the Markov chain; predicting the risk probability of the early-warning driver based on a metabolism gray GM (1,1) model, and fusing the risk probability prediction results of the two times to obtain a comprehensive risk early-warning probability; and outputting the driver with high alarm probability in the next day according to the comprehensive risk early warning probability and forming a space early warning and improving report.

Description

Driver space risk early warning method based on public transportation historical alarm data
Technical Field
The invention relates to the technical field of traffic safety warning, in particular to a driver space risk early warning method based on public traffic historical warning data.
Background
The bus driver is easy to generate fatigue, distraction and other conditions because of long-term running, and meets some road conditions which are not good running road sections, and is easy to cause visual sense fatigue influence of the driver, so as to cause accidents.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The present invention has been made in view of the above-mentioned conventional problems.
Therefore, the invention provides a driver space risk early warning method based on public traffic historical warning data, which can solve the problem of safety risk in the existing public traffic driver driving process.
In order to solve the technical problems, the invention provides the following technical scheme: acquiring driving alarm data of a bus driver, transmitting the driving alarm data into a database, and dividing a road analysis area according to a city where the alarm data is located; carrying out risk classification on the bus drivers, and carrying out spatial connection on the alarm data and the divided risk analysis areas; connecting alarm data of different types of high-risk drivers with the divided risk analysis areas in space; comparing the individual high-risk alarm analysis area with the common high-risk alarm analysis area, and predicting the risk probability of the early-warning driver based on the Markov chain; predicting the risk probability of the early-warning driver based on a metabolism gray GM (1,1) model, and fusing the risk probability prediction results of the two times to obtain a comprehensive risk early-warning probability; and outputting the drivers with high alarm probability in the next day according to the comprehensive risk early warning probability and forming a space early warning prompt report.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: the input of the database comprises that the bus running alarm data is obtained in real time through a wireless transmission system installed on a bus, namely the alarm data of abnormal behaviors of the bus or a driver in the running state of a bus driver; the abnormal behaviors comprise smoking, mobile phone hitting, yawning and right expectation left behaviors of the driver during the vehicle passing process, and lane departure, forward collision, rapid acceleration and rapid deceleration of the driven vehicle; the main fields contained in the public transport driving alarm data comprise the date and time of the alarm, the serial number of the vehicle in which the alarm occurs, the bus line number in which the alarm occurs, the instantaneous speed of the public transport in which the alarm occurs, the serial number of the driver in which the alarm occurs, the name of the driver in which the alarm occurs, the longitude and latitude coordinates of the place in which the alarm occurs, and the type of the abnormal driving behavior in which the alarm occurs; and recording the bus driving alarm data into the database to prepare for data analysis.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: dividing the road analysis area comprises the steps of obtaining geographic file data of the urban road by utilizing a web crawler technology, wherein the geographic file data are of line types; breaking the lines at the intersection points by using a GIS space breaking technology to be used as the road to be divided at the intersection; generating buffer area roads with a buffer range threshold value of 20m-40m for all roads by a GIS buffer area generation technology; performing differential division according to the levels of different roads, and selecting a threshold value, wherein at the moment, the geographic file data is converted into a surface type from the line type; and taking the road analysis area as a road analysis area according to the divided buffer area, and carrying out digital numbering on each road analysis area.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: the risk classification comprises the steps of carrying out cluster analysis on the alarm types of the drivers by using a K-means clustering algorithm in unsupervised learning according to the alarm data of the drivers in a period of time; the divided clusters are 2 types, and two types of drivers with different labels are divided, namely two types of drivers with different tendency alarm types; performing data binning on the alarm occurrence times of the drivers with different inclination alarm types by using a natural discontinuity method; the classification type is 3 grades, which respectively correspond to low risk, medium risk and high risk; drivers in the high risk category in each level are screened out.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: the step of connecting the alarm data with the divided risk analysis areas in space comprises the step of connecting the alarm data with the road analysis areas by using a GIS space and counting the number of alarm points contained in each road analysis area; according to the alarm data points and the longitude and latitude coordinates of the road analysis area, marking the road analysis area containing the alarm points as the risk analysis area; classifying the risk analysis areas based on the drivers with different tendency alarm types according to the same classification principle of the drivers to obtain three types of risk analysis areas with different alarm types; classifying the classified risk analysis areas into 3 grades, namely low, medium and high risk analysis areas by using a natural discontinuity point method; and respectively extracting the analysis area numbers of the high-risk analysis areas in different categories, and taking the analysis area numbers as the roads of the dangerous areas shared by the drivers, wherein the high-risk analysis areas are a category of group set and comprise a plurality of analysis areas.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: and performing space connection on the alarm data of the high-risk drivers of different types and the divided risk analysis areas, namely performing GIS space connection on the classified and classified high-risk driver alarm data and the divided initial road analysis areas to obtain the alarm analysis areas of the high-risk drivers of different types, and taking the alarm analysis areas as the individual high-risk alarm risk analysis areas.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: the comparison comprises the steps of comparing the individual risk analysis areas and the common high risk analysis areas of the high risk drivers of different types, and judging whether the place where the driver gives an alarm is an area common behavior or not; and if a difference analysis area appears in the comparison, the driver is considered to have a different warning behavior in the difference analysis area, the driver is not a common warning behavior, the driver needs to be warned, and the driver is taken as the driver needing to be warned in a period of time and is marked as a warning driver.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: constructing a Markov chain state transition matrix p for the different road section with the alarm of the early-warning driver, wherein the matrix is a matrix with two rows and two columns, the row represents the alarm state of the current day, namely whether the alarm occurs on the road section on the current day, the occurrence is 1, the non-occurrence is 0, and the column represents the alarm state of the next day, namely whether the alarm occurs on the road section on the next day, the occurrence is 1, and the non-occurrence is 0;
Figure BDA0003044887900000031
wherein ,pijRepresenting state transition probabilities, e.g. p11The probability that the alarm occurs in the first day and the alarm occurs in the second day is represented, and obviously, four conditions are provided, namely, the alarm occurs in the first day, the alarm occurs in the second day, the alarm occurs in the first day, the alarm does not occur in the second day, the alarm does not occur in the first day, the alarm occurs in the second day, the alarm does not occur in the first day, and the alarm does not occur in the second day; the probability in the matrix is calculated and calibrated through the alarm data of the early-warning driver in the previous month; grouping the alarm data of the early-warning driver in pairs one month before, namely, grouping the alarm data in each group for 2 days, and deleting the alarm data until the alarm data cannot be grouped in the last day; spatially connecting the alarm data of the previous month with an analysis area of an individual road section where the driver is early warned to obtain data of whether the driver in the previous month alarms on the road section; according to the pairwise grouping, counting the occurrence frequency of four conditions in the state transition matrix, and solving the transition probability in the matrix by using the following formula;
Figure BDA0003044887900000041
wherein ,nijRepresenting the total number of occurrences of a certain set of conditions, NijRepresents the total number of packets in total; constructing a state distribution matrix according to whether the driver gives an alarm on the road section on the same day,
Snow＝[s1,s2]
wherein ,s1Representative of occurrence of an alarm, s2Representing no alarm, if alarm occurs on the same day, the alarm is 1,0]If no alarm occurs on the same day, the alarm is [0,1 ]]；
The probability of warning the driver to give an alarm the next day is Stom＝SnowP, the multiplication in the formula is a matrix multiplication.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: predicting the risk of the alarm data of the early-warning driver in the previous month by utilizing a metabolism gray GM (1,1) model, wherein the predicted target is the behavior of whether the early-warning driver gives an alarm in a certain road section, the prediction process of the metabolism gray GM (1,1) model comprises the steps of generating a new more regular discrete data column weakening the randomness by once accumulation, establishing a differential equation model, obtaining an approximate estimation value of the original data generated by the accumulation and subtraction of the solution at a discrete point, and predicting the subsequent development of the original data,
x(0)＝(x(0)(1),x(0)(2),…,x(0)(n))
wherein ,x(0)Representing the original data vector, x(0)(n) representing the original data value of the nth row, accumulating the original number row to obtain a new number row,
x(1)＝(x(1)(1),x(1)(2),…,x(1)(n))
wherein ,x(1)Representing accumulated data vector, x(1)(n) represents the accumulated data value of the nth row, and the new row is the row x: (n)0) The 1-time accumulation of (a) generates a sequence of numbers,
Figure BDA0003044887900000051
wherein ,x(1)(k) Represents a sequence of numbers x(0)Accumulation of corresponding front k items of data;
the gray derivative equation is defined as follows,
d(k)＝x(0)(k)＝x(1)(k)-x(1)(k-1)
defining a sequence x(1)(k) The close-proximity average of (a) is as follows,
Figure BDA0003044887900000052
wherein the new sequence z(1)(k)＝(z(1)(1),z(1)(2),…,z(1)(n)) is x: (1) The sequence of adjacent mean values;
the gray differential equation for GM (1,1) is defined as a first order linear differential equation,
d(k)+az(1)(k)＝b
then equation x(0)(k)+az(1)(k) B is the basic form of the GM (1,1) model;
wherein ,x(0)(k) Is the gray derivative, -a is the coefficient of development, z(1)(k) B is the amount of grey contribution;
the introduction of the matrix vector notation,
Figure BDA0003044887900000053
the GM (1,1) model is as follows,
Y＝Bu
the least square method is used to obtain the estimated values of the parameters a and b, then the whitening equation of the GM (1,1) model is as follows,
Figure BDA0003044887900000054
wherein ,
Figure BDA0003044887900000055
is an estimate of the parameter b that is,
Figure BDA0003044887900000056
is an estimate of the parameter a;
the whitening equation is subjected to integral solution, the calculated value is subtracted and reduced to obtain a predicted value,
Figure BDA0003044887900000057
predicting a new value every time, accumulating the new value to the back of the original data, and deleting the first data at the beginning in the original data so as to carry out metabolism behavior; dividing the historical data into a training set of 70 percent and a verification set of 30 percent for model training to obtain a trained metabolism GM (1,1) modelThe accuracy of which is taken as the confidence probability p of the modelgm(ii) a Predicting whether the driver generates an alarm Re the next day by using the model, and if the alarm Re is 1, not generating the alarm Re is 0; p is a radical of2＝Re*pgmAnd obtaining a metabolism gray GM (1,1) model for predicting the probability of the driver giving an alarm.
As an optimal scheme of the driver space risk early warning method based on the public transportation historical warning data, the method comprises the following steps: obtaining the comprehensive risk pre-warning probability includes,
pp＝λp2+(1-λ)Stom
wherein, lambda is generally 0.5; and forming a space early warning prompt report of a road section warning type, and sending the space early warning prompt report to a driver to remind the driver of needing to drive carefully in the area and the road section.
The invention has the beneficial effects that: the invention can improve the safety quality of the bus driver in the driving process, avoid fatigue driving in time, effectively prompt the driver to have driving safety standard and greatly reduce the occurrence probability of traffic accidents.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
FIG. 1 is a schematic flow chart of a driver spatial risk early warning method based on public transportation historical warning data according to an embodiment of the invention;
FIG. 2 is a database data diagram of a driver spatial risk early warning method based on public transportation historical warning data according to an embodiment of the invention;
FIG. 3 is a first type table data diagram of a driver spatial risk early warning method based on public transportation historical warning data according to an embodiment of the invention;
FIG. 4 is a second type table data diagram of a driver spatial risk early warning method based on public transportation historical warning data according to an embodiment of the invention;
FIG. 5 is a schematic diagram of a road section analysis area establishment of a public transportation historical alarm data-based driver spatial risk early warning method according to an embodiment of the invention;
FIG. 6 is a schematic diagram of a driver cluster analysis result of the public transportation historical alarm data-based driver spatial risk early warning method according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of driver warning data binning of a bus historical warning data-based driver spatial risk early warning method according to an embodiment of the invention;
FIG. 8 is a schematic diagram of extracting data of different types of high-risk drivers according to the public transportation historical warning data-based driver space risk early warning method in one embodiment of the present invention;
fig. 9 is a schematic diagram of a road analysis area common hazard area of a driver spatial risk early warning method based on public transportation historical warning data according to an embodiment of the present invention;
fig. 10 is a schematic diagram of an individual dangerous area in a road analysis area of a driver space risk early warning method based on public transportation historical warning data according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
The present invention will be described in detail with reference to the drawings, wherein the cross-sectional views illustrating the structure of the device are not enlarged partially in general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Referring to fig. 1, for a first embodiment of the present invention, a driver spatial risk early warning method based on public transportation historical warning data is provided, which is characterized in that: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
s1: and acquiring the driving alarm data of the bus driver, transmitting the driving alarm data into a database, and dividing a road analysis area according to the city where the alarm data is located. It should be noted that the input database includes:
the bus running alarm data is obtained in real time through a wireless transmission system installed on a bus, namely the alarm data of abnormal behaviors of the bus or a driver in the running state of a bus driver;
the abnormal behaviors comprise smoking, mobile phone calling, yawning and right expectation behaviors of a driver during the vehicle passing, and lane departure, forward collision, rapid acceleration and rapid deceleration of a driving vehicle;
the main fields contained in the public transport driving alarm data comprise the date and time of the alarm, the serial number of the vehicle in which the alarm occurs, the bus line number in which the alarm occurs, the instantaneous speed of the public transport in which the alarm occurs, the serial number of the driver in which the alarm occurs, the name of the driver in which the alarm occurs, the longitude and latitude coordinates of the place where the alarm occurs, and the type of the abnormal driving behavior in which the alarm occurs;
and recording the bus driving alarm data into a database to prepare for data analysis.
Further, dividing the road analysis area includes:
acquiring geographic file data of a road in which the urban road is located by utilizing a web crawler technology, wherein the geographic file data is of a line type;
breaking the lines at the intersection points by using a GIS space breaking technology to be used as the road to be divided at the intersection;
generating buffer area roads with a buffer range threshold value of 20m-40m for all roads by a GIS buffer area generation technology;
distinguishing and dividing according to the levels of different roads, and selecting a threshold value, wherein at the moment, the geographic file data is converted into a surface type from a line type;
and taking the divided buffer areas as road analysis areas, and numbering each road analysis area in a digital mode.
S2: and carrying out risk classification on the bus drivers, and carrying out spatial connection on the alarm data and the divided risk analysis areas. It should be noted that in this step, the risk classification includes:
according to alarm data generated by a driver within a period of time, carrying out cluster analysis on the alarm types generated by the driver by using a K-means clustering algorithm in unsupervised learning;
the divided clusters are 2 types, and two types of drivers with different labels are divided, namely two types of drivers with different tendency alarm types;
performing data binning on the alarm occurrence times of drivers with different inclination alarm types by using a natural discontinuity method;
the classification type is 3 grades, which respectively correspond to low risk, medium risk and high risk;
drivers in the high risk category in each level are screened out.
Specifically, the spatial connection of the alarm data and the divided risk analysis areas includes:
connecting the alarm data with the road analysis areas by using a GIS space, and counting the number of alarm points contained in each road analysis area;
according to the alarm data points and the longitude and latitude coordinates of the road analysis area, marking the road analysis area containing the alarm points as a risk analysis area;
classifying the risk analysis areas based on drivers with different tendency alarm types according to the same classification principle of the drivers to obtain three types of risk analysis areas with different alarm types;
classifying the classified risk analysis areas into 3 grades, namely low, medium and high risk analysis areas by using a natural discontinuity point method;
and respectively extracting the analysis area numbers of the high-risk analysis areas in different categories, and taking the analysis area numbers as the roads of the dangerous areas shared by the drivers, wherein the high-risk analysis areas are a category of group set and comprise a plurality of analysis areas.
S3: and connecting the alarm data of different types of high-risk drivers with the divided risk analysis areas in space. Among them, it is also to be noted that:
and performing GIS space connection on the classified and graded high-risk driver alarm data and the divided initial road analysis areas to obtain analysis areas for alarming of different types of high-risk drivers, and taking the analysis areas as individual high-risk alarm risk analysis areas.
S4: and comparing the individual high-risk alarm analysis area with the common high-risk alarm analysis area, and predicting the risk probability of the early-warning driver based on the Markov chain. It should be further noted that the comparison includes:
comparing individual risk analysis areas and common high risk analysis areas of different types of high risk drivers, and judging whether the place where the driver gives an alarm is an area common behavior or not;
if the difference analysis area appears in the comparison, the driver is considered to have a different warning behavior in the difference analysis area, but not a common warning behavior, the driver needs to be warned, and the driver is taken as the driver needing to be warned in a period of time and is marked as a warning driver.
Specifically, this embodiment also needs to be described as follows:
constructing a state transition matrix p of a Markov chain for a different road section giving an alarm to an early-warning driver, wherein the matrix is a matrix with two rows and two columns, the row represents the alarm state of the current day, namely whether the alarm occurs on the road section on the current day, the occurrence is equal to 1, the non-occurrence is equal to 0, and the column represents the alarm state of the next day, namely whether the alarm occurs on the road section on the second day, the occurrence is equal to 1, and the non-occurrence is equal to 0;
Figure BDA0003044887900000101
wherein ,pijRepresenting state transition probabilities, e.g. p11The probability that the alarm occurs in the first day and the alarm occurs in the second day is represented, and obviously, four conditions are provided, namely, the alarm occurs in the first day, the alarm occurs in the second day, the alarm occurs in the first day, the alarm does not occur in the second day, the alarm does not occur in the first day, the alarm occurs in the second day, the alarm does not occur in the first day, and the alarm does not occur in the second day;
the probability in the matrix is calculated and calibrated by warning data of a driver in the previous month;
grouping the alarm data of the early-warning driver in a month in pairs, namely, the time of 2 days in each group, and deleting the alarm data until the alarm data cannot be grouped in the last day;
spatially connecting the alarm data of the previous month with an analysis area of an individual road section where the driver is early warned to obtain data of whether the driver in the previous month alarms on the road section;
according to the pairwise grouping, counting the occurrence times of four conditions in the state transition matrix, and solving the transition probability in the matrix by using the following formula;
Figure BDA0003044887900000102
wherein ,nijRepresenting the total number of occurrences of a certain set of conditions, NijRepresents the total number of packets in total;
constructing a state distribution matrix according to whether the driver gives an alarm on the road section on the same day,
Snow＝[s1,s2]
wherein ,s1Representative of occurrence of an alarm, s2Representing no alarm, if alarm occurs on the same day, the alarm is 1,0]If no alarm occurs on the same day, the alarm is [0,1 ]]；
The probability of warning the driver to give an alarm the next day is Stom＝SnowP, the multiplication in the formula is a matrix multiplication.
S5: and predicting the risk probability of the early-warning driver based on a metabolism gray GM (1,1) model, and fusing the risk probability prediction results of the two times to obtain the comprehensive risk early-warning probability. Wherein, it needs to be explained again that:
the risk prediction is carried out on the alarm data of the early-warning driver in the previous month by utilizing a metabolism gray GM (1,1) model, the prediction target is the behavior of whether the early-warning driver gives an alarm in a certain road section, the prediction process of the metabolism gray GM (1,1) model comprises the steps of,
generating new discrete data columns with more regularity for weakening randomness by once accumulation, establishing a differential equation model, obtaining an approximate estimation value of the original data generated by accumulating and subtracting solutions at discrete points, thereby predicting the subsequent development of the original data,
x(0)＝(x(0)(1),x(0)(2),…,x(0)(n))
wherein ,x(0)Representing the original data vector, x(0)(n) representing the original data value of the nth row, accumulating the original data value to obtain a new data value,
x(1)＝(x(1)(1),x(1)(2),…,x(1)(n))
wherein ,x(1) Representing accumulated data vector, x(1)(n) represents the accumulated data value of the nth row, and the new row is the row x: (n)0) The 1-time accumulation of (a) generates a sequence of numbers,
Figure BDA0003044887900000111
wherein ,x(1)(k) Represents a sequence of numbers x(0)Accumulation of corresponding front k items of data;
the gray derivative equation is defined as follows,
d(k)＝x(0)(k)＝x(1)(k)-x(1)(k-1)
defining a sequence x(1)(k) The close-proximity average of (a) is as follows,
Figure BDA0003044887900000121
wherein the new sequence z(1)(k)＝(z(1)(1),z(1)(2),…,z(1)(n)) is x: (1) The sequence of adjacent mean values;
the gray differential equation for GM (1,1) is defined as a first order linear differential equation,
d(k)+az(1)(k)＝b
then equation x(0)(k)+az(1)(k) B isBasic form of GM (1,1) model;
wherein ,x(0)(k) Is the gray derivative, -a is the coefficient of development, z(1)(k) B is the amount of grey contribution;
the introduction of the matrix vector notation,
Figure BDA0003044887900000122
the GM (1,1) model is as follows,
Y＝Bu
the least square method is used to obtain the estimated values of the parameters a and b, then the whitening equation of the GM (1,1) model is as follows,
Figure BDA0003044887900000123
wherein ,
Figure BDA0003044887900000124
is an estimate of the parameter b that is,
Figure BDA0003044887900000125
is an estimate of the parameter a;
the whitening equation is subjected to integral solution, the calculated value is subtracted and reduced to obtain a predicted value,
Figure BDA0003044887900000126
predicting a new value every time, accumulating the new value to the back of the original data, and deleting the first data at the beginning in the original data so as to carry out metabolism behavior;
dividing historical data into a training set of 70 percent and a verification set of 30 percent for model training to obtain a trained metabolism GM (1,1) model, and taking the accuracy as the confidence probability p of the modelgm；
Predicting whether the driver generates an alarm Re the next day by using the model, if the alarm Re is 1, and if not, recording the alarm Re as 0;
p2＝Re*pgmand obtaining a metabolism gray GM (1,1) model for predicting the probability of the driver giving an alarm.
S6: and outputting the drivers with high alarm probability in the next day according to the comprehensive risk early warning probability and forming a space early warning prompt report. It should be explained again in this step that obtaining the comprehensive risk early warning probability includes:
pp＝λp2+(1-λ)Stom
wherein, lambda is generally 0.5;
and forming a space early warning prompt report of a road section warning type, and sending the space early warning prompt report to a driver to remind the driver of needing to drive carefully in the area and the road section.
Preferably, in this embodiment, it should be further noted that a GIS technology (geographic information system) is a product of crossing multiple disciplines, and it provides multiple spatial and dynamic geographic information in real time by using geographic model analysis based on geographic space, and the GIS technology in this embodiment converts data information uploaded in a database into geographic graphic display for analysis, and forms a spatial early warning prompt by combining with a spatial connection policy, so as to further specify the driving safety of a driver, and specifically, the actual operation is implemented by importing a software-written code program into a computing system, and part of the operation codes are as follows:
Figure BDA0003044887900000131
Figure BDA0003044887900000141
example 2
Referring to fig. 2 to 10, a second embodiment of the present invention is different from the first embodiment in that an experimental test of a driver spatial risk early warning method based on public transportation history warning data is provided, which specifically includes:
taking the bus driving alarm data in a certain city in 2020 and 2021 as an example, the example analysis is carried out, and the mainly used software is python programming software, an SQL Server database and qgis open source geographic software.
(1) Initial data introduction
Referring to fig. 2, alarm data in the bus driving process is obtained through a vehicle-mounted device and stored in a database, the alarm data mainly comprises two types of tables, the first type of table is named as "bus + date" and mainly records the alarm behavior of a vehicle, the second type of table is named as "count + date", the times of different alarm behaviors of a driver are mainly counted, the first type of table is shown in fig. 3, the second type of table is shown in fig. 4, for privacy, the name of a person and the line name appearing in the table are hidden, and the column of "exceptingname" in the first type of table represents the type of alarm occurring.
(2) Dividing urban road analysis area
Referring to fig. 5, a partial road analysis area of a certain city is obtained according to the method of the road analysis area.
(3) Driver risk classification
The drivers are classified into two categories by using the K-means clustering algorithm, and the results are shown in FIG. 6, wherein one column of labels is the clustering result.
By using the natural breakpoint method, the three risk levels of low, medium and high are classified for different types of drivers, and as a result, as shown in fig. 7, it can be seen that the high risk interval of the first type of drivers is (192,639), the high risk interval of the second type of drivers is (332,814), all the intervals are left-open and right-closed, and the drivers with the secondary alarm number in the interval are extracted.
(4) The road analysis area is connected with the data space
Referring to fig. 8, in order to obtain a common danger zone of the road analysis area, referring to fig. 9, an individual danger zone of a high-risk driver is obtained.
(5) Deriving driver risk probability predictions
The risk probability of 10 high-risk drivers in different types in a certain road section is predicted, and is specifically shown in table 1:
table 1: a risk prediction probability table.
Figure BDA0003044887900000151
The driver 6 can be found to have a high risk according to the probability, and the driver can be warned in an early warning manner by combining the position of the road section where the driver occurs.
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.

Claims (10)

1. A driver space risk early warning method based on public traffic historical warning data is characterized by comprising the following steps: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
acquiring driving alarm data of a bus driver, transmitting the driving alarm data into a database, and dividing a road analysis area according to a city where the alarm data is located;
carrying out risk classification on the bus drivers, and carrying out spatial connection on the alarm data and the divided risk analysis areas;
connecting alarm data of different types of high-risk drivers with the divided risk analysis areas in space;
comparing the individual high-risk alarm analysis area with the common high-risk alarm analysis area, and predicting the risk probability of the early-warning driver based on the Markov chain;
predicting the risk probability of the early-warning driver based on a metabolism gray GM (1,1) model, and fusing the risk probability prediction results of the two times to obtain a comprehensive risk early-warning probability;
and outputting the drivers with high alarm probability in the next day according to the comprehensive risk early warning probability and forming a space early warning prompt report.
2. The bus history alarm data-based driver space risk early warning method as claimed in claim 1, wherein: entering the database includes entering a database of the data,
the bus running alarm data is obtained in real time through a wireless transmission system installed on a bus, namely the alarm data of abnormal behaviors of the bus or a driver in the running state of a bus driver;
the abnormal behaviors comprise smoking, mobile phone hitting, yawning and right expectation left behaviors of the driver during the vehicle passing process, and lane departure, forward collision, rapid acceleration and rapid deceleration of the driven vehicle;
the main fields contained in the public transport driving alarm data comprise the date and time of the alarm, the serial number of the vehicle in which the alarm occurs, the bus line number in which the alarm occurs, the instantaneous speed of the public transport in which the alarm occurs, the serial number of the driver in which the alarm occurs, the name of the driver in which the alarm occurs, the longitude and latitude coordinates of the place in which the alarm occurs, and the type of the abnormal driving behavior in which the alarm occurs;
and recording the bus driving alarm data into the database to prepare for data analysis.
3. The driver spatial risk early warning method based on the public transportation historical warning data as claimed in claim 1 or 2, characterized in that: the dividing of the road analysis region may include,
acquiring geographic file data of a road in which the urban road is located by utilizing a web crawler technology, wherein the geographic file data is of a line type;
breaking the lines at the intersection points by using a GIS space breaking technology to be used as the road to be divided at the intersection;
generating buffer area roads with a buffer range threshold value of 20m-40m for all roads by a GIS buffer area generation technology;
performing differential division according to the levels of different roads, and selecting a threshold value, wherein at the moment, the geographic file data is converted into a surface type from the line type;
and taking the road analysis area as a road analysis area according to the divided buffer area, and carrying out digital numbering on each road analysis area.
4. The bus history alarm data-based driver space risk early warning method as claimed in claim 3, wherein: the risk classification includes a list of the risk categories,
according to the alarm data generated by the driver within a period of time, carrying out cluster analysis on the alarm types generated by the driver by using a K-means clustering algorithm in unsupervised learning;
the divided clusters are 2 types, and two types of drivers with different labels are divided, namely two types of drivers with different tendency alarm types;
performing data binning on the alarm occurrence times of the drivers with different inclination alarm types by using a natural discontinuity method;
the classification type is 3 grades, which respectively correspond to low risk, medium risk and high risk;
drivers in the high risk category in each level are screened out.
5. The bus history alarm data-based driver space risk early warning method as claimed in claim 4, wherein: spatially linking the alarm data with the partitioned risk analysis areas comprises,
connecting the alarm data with the road analysis areas by using a GIS space, and counting the number of alarm points contained in each road analysis area;
according to the alarm data points and the longitude and latitude coordinates of the road analysis area, marking the road analysis area containing the alarm points as the risk analysis area;
classifying the risk analysis areas based on the drivers with different tendency alarm types according to the same classification principle of the drivers to obtain three types of risk analysis areas with different alarm types;
classifying the classified risk analysis areas into 3 grades, namely low, medium and high risk analysis areas by using a natural discontinuity point method;
and respectively extracting the analysis area numbers of the high-risk analysis areas in different categories, and taking the analysis area numbers as the roads of the dangerous areas shared by the drivers, wherein the high-risk analysis areas are a category of group set and comprise a plurality of analysis areas.
6. The bus history alarm data-based driver space risk early warning method as claimed in claim 5, wherein: spatially linking the different types of high risk driver alert data with the already partitioned risk analysis areas includes,
and performing GIS space connection on the classified and graded high-risk driver alarm data and the divided initial road analysis areas to obtain analysis areas for alarming of different types of high-risk drivers, and taking the analysis areas as individual high-risk alarm risk analysis areas.
7. The bus history alarm data-based driver space risk early warning method as claimed in claim 6, wherein: the comparison includes the comparison of the number of pixels,
comparing the individual risk analysis areas and the common high risk analysis areas of the high risk drivers of different types, and judging whether the place where the driver gives an alarm is an area common behavior or not;
and if a difference analysis area appears in the comparison, the driver is considered to have a different warning behavior in the difference analysis area, the driver is not a common warning behavior, the driver needs to be warned, and the driver is taken as the driver needing to be warned in a period of time and is marked as a warning driver.
8. The bus history alarm data-based driver space risk early warning method as claimed in claim 7, wherein: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
constructing a Markov chain state transition matrix p for the different road section of the early-warning driver with the alarm, wherein the matrix is a matrix with two rows and two columns, the row represents the alarm state of the current day, namely whether the alarm occurs on the road section on the current day, the occurrence is 1, the non-occurrence is 0, and the column represents the alarm state of the next day, namely whether the alarm occurs on the road section on the next day, the occurrence is 1, and the non-occurrence is 0;
Figure FDA0003044887890000031
wherein ,pijRepresenting state transition probabilities, e.g. p11The probability that the alarm occurs in the first day and the alarm occurs in the second day is represented, and obviously, four conditions are provided, namely, the alarm occurs in the first day, the alarm occurs in the second day, the alarm occurs in the first day, the alarm does not occur in the second day, the alarm does not occur in the first day, the alarm occurs in the second day, the alarm does not occur in the first day, and the alarm does not occur in the second day;
the probability in the matrix is calculated and calibrated through the alarm data of the early-warning driver in the previous month;
grouping the alarm data of the early-warning driver in pairs one month before, namely, grouping the alarm data in each group for 2 days, and deleting the alarm data until the alarm data cannot be grouped in the last day;
spatially connecting the alarm data of the previous month with an analysis area of an individual road section where the driver is early warned to obtain data of whether the driver in the previous month alarms on the road section;
according to the pairwise grouping, counting the occurrence frequency of four conditions in the state transition matrix, and solving the transition probability in the matrix by using the following formula;
Figure FDA0003044887890000041
wherein ,nijRepresenting the total number of occurrences of a certain set of conditions, NijRepresents the total number of packets in total;
constructing a state distribution matrix according to whether the driver gives an alarm on the road section on the same day,
Snow＝[s1,s2]
wherein ,s1Representative of occurrence of an alarm, s2If no alarm occurs on the representative, the alarm occurs on the same dayAlarm is [1,0 ]]If no alarm occurs on the same day, the alarm is [0,1 ]]；
The probability of warning the driver to give an alarm the next day is Stom＝SnowP, the multiplication in the formula is a matrix multiplication.
9. The bus history alarm data-based driver space risk early warning method as claimed in claim 8, wherein: utilizing a metabolism gray GM (1,1) model to carry out risk prediction on alarm data of an early-warning driver in the previous month, wherein the prediction target is the behavior of whether the early-warning driver gives an alarm in a certain road section, and the prediction process of the metabolism gray GM (1,1) model comprises the steps of,
generating new discrete data columns with more regularity for weakening randomness by once accumulation, establishing a differential equation model, obtaining an approximate estimation value of the original data generated by accumulating and subtracting solutions at discrete points, thereby predicting the subsequent development of the original data,
x(0)＝(x(0)(1),x(0)(2),…,x(0)(n))
wherein ,x(0)Representing the original data vector, x(0)(n) representing the original data value of the nth row, accumulating the original number row to obtain a new number row,
x(1)＝(x(1)(1),x(1)(2),…,x(1)(n))
wherein ,x(1)Representing accumulated data vector, x(1)(n) represents the accumulated data value of the nth column, and the new column is the number column x(0)The 1-time accumulation of (a) generates a sequence of numbers,
Figure FDA0003044887890000051
wherein ,x(1)(k) Represents a sequence of numbers x(0)Accumulation of corresponding front k items of data;
the gray derivative equation is defined as follows,
d(k)＝x(0)(k)＝x(1)(k)-x(1)(k-1)
defining a sequence x(1)(k) The close-proximity average of (a) is as follows,
Figure FDA0003044887890000052
wherein the new sequence z(1)(k)＝(z(1)(1),z(1)(2),…,z(1)(n)) is x(1)The sequence of adjacent mean values;
the gray differential equation for GM (1,1) is defined as a first order linear differential equation,
d(k)+az(1)(k)＝b
then equation x(0)(k)+az(1)(k) B is the basic form of the GM (1,1) model;
wherein ,x(0)(k) Is the gray derivative, -a is the coefficient of development, z(1)(k) B is the amount of grey contribution;
the introduction of the matrix vector notation,
Figure FDA0003044887890000053
the GM (1,1) model is as follows,
Y＝Bu
the least square method is used to obtain the estimated values of the parameters a and b, then the whitening equation of the GM (1,1) model is as follows,
Figure FDA0003044887890000054
wherein ,
Figure FDA0003044887890000055
is an estimate of the parameter b that is,
Figure FDA0003044887890000056
is an estimate of the parameter a;
the whitening equation is subjected to integral solution, the calculated value is subtracted and reduced to obtain a predicted value,
Figure FDA0003044887890000057
predicting a new value every time, accumulating the new value to the back of the original data, and deleting the first data at the beginning in the original data so as to carry out metabolism behavior;
dividing historical data into a training set of 70 percent and a verification set of 30 percent for model training to obtain a trained metabolism GM (1,1) model, and taking the accuracy as the confidence probability p of the modelgm；
Predicting whether the driver generates an alarm Re the next day by using the model, and if the alarm Re is 1, not generating the alarm Re is 0;
p2＝Re*pgmand obtaining a metabolism gray GM (1,1) model for predicting the probability of the driver giving an alarm.
10. The bus history alarm data-based driver space risk early warning method as claimed in claim 9, wherein: obtaining the comprehensive risk pre-warning probability includes,
pp＝λp2+(1-λ)Stom
wherein, lambda is generally 0.5;
and forming a space early warning prompt report of a road section warning type, and sending the space early warning prompt report to a driver to remind the driver of needing to drive carefully in the area and the road section.
CN202110468206.7A 2021-04-28 2021-04-28 Driver space risk early warning method based on public transport history alarm data Active CN113095713B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110468206.7A CN113095713B (en) 2021-04-28 2021-04-28 Driver space risk early warning method based on public transport history alarm data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110468206.7A CN113095713B (en) 2021-04-28 2021-04-28 Driver space risk early warning method based on public transport history alarm data

Publications (2)

Publication Number Publication Date
CN113095713A true CN113095713A (en) 2021-07-09
CN113095713B CN113095713B (en) 2023-10-20

Family

ID=76681331

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110468206.7A Active CN113095713B (en) 2021-04-28 2021-04-28 Driver space risk early warning method based on public transport history alarm data

Country Status (1)

Country Link
CN (1) CN113095713B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113624514A (en) * 2021-08-17 2021-11-09 中国汽车技术研究中心有限公司 Test method, system, electronic device and medium for driver state monitoring product
CN114913398A (en) * 2022-05-10 2022-08-16 武汉长江通信智联技术有限公司 Method for identifying real-time driving risk and judging risk level of passenger driver

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101182802A (en) * 2007-12-06 2008-05-21 哈尔滨工程大学 Consecutive pressurization system failure prediction, diagnostic equipment, failure prediction and diagnostic method
CN103033761A (en) * 2012-12-17 2013-04-10 哈尔滨工业大学 Lithium ion battery residual life forecasting method of dynamic gray related vector machine
US20150314727A1 (en) * 2012-12-17 2015-11-05 Toyota Jidosha Kabushiki Kaisha Motive power operation guidance apparatus
CN106095963A (en) * 2016-06-17 2016-11-09 上海经达信息科技股份有限公司 Vehicle drive behavior analysis big data public service platform under the Internet+epoch
CN106127586A (en) * 2016-06-17 2016-11-16 上海经达信息科技股份有限公司 Vehicle insurance rate aid decision-making system under big data age
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
US20180143754A1 (en) * 2016-11-22 2018-05-24 Sarfaraz K. Niazi Vehicle steering and control device (vscd)
CN109572550A (en) * 2018-12-28 2019-04-05 西安航空学院 A kind of wheelpath prediction technique, system, computer equipment and storage medium
CN109829601A (en) * 2018-12-07 2019-05-31 深圳大学 A kind of driver drives the prediction technique and system of vehicle risk behavior
CN110176153A (en) * 2019-05-20 2019-08-27 重庆大学 A kind of blind area vehicle collision prewarning method based on edge calculations
CN110705628A (en) * 2019-09-26 2020-01-17 长安大学 Method for detecting risk level of driver based on hidden Markov model
CN111881952A (en) * 2020-07-14 2020-11-03 扬州大学 Driver tendency analysis method based on early warning big data
CN111898878A (en) * 2020-07-14 2020-11-06 扬州大学 Vehicle risk space-time distribution analysis method based on early warning big data

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101182802A (en) * 2007-12-06 2008-05-21 哈尔滨工程大学 Consecutive pressurization system failure prediction, diagnostic equipment, failure prediction and diagnostic method
CN103033761A (en) * 2012-12-17 2013-04-10 哈尔滨工业大学 Lithium ion battery residual life forecasting method of dynamic gray related vector machine
US20150314727A1 (en) * 2012-12-17 2015-11-05 Toyota Jidosha Kabushiki Kaisha Motive power operation guidance apparatus
CN106095963A (en) * 2016-06-17 2016-11-09 上海经达信息科技股份有限公司 Vehicle drive behavior analysis big data public service platform under the Internet+epoch
CN106127586A (en) * 2016-06-17 2016-11-16 上海经达信息科技股份有限公司 Vehicle insurance rate aid decision-making system under big data age
US20180143754A1 (en) * 2016-11-22 2018-05-24 Sarfaraz K. Niazi Vehicle steering and control device (vscd)
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
CN109829601A (en) * 2018-12-07 2019-05-31 深圳大学 A kind of driver drives the prediction technique and system of vehicle risk behavior
CN109572550A (en) * 2018-12-28 2019-04-05 西安航空学院 A kind of wheelpath prediction technique, system, computer equipment and storage medium
CN110176153A (en) * 2019-05-20 2019-08-27 重庆大学 A kind of blind area vehicle collision prewarning method based on edge calculations
CN110705628A (en) * 2019-09-26 2020-01-17 长安大学 Method for detecting risk level of driver based on hidden Markov model
CN111881952A (en) * 2020-07-14 2020-11-03 扬州大学 Driver tendency analysis method based on early warning big data
CN111898878A (en) * 2020-07-14 2020-11-06 扬州大学 Vehicle risk space-time distribution analysis method based on early warning big data

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王春莲: ""智慧城市中物联网及云计算技术的应用"", 《电子技术与软件工程》, no. 13 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113624514A (en) * 2021-08-17 2021-11-09 中国汽车技术研究中心有限公司 Test method, system, electronic device and medium for driver state monitoring product
CN114913398A (en) * 2022-05-10 2022-08-16 武汉长江通信智联技术有限公司 Method for identifying real-time driving risk and judging risk level of passenger driver

Also Published As

Publication number Publication date
CN113095713B (en) 2023-10-20

Similar Documents

Publication Publication Date Title
CN104834977B (en) Traffic alert grade prediction technique based on learning distance metric
CN108596409B (en) Method for improving accident risk prediction precision of traffic hazard personnel
Hossain et al. A real-time crash prediction model for the ramp vicinities of urban expressways
CN110807930B (en) Dangerous vehicle early warning method and device
CN109493566B (en) Fatigue driving prediction method based on GPS data
CN108717786B (en) Traffic accident cause mining method based on universality meta-rule
Le et al. Traffic accident hotspot identification by integrating kernel density estimation and spatial autocorrelation analysis: a case study
CN113095713A (en) Driver space risk early warning method based on public transportation historical alarm data
CN103996287A (en) Vehicle forced lane changing decision-making method based on decision-making tree model
CN109598931B (en) Group division and difference analysis method and system based on traffic safety risk
CN113988476A (en) Dynamic assessment prediction method for road transportation safety risk
Takale et al. Road accident prediction model using data mining techniques
CN114299742B (en) Speed limit information dynamic identification and update recommendation method for expressway
Fu Development of dynamic travel demand models for hurricane evacuation
CN117745079A (en) Platform system based on urban security risk assessment model
CN117475584A (en) Mountain road collapse disaster weather risk early warning system and computer equipment
CN114139984B (en) Urban traffic accident risk prediction method based on flow and accident cooperative sensing
CN116596307A (en) Method for constructing driver security portrait model based on public transport operation security data
CN114580874B (en) Multidimensional distributed data analysis system suitable for highway risk assessment
Ali et al. Employment of instrumented vehicles to identify real-time snowy weather conditions on freeways using supervised machine learning techniques–A naturalistic driving study
Pagany et al. Risk Prediction of Wildlife-vehicle Collisions Comparing Machine Learning Methods and Data Use
Hossain et al. Development of a real-time crash prediction model for urban expressway
CN118194727B (en) Expressway tunnel fire evacuation research method and system considering uncertainty
JP7414502B2 (en) Information processing device, information processing method and program
Levine Journey to crime Estimation

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