CN114118795A - Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway - Google Patents
Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway Download PDFInfo
- Publication number
- CN114118795A CN114118795A CN202111417767.0A CN202111417767A CN114118795A CN 114118795 A CN114118795 A CN 114118795A CN 202111417767 A CN202111417767 A CN 202111417767A CN 114118795 A CN114118795 A CN 114118795A
- Authority
- CN
- China
- Prior art keywords
- risk
- road
- intelligent
- highway
- intelligent heavy
- 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.)
- Pending
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
- 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
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention relates to a safety risk assessment grading and dynamic early warning method for an intelligent heavy-load highway, which comprises the following steps: 1) constructing an integral element framework of intelligent heavy-load highway risk assessment; 2) acquiring risk influence coefficients of all dynamic and static elements from the two aspects of probability and severity according to historical data; 3) determining the operation safety risk level of the intelligent heavy-duty highway by adopting a risk degree calculation model and a grading method according to the whole element frame and the element risk influence coefficient of the intelligent heavy-duty highway risk assessment; 4) and according to the measured data, realizing dynamic risk early warning through an early warning platform. Compared with the prior art, the method can comprehensively, clearly, conveniently, rationalize and standardize the evaluation of the running safety risk of the intelligent heavy-load highway, can realize dynamic risk display and early warning according to the change of the dynamic elements of road traffic, and has the advantages of novelty, scientificity, practicability and the like.
Description
Technical Field
The invention relates to the technical field of road traffic safety assessment, in particular to a safety risk assessment grading and dynamic early warning method for an intelligent heavy-load highway.
Background
The automatic driving automobile (intelligent automobile) is a high point of science and technology for disputed occupation of countries in the world, corresponding policies are set for automatic driving or development of the intelligent automobile industry in the United states, European countries, Japan and Korean, for example, 1 month in 2020, 4.0 plans for automatic driving are issued in the United states, and the leading position of the technology of the United states in the field of automatic driving is ensured. In addition to relevant standards or guidelines established at the national level, japan also establishes relevant traffic law regulations governing the behavior of autonomous vehicles. In 2019, in China, it is clearly proposed to strengthen the research and development of intelligent networked automobiles (intelligent automobiles, automatic driving and vehicle-road cooperation) and form an autonomous and controllable complete industrial chain, so that the development of the intelligent automobiles has important strategic significance.
However, in view of the current technical development level, it is difficult to implement L5-level full automatic driving, so that road testing and demonstration operation of intelligent automobiles in typical application scenarios for automatic driving are actively explored in some cities, especially major domestic major cities, and scene-driven automatic driving is implemented by combining an intelligent road and vehicle road cooperative information real-time transmission mechanism. Compared with urban roads and other grades of roads, on one hand, the intelligent heavy-load expressway is an important scene for landing and demonstration application of the automatic driving technology, and on the other hand, the expressway scene complexity is relatively low, no traffic participants such as pedestrians and non-motor vehicles exist, meanwhile, the expressway facility construction quality is better, and intelligent automobiles in the current technical level can be driven more conveniently if mark lines are clearer, the sight distance is good and the like; on the other hand, the highway has large transportation capacity, and for China, many highways are heavy-load traffic with large traffic capacity, and for intelligent vehicles, certain safety risks exist while the demand for large transportation capacity is high. The 'intelligence' of the intelligent heavy-load expressway is embodied in two aspects of 'intelligence of vehicles' and 'intelligence of roads'. "vehicle intelligence" means that some vehicles in the highway have a certain intelligence level, and mainly depends on the positioning, perception level and decision-making and planning method of the intelligent vehicles. The 'road intelligence' means that the road can acquire vehicle kinematic state information, road traffic environment information, climate environment information and the like in real time.
Under the background, in order to realize controllable risks of road testing and demonstration application in an intelligent heavy-duty highway scene and provide suggestions for highway scene selection and intelligent vehicle policy making, a set of complete dynamic risk assessment and early warning method for the intelligent heavy-duty highway scene is urgently needed, and at present, domestic and foreign research is relatively deficient.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide an intelligent heavy-duty highway safety risk assessment grading and dynamic early warning method, and aims to realize risk controllability in the demonstration application process of the intelligent heavy-duty highway, provide a selection suggestion of an intelligent heavy-duty highway scene for a decision maker and make a vehicle enterprise know the driving capacity of an intelligent vehicle on an actual road.
The purpose of the invention can be realized by the following technical scheme:
a safety risk degree evaluation grading and dynamic early warning method for an intelligent heavy-load highway comprises the following steps:
1) determining dynamic and static risk elements related to the running safety of the intelligent heavy-duty highway, dividing road sections, and constructing an integral element framework of the intelligent heavy-duty highway risk assessment;
2) acquiring risk influence coefficients of various dynamic and static elements from two aspects of probability and severity according to historical data including traffic flow data, manual driving vehicle accident data and intelligent vehicle takeover data;
3) determining the operation safety risk level of the intelligent heavy-duty highway by adopting a risk degree calculation model and a grading method according to the whole element frame and the element risk influence coefficient of the intelligent heavy-duty highway risk assessment;
4) and according to the measured data, realizing dynamic risk early warning through an early warning platform.
In the step 1), the static risk elements related to the operation safety of the intelligent heavy-duty highway comprise traditional risk elements and emerging risk elements, wherein the traditional risk elements comprise:
the general elements are as follows: lane number, lane width, gradient, curvature, road flatness, road surface skid resistance, center isolation belt type, sight distance, speed management measures, anti-collision guardrail type, emergency lane width, edge vibration belt, mark line, road sign board and anti-dazzle facility;
service area related elements: service area scale, service area spacing;
the number of toll lanes and the number of ETC in the toll station;
ramp related elements: the number of ramp lanes, the linear induction of ramp roads, the length of ramp and the visual distance of ramp;
tunnel-related elements: tunnel entrance type, tunnel interior light, tunnel maintenance way and tunnel water-proof and drainage facility;
relevant factors of the bridge: the novel risk factors comprise vehicle-road cooperative equipment, a novel lane line and a novel curb belt;
the emerging risk elements include:
vehicle-road cooperative equipment, vehicle-road communication modes and performances, high-precision maps, novel lane lines, novel traffic signs and novel road borders;
intelligent heavy haul highway operational safety related dynamic risk factors include speed, traffic flow, and weather condition changes.
In the step 1), road sections are divided by adopting an indefinite length method according to static risk elements related to the running safety of the intelligent heavy-load highway, and the method specifically comprises the following steps:
the method comprises the steps of firstly preliminarily dividing the road sections into large road sections according to the types and the line shapes of the facilities of the road sections, and then subdividing the large road sections into small road sections according to static risk elements.
In the step 1), the whole element framework for intelligent heavy-duty highway risk assessment is composed of an element basic framework and a facility type element framework, wherein the element basic framework specifically comprises the following components:
the facility type element frame is specifically as follows:
the step 2) specifically comprises the following steps:
21) the multi-source data processing and fusion specifically comprises the following steps:
carrying out data splicing on the acquired map data, the acquired vehicle end data and the acquired road section data according to the time and the space of the data to obtain a data set corresponding to the integral element frame;
22) acquiring the importance degree of the risk factors of the intelligent heavy-load expressway and sequencing the risk factors;
23) calibrating risk influence coefficients of various dynamic and static elements;
24) and correcting the risk coefficient of the wind influence of each dynamic and static element.
In the step 22), the importance degrees are sequentially ranked from high to low as the average traffic volume of the section of the road section, the average running speed of the section, the ratio of the large vehicle, the weather state, the marking line, the road side equipment condition, the road side cooperation condition, the road side object type, the type of the anti-collision guardrail, the type of the central isolation belt, the distance from the road side object, the road indicating sign, the curvature, the road surface anti-skid capability, the road flatness, the number of lanes, the gradient, the emergency lane, the quality of the curve, the edge vibration belt, the lane width, the sight distance and the anti-dazzle facility.
And 24), when more intelligent vehicle accident data or strong risk avoiding takeover data are obtained, correcting the risk influence coefficient by adopting a Bayesian network and a heuristic algorithm model.
In the step 3), calculating the operation safety risk degree according to the risk degree calculation model specifically comprises the following steps:
31) acquiring the types of main traffic accidents of the intelligent heavy-load highway, including co-directional rear-end collision and lateral collision between vehicles and single-vehicle accidents between the vehicles and the road;
32) evaluating the safety risk of the road section from the two aspects of accident probability and accident severity, and correcting the risk degree by combining the dynamic characteristics of the traffic flow to obtain the dynamic risk degree R of the road section corresponding to the type of the traffic accidentk(i) Then, there are:
Rk(i)＝Pk(i)×Sk(i)×rv×rf×rc
where k is the accident type, Rk(i) Degree of risk of accident k for road section i, Pk(i) Probability of occurrence of accident k for road section i, Sk(i) Severity of the accident k on the road section i, rvIs a velocity influence coefficient, rfIn order to be a traffic flow influence coefficient,rcis a traffic composition influence coefficient;
33) and calculating the risk degree of the road section, namely:
wherein, R (i) is the risk degree of the road section i, which represents the safety risk of the road section under the road traffic condition and the environmental condition, n is the type of accident that may occur, and for the expressway scene, the value of n is 3;
34) and calculating the risk degree of the line, namely:
where, ROL is a risk degree of the link, which represents a safety risk of the link under the road traffic condition and the environmental condition, m is the number of the links included in the link, and l (i) is the length of the link i.
And in the step 3), according to the risk degree calculation result and combining with expert scoring, determining 4 risk grades of red, orange, yellow and blue which are sequentially divided into the risks of the intelligent heavy-load expressway from high to low, and determining a risk grade division threshold value.
The step 4) is specifically as follows:
the intelligent heavy-load expressway risk dynamic early warning platform is established by integrally accessing multi-source actual measurement data and splicing, and utilizing integrated safety risk assessment and grading models, and specifically comprises the following steps:
41) integrated access and splicing of multi-source measured data:
splicing and deeply fusing data acquired by a plurality of data acquisition sensors on the intelligent heavy-duty highway to form a full-line real-time characteristic data set required by the evaluation of the running safety risk level of the intelligent heavy-duty highway;
42) safety risk assessment and hierarchical model integration:
the full-line real-time characteristic data set is accessed into a safety risk assessment and grading model, so that the real-time assessment of the running safety risk grade of the intelligent heavy-load highway is realized;
43) and constructing an intelligent dynamic risk early warning platform of the heavy-duty highway, and realizing visual dynamic early warning of the intelligent heavy-duty highway risk.
Compared with the prior art, the invention has the following advantages:
according to the method, the main characteristics of the intelligent heavy-load automobile and the highway are fully considered, all-weather display and dynamic early warning of the scene of the intelligent heavy-load highway are realized through the steps of risk element combing, element evaluation frame making, element influence coefficient calibration based on accident data and takeover data, safety risk model calculation, risk grading and the like, not only can the safety and the controllability of risks be realized, but also suggestions can be provided for selection of the scene of the intelligent heavy-load highway, and the driving capacity of the intelligent automobile and the degree of adaptation to the road environment are known at the same time.
The invention provides a risk assessment method for the intelligent vehicle to run on the heavy-load highway by combining vehicles, road facilities, traffic environment, weather environment and the like aiming at the typical road scene type of the highway and combining the characteristics of the intelligent vehicle, and the method has better innovation.
Thirdly, based on the actual characteristics of the intelligent vehicle and the heavy-load highway, risk influence coefficients of various dynamic and static elements are calibrated from two aspects of probability and severity through risk element combing and framework formulation, accident cause and intelligent vehicle takeover cause analysis, and an intelligent heavy-load highway risk degree calculation model and a risk grade division method are further provided, and dynamic display and real-time early warning of risk grades are realized.
The intelligent heavy-duty highway-oriented operation safety risk degree evaluation grading and dynamic early warning method is a set of systematic flow, can evaluate the operation safety risk of the intelligent heavy-duty highway simply and efficiently in real time, is practically applied and verified in a plurality of engineering projects, and meanwhile, the model and the method can be expanded to different scenes and can be continuously optimized and perfected according to more and more data input.
Drawings
FIG. 1 is an overall step framework diagram of the present invention.
FIG. 2 is an overall process flow diagram of the present invention.
Fig. 3 is an example of an interface of an intelligent heavy-duty highway operation safety risk assessment, analysis and early warning platform, wherein fig. 3a is an example of risk during peak-off period, and fig. 3b is an example of risk during peak-off period.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All embodiments that can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention belong to the protection scope of the present invention.
As shown in fig. 1 and 2, the invention provides a safety risk assessment grading and dynamic early warning method for an intelligent heavy-duty highway, which comprises the following steps:
1) aiming at the main characteristics of intelligent heavy-duty automobiles and expressways, comprehensively carding dynamic and static risk factors related to the running safety of the intelligent heavy-duty expressways to form an intelligent heavy-duty expressway risk assessment element framework;
2) according to historical data (including traffic flow data, manual driving vehicle accident data and intelligent vehicle takeover data), through accident cause and takeover cause analysis, risk influence coefficients of various dynamic and static elements are obtained from two aspects of probability and severity;
3) determining the operation safety risk level of the intelligent heavy-load expressway by using a risk degree calculation model and a grading method according to the risk evaluation framework and the element risk influence coefficient;
4) and developing an intelligent heavy-load highway operation safety risk assessment, analysis and early warning platform by utilizing the measured data, and realizing dynamic risk early warning.
The details of each step are described in detail below
In step 1), according to the main characteristics of the intelligent heavy-duty automobile and the highway, dynamic and static risk elements affecting the safety of the intelligent automobile are identified and divided into sections, and an intelligent heavy-duty highway risk assessment element frame is formulated, which specifically comprises the following steps:
11) distinguishing and analyzing the main characteristics of the intelligent automobile and the highway;
the intelligent automobile is different from a common manually-driven vehicle, and the behaviors of positioning, perception, decision, planning and the like of the vehicle are greatly different from the behaviors of human drivers under the condition of machine vision, so that the influence of the intelligent automobile is considered when the safety risk of the operation of the intelligent heavy-load highway is evaluated, meanwhile, the highway also has more typical characteristics such as an upper ramp, a lower ramp, a vertical crossing, opposite direction hard isolation and the like, and a model and a method are provided only aiming at the scene of the highway;
12) combing static risk factors related to the running safety of the intelligent heavy-load highway;
the static risk elements related to the running safety of the intelligent heavy-load expressway comprise traditional risk elements and emerging risk elements, wherein the traditional risk elements comprise lane number, lane width, gradient, curvature, pavement evenness, mark and marking line definition degree, mark and marking line perfection degree and the like; emerging risk elements include roadway coordination devices, novel lane lines, novel curb belts, and the like.
13) A method for road section division according to static risk elements is provided;
there are two methods for road segment division, one is a fixed length method, the other is an indefinite length method, the fixed length method means that the road segment is divided according to fixed length; the method for dividing the road sections by the indefinite length method is characterized in that roads with the same main road environment elements are divided into sections according to the difference of road environments.
14) Combing dynamic risk factors related to the running safety of the intelligent heavy-load highway;
the dynamic risk factors related to the running safety of the intelligent heavy-load expressway comprise speed, traffic flow, weather condition change and the like, and the main factors are the main reasons for causing the difference of risks in different time periods of the same road section.
15) And forming an intelligent heavy-load highway risk assessment element framework.
And classifying and dividing the dynamic and static risk factors to form an intelligent heavy-load highway risk assessment factor integral frame structure.
In step 2), multi-source data processing and fusion are required to be carried out on dynamic and static risk factor data, manual driving vehicle accident data, takeover data in the intelligent vehicle testing process and the like, then accident cause analysis and takeover cause analysis are carried out, and calibration and correction of risk coefficients of various dynamic and static factors are realized, and the method specifically comprises the following steps:
21) multi-source data processing and fusion;
for an intelligent heavy-load highway, data are often acquired through various means, such as vehicle-mounted orange video data, vehicle-mounted sensor data, data acquired by a roadside sensor and the like. And splicing and fusing data in different sensor ranges and different types according to time and space to obtain the dynamic and static risk element data set so as to carry out subsequent operation safety risk assessment.
22) Analyzing the cause of the manual driving accident and the reason for taking over by the intelligent vehicle;
by analyzing the domestic traffic accident records and combining the statistics of the international road assessment organization (IRAP) and the Chinese road assessment organization (China RAP) on traffic accidents under different road elements, the importance ranking of the elements related to the road traffic accidents is determined, then the taking-over reasons under the critical state are analyzed according to the actual road test data of intelligent vehicles, the environmental factors mainly related to the automatic driving function of the vehicles when the faults occur are found out, and the importance ranking of the elements is corrected.
23) Calibrating risk influence coefficients of various dynamic and static elements;
according to the ranking of the importance degree of the elements obtained by analyzing the cause of the artificial driving accident and the reason for taking over the intelligent vehicle, the influence of the elements on the operation risk is further determined by combining the actual accident statistics of the road to be evaluated and the analysis of the substitute traffic risk indexes (such as the Time To Collision (TTC) and the like), and the calibration of the influence coefficient of each element is realized
24) Correcting the wind influence risk coefficient of each dynamic and static element;
when more intelligent vehicle accident data or strong risk-avoiding takeover data (accidents do not take over) can be obtained, the risk influence coefficient can be corrected by utilizing models such as a Bayesian network and a heuristic algorithm.
In the step 3), the operation safety risk degree and the risk level of the intelligent heavy-load highway are obtained by combining the risk influence coefficients of all the elements calibrated and corrected in the step 2) according to the risk evaluation element framework of the intelligent heavy-load highway formulated in the step 1) and by a model calculation and risk classification method. The method comprises the following steps:
31) calculating a model aiming at the risk degree of the intelligent heavy-load expressway;
311) according to the actual situation of the intelligent heavy-load expressway, the types of possible traffic accidents are distinguished, and through actual investigation and accident recording, three types of traffic accidents mainly occur on the expressway by the intelligent heavy-load vehicle, including co-directional rear-end collision and lateral collision between vehicles and single-vehicle accidents between the vehicles and the road.
312) Evaluating the safety risk of the road section from two aspects of accident probability and accident severity; and correcting the risk degree by combining with the dynamic characteristics (speed, traffic flow and large vehicle proportion) of the traffic flow to obtain the dynamic risk degree of the road section corresponding to a certain accident type.
Rk(i)＝Pk(i)×Sk(i)×rv×rf×rc
Wherein k is an accident type; rk(i) The risk degree of the accident k occurring on the road section i; pk(i) The probability of the accident k occurring on the road section i is shown; sk(i) The severity of the accident k on the road section i; the influence coefficients of the accident occurrence probability and the accident severity mainly depend on the investigationThe road facility factors and the environmental factors are jointly determined; r isvThe speed is defined as the average driving speed (meter/second) of a road section as a speed influence coefficient; r isfAs a traffic flow influence coefficient, traffic flow is defined as a link average hourly traffic volume (vehicles/hour); r iscFor the influence coefficient of traffic composition, the highway scene mainly refers to the proportion of large vehicles, namely the proportion of the number of passing medium-large trucks and medium-large buses in the total number of vehicles in unit time.
313) And calculating the risk degree of the road section.
Wherein, R (i) is a road section risk degree which represents the safety risk of the road section under the road traffic condition and the environmental condition; n is the type of accident that may occur, and for a highway scenario, n is 3, i.e., co-directional rear-end collisions and side-to-side collisions between vehicles, as well as single-car accidents between vehicles and roads may occur.
314) And calculating the risk degree of the line.
Wherein, ROL (Risk of line) is a line risk degree, which represents the safety risk of the line under the road traffic condition and the environmental condition; m is the number of road sections contained in the line; l (i) is the length of the section i.
32) A risk classification method for intelligent heavy-duty expressways;
according to the risk degree calculation result, combining with expert scoring, determining 4 risk grades (risks are reduced in sequence) for dividing the risks of the intelligent heavy-load expressway into red, orange, yellow and blue, and determining a risk grade division threshold value.
In step 4), the integrated access of multisource measured data is performed, splicing is performed, an intelligent heavy-load highway risk dynamic early warning platform is constructed by utilizing an integrated security risk assessment and grading model, and the method specifically comprises the following steps:
41) a multi-source measured data integrated access and splicing method;
the data collected by a plurality of data collecting sensors on the intelligent heavy-duty highway are spliced and deeply fused to form a full-line real-time characteristic data set required by the operation safety risk level evaluation of the intelligent heavy-duty highway.
42) Safety risk assessment and hierarchical model integration;
and (3) accessing the data set into the calculation model in the step 3) to realize the real-time evaluation of the intelligent heavy-load highway operation safety risk level.
43) Intelligent heavy-load highway risk dynamic early warning platform.
By utilizing APP development tools such as MATLAB, VB and the like, a visual platform is developed, and intelligent heavy-load expressway risk dynamic early warning and display are realized.
Examples
The safety risk assessment grading and dynamic early warning method for the intelligent heavy-load highway comprises the following steps:
step 1: aiming at the main characteristics of intelligent heavy-duty automobiles and expressways, the dynamic and static risk factors related to the running safety of the intelligent heavy-duty expressways are comprehensively combed to form an intelligent heavy-duty expressway risk assessment factor frame, and the method comprises the following steps:
11) distinguishing and analyzing the main characteristics of the intelligent automobile and the highway;
the intelligent automobile: the intelligent automobile realizes the functions of vehicle positioning, perception, decision, planning and the like through a built-in algorithm.
The common positioning methods can be classified into the following categories:
(1) satellite positioning (GPS positioning, BDS positioning, etc.);
(2) high-precision map + radar positioning;
(3) inertial navigation positioning;
(4) the camera is locally positioned.
The existing intelligent automobile, especially the intelligent automobile above the L3 level, is often combined by a plurality of positioning modes, and the commonly used sensing hardware comprises:
(1) a long-range millimeter wave radar;
(2) a short-range millimeter wave radar;
(3) a laser radar;
(4) vehicle-mounted camera.
Common decision algorithms are:
(1) a physical-based behavioral characteristic decision model;
(2) an Artificial Intelligence (AI) machine learning based decision model;
(3) and (3) a decision model (such as a game model and the like) based on the behavior-consciousness of the driver.
The planning comprises path planning and speed planning, and comprises the following steps aiming at different intelligent vehicles:
(1) setting a path;
(2) and (5) planning the path in real time.
The highway has the characteristics of high speed, large traffic flow and the like, so that the speed and the flow are important factors influencing the running safety risk of the highway; in addition, the expressway has no complicated and various traffic participants such as non-motor vehicles, pedestrians and the like and traffic environments, the road facility conditions are relatively good, the entrance and the exit are controlled completely, generally speaking, the mark lines are relatively clear, and generally speaking, the demonstration application of the intelligent automobile on the expressway is easier to land according to the positioning, sensing, decision-making and planning modes of the intelligent automobile.
12) Combing static risk factors related to the running safety of the intelligent heavy-load highway;
according to the characteristics of the intelligent automobile and the highway, static risk factors related to the operation safety of the intelligent heavy-load highway are summarized and summarized as shown in table 1.
TABLE 1 static factors of safety risk of intelligent heavy-duty highway operation
13) A method for road section division according to static risk elements is provided;
there are two methods for segment division, one is a fixed length method and the other is an indefinite length method. The fixed length method is to divide road sections according to fixed length; the indefinite length method refers to dividing a road having the same main road environment elements into sections according to the difference of road environments. In the invention, in order to ensure scientific and reasonable risk assessment of the road section and facilitate subsequent risk calculation, the road section is divided by adopting an indefinite length method. The method is characterized in that the method comprises the steps of firstly preliminarily dividing the road into large road sections according to road facility types (common channel road sections, bridge sections, tunnel sections, interchange ramp sections and service sections) and line shapes (straight line sections and curve sections), and then dividing the large road sections into small road sections according to other static factors (such as gradient, marking and marking conditions and the like) to form a refined road section division scheme of the intelligent heavy-load high-speed line.
14) And combing dynamic risk factors related to the running safety of the intelligent heavy-load highway.
According to the characteristics of the intelligent automobile and the highway, dynamic risk factors such as traffic, climate environment, traffic participants and the like related to the operation safety of the intelligent heavy-load highway are summarized and summarized as shown in table 2.
TABLE 2 Intelligent heavy-duty highway operation safety dynamic risk factor
15) And forming an intelligent heavy-load highway risk assessment element framework.
Basic element framework: the characteristics of dynamic and static risk elements and intelligent automobiles are combined to form the following basic framework of the risk assessment elements of the intelligent heavy-load expressway as shown in table 3.
TABLE 3 basic framework of intelligent heavy-duty highway risk assessment elements
Combining the characteristics of the expressway, considering the influence of different facility types on the operation safety risk, a facility type element frame is formed, and the basic element frame is supplemented under the corresponding different facility types, as shown in table 4.
TABLE 4 facility type element framework
Thus, an integral element framework aiming at intelligent heavy-load highway risk assessment is formed.
Step 2: according to historical data (including traffic flow data, manual driving vehicle accident data and intelligent vehicle takeover data), through accident cause and takeover cause analysis, risk influence coefficients of various dynamic and static elements are obtained from two aspects of probability and severity; the method comprises the following steps:
21) multi-source data processing and fusion;
and (2) acquiring data through different ways, including map data, vehicle end acquired data and road section acquired data, and splicing the data through data time and space to obtain a data set corresponding to the element frame in the step (1).
22) Analyzing the cause of the manual driving accident and the reason for taking over by the intelligent vehicle;
by analyzing the domestic traffic accident records and combining the statistics of the international road assessment organization (IRAP) and the Chinese road assessment organization (China RAP) on traffic accidents under different road elements, the importance ranking of the elements related to the road traffic accidents is determined, then the taking-over reasons under the critical state are analyzed according to the actual road test data of intelligent vehicles, the environmental factors mainly related to the automatic driving function of the vehicles when the faults occur are found out, the importance ranking of the elements is corrected, and the ranking of the importance degrees of the corrected elements is obtained and is shown in a table 5.
TABLE 5 Intelligent heavy-duty highway Risk elements importance ranking
23) Calibrating risk influence coefficients of various dynamic and static elements;
according to the above-mentioned importance ranking of the importance degree of the elements, further according to the statistics of the intelligent automobile takeover data, the influence coefficients of the factors are calibrated and are shown in table 6.
TABLE 6 Risk impact coefficients of dynamic and static elements
24) And correcting the risk coefficient of the wind influence of each dynamic and static element.
When more intelligent vehicle accident data or strong risk-avoiding takeover data (accidents do not take over) can be obtained, the risk influence coefficient can be corrected by utilizing models such as a Bayesian network and a heuristic algorithm.
And step 3: obtaining the operation safety risk degree and the risk grade of the intelligent heavy-duty highway by a model calculation and risk grading method according to the risk evaluation element framework of the intelligent heavy-duty highway formulated in the step 1 and by combining the risk influence coefficients of all the elements calibrated and corrected in the step 2, and comprising the following steps of:
31) calculating a model aiming at the risk degree of the intelligent heavy-load expressway;
311) according to the actual situation of the intelligent heavy-load expressway, the types of possible traffic accidents are distinguished, and through actual investigation and accident recording, three types of traffic accidents mainly occur on the expressway by the intelligent heavy-load vehicle, including co-directional rear-end collision and lateral collision between vehicles and single-vehicle accidents between the vehicles and the road.
312) Evaluating the safety risk of the road section from two aspects of accident probability and accident severity; and correcting the risk degree by combining with the dynamic characteristics (speed, traffic flow and large vehicle proportion) of the traffic flow to obtain the dynamic risk degree of the road section corresponding to a certain accident type.
Rk(i)＝Pk(i)×Sk(i)×rv×rf×rc
Wherein k is an accident type; rk(i) The risk degree of the accident k occurring on the road section i; pk(i) The probability of the accident k occurring on the road section i is shown; sk(i) The severity of the accident k on the road section i; the influence coefficients of the accident occurrence probability and the accident severity are mainly determined by the road facility factors and the environmental factors in the investigation; r isvFor the speed influence coefficient, the speed is defined as the average driving speed (meter/second) of the road section; r isfAs a traffic flow influence coefficient, traffic flow is defined as a link average hourly traffic volume (vehicles/hour); r iscFor the influence coefficient of traffic composition, the highway scene mainly refers to the proportion of large vehicles, namely the proportion of the number of passing medium-large trucks and medium-large buses in the total number of vehicles in unit time.
313) And calculating the risk degree of the road section.
Wherein, R (i) is the risk degree of the road section i, and represents the safety risk of the road section under the road traffic condition and the environmental condition; n is the type of accident that may occur, and for a highway scenario, n is 3, i.e., co-directional rear-end collisions and side-to-side collisions between vehicles, as well as single-car accidents between vehicles and roads may occur.
314) And calculating the risk degree of the line.
Wherein, ROL (Risk of line) is a line risk degree, which represents the safety risk of the line under the road traffic condition and the environmental condition; m is the number of road sections contained in the line; l (i) is the length of the section i.
32) A risk classification method for intelligent heavy-duty expressways;
according to the risk degree calculation result, combining with expert scoring, determining 4 risk grades (risks are reduced in sequence) for dividing the risks of the intelligent heavy-load expressway into red, orange, yellow and blue, and determining a risk grade division threshold value.
Step four: the intelligent heavy-load expressway risk dynamic early warning platform is established by integrally accessing multi-source actual measurement data and splicing, and utilizing integrated safety risk assessment and grading models, and comprises the following steps:
41) a multi-source measured data integrated access and splicing method.
The data collected by a plurality of data collecting sensors on the intelligent heavy-duty highway are spliced and deeply fused to form a full-line real-time characteristic data set required by the operation safety risk level evaluation of the intelligent heavy-duty highway.
42) And safety risk assessment and hierarchical model integration.
And (3) accessing the data set into the calculation model in the step 3) to realize the real-time evaluation of the intelligent heavy-load highway operation safety risk level.
43) Intelligent heavy-load highway risk dynamic early warning platform.
An APP development tool of MATLAB is utilized to develop a visual platform, and intelligent heavy-load highway risk dynamic early warning and display cases are realized, for example, as shown in FIG. 3.
Claims (10)
1. A safety risk degree evaluation grading and dynamic early warning method for an intelligent heavy-load highway is characterized by comprising the following steps:
1) determining dynamic and static risk elements related to the running safety of the intelligent heavy-duty highway, dividing road sections, and constructing an integral element framework of the intelligent heavy-duty highway risk assessment;
2) acquiring risk influence coefficients of various dynamic and static elements from two aspects of probability and severity according to historical data including traffic flow data, manual driving vehicle accident data and intelligent vehicle takeover data;
3) determining the operation safety risk level of the intelligent heavy-duty highway by adopting a risk degree calculation model and a grading method according to the whole element frame and the element risk influence coefficient of the intelligent heavy-duty highway risk assessment;
4) and according to the measured data, realizing dynamic risk early warning through an early warning platform.
2. The method as claimed in claim 1, wherein in step 1), the static risk elements related to the operation safety of the intelligent heavy-duty highway include traditional risk elements and emerging risk elements, and the traditional risk elements include:
the general elements are as follows: lane number, lane width, gradient, curvature, road flatness, road surface skid resistance, center isolation belt type, sight distance, speed management measures, anti-collision guardrail type, emergency lane width, edge vibration belt, mark line, road sign board and anti-dazzle facility;
service area related elements: service area scale, service area spacing;
the number of toll lanes and the number of ETC in the toll station;
ramp related elements: the number of ramp lanes, the linear induction of ramp roads, the length of ramp and the visual distance of ramp;
tunnel-related elements: tunnel entrance type, tunnel interior light, tunnel maintenance way and tunnel water-proof and drainage facility;
relevant factors of the bridge: the novel risk factors comprise vehicle-road cooperative equipment, a novel lane line and a novel curb belt;
the emerging risk elements include:
vehicle-road cooperative equipment, vehicle-road communication modes and performances, high-precision maps, novel lane lines, novel traffic signs and novel road borders;
intelligent heavy haul highway operational safety related dynamic risk factors include speed, traffic flow, and weather condition changes.
3. The method for assessing and grading the safety risk degree and dynamically early warning the intelligent heavy-duty highway according to claim 1, wherein in the step 1), the road sections are divided by adopting an indefinite length method according to static risk elements related to the running safety of the intelligent heavy-duty highway, and the method specifically comprises the following steps:
the method comprises the steps of firstly preliminarily dividing the road sections into large road sections according to the types and the line shapes of the facilities of the road sections, and then subdividing the large road sections into small road sections according to static risk elements.
4. The method as claimed in claim 1, wherein in step 1), the whole element framework of the risk assessment of the intelligent heavy-duty highway is composed of an element basic framework and a facility type element framework, and the element basic framework is specifically:
the facility type element frame is specifically as follows:
5. the method for assessing and grading the safety risk degree and dynamically early warning the intelligent heavy-duty highway according to claim 4, wherein the step 2) comprises the following steps:
21) the multi-source data processing and fusion specifically comprises the following steps:
carrying out data splicing on the acquired map data, the acquired vehicle end data and the acquired road section data according to the time and the space of the data to obtain a data set corresponding to the integral element frame;
22) acquiring the importance degree of the risk factors of the intelligent heavy-load expressway and sequencing the risk factors;
23) calibrating risk influence coefficients of various dynamic and static elements;
24) and correcting the risk coefficient of the wind influence of each dynamic and static element.
6. The method as claimed in claim 5, wherein the importance levels in step 22) are sequentially ranked from high to low as average traffic volume of section, average driving speed in section, ratio of buses, weather conditions, marking lines, road-side equipment conditions, road-side coordination conditions, road-side object types, anti-collision guardrail types, central isolation zone types, distance to road-side object distances, road-indicating signs, curvature, road surface anti-skid capability, road flatness, lane number, gradient, emergency lane, curve quality, edge vibration zone, lane width, sight distance, and anti-dazzle facilities.
7. The method as claimed in claim 5, wherein in step 24), when more intelligent vehicle accident data or strong risk avoidance takeover data are obtained, the risk influence coefficient is modified by using a Bayesian network and a heuristic algorithm model.
8. The method for assessing and grading the safety risk degree and dynamically early warning the intelligent heavy-duty highway according to claim 1, wherein the step 3) of calculating the operation safety risk degree according to the risk degree calculation model specifically comprises the following steps:
31) acquiring the types of main traffic accidents of the intelligent heavy-load highway, including co-directional rear-end collision and lateral collision between vehicles and single-vehicle accidents between the vehicles and the road;
32) evaluating the safety risk of the road section from the two aspects of accident probability and accident severity, and correcting the risk degree by combining the dynamic characteristics of the traffic flow to obtain the dynamic risk degree R of the road section corresponding to the type of the traffic accidentk(i) Then, there are:
Rk(i)＝Pk(i)×Sk(i)×rv×rf×rc
where k is the accident type, Rk(i) Degree of risk of accident k for road section i, Pk(i) Probability of occurrence of accident k for road section i, Sk(i) Severity of the accident k on the road section i, rvIs a velocity influence coefficient, rfAs a traffic flow influence coefficient, rcIs a traffic composition influence coefficient;
33) and calculating the risk degree of the road section, namely:
wherein, R (i) is the risk degree of the road section i, which represents the safety risk of the road section under the road traffic condition and the environmental condition, n is the type of accident that may occur, and for the expressway scene, the value of n is 3;
34) and calculating the risk degree of the line, namely:
and the ROL is a line risk degree and represents the safety risk of the line under the road traffic condition and the environmental condition, m is the number of the road sections contained in the line, and Li is the length of the road section i.
9. The method as claimed in claim 8, wherein in the step 3), the risk of the intelligent heavy-duty highway is determined to be divided into 4 risk levels of red, orange, yellow and blue from high to low according to the risk calculation result and the expert score, and the risk level division threshold is determined.
10. The method for assessing and grading the safety risk degree and dynamically early warning the intelligent heavy-duty highway according to claim 1, wherein the step 4) specifically comprises:
the intelligent heavy-load expressway risk dynamic early warning platform is established by integrally accessing multi-source actual measurement data and splicing, and utilizing integrated safety risk assessment and grading models, and specifically comprises the following steps:
41) integrated access and splicing of multi-source measured data:
splicing and deeply fusing data acquired by a plurality of data acquisition sensors on the intelligent heavy-duty highway to form a full-line real-time characteristic data set required by the evaluation of the running safety risk level of the intelligent heavy-duty highway;
42) safety risk assessment and hierarchical model integration:
the full-line real-time characteristic data set is accessed into a safety risk assessment and grading model, so that the real-time assessment of the running safety risk grade of the intelligent heavy-load highway is realized;
43) and constructing an intelligent dynamic risk early warning platform of the heavy-duty highway, and realizing visual dynamic early warning of the intelligent heavy-duty highway risk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111417767.0A CN114118795A (en) | 2021-11-26 | 2021-11-26 | Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111417767.0A CN114118795A (en) | 2021-11-26 | 2021-11-26 | Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114118795A true CN114118795A (en) | 2022-03-01 |
Family
ID=80373739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111417767.0A Pending CN114118795A (en) | 2021-11-26 | 2021-11-26 | Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114118795A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116432448A (en) * | 2023-04-06 | 2023-07-14 | 西南交通大学 | Variable speed limit optimization method based on intelligent network coupling and driver compliance |
| CN117172424A (en) * | 2023-10-31 | 2023-12-05 | 江苏科运智慧交通科技有限公司 | Method for enhancing transmission effectiveness of road safety warning information |
| CN117194930A (en) * | 2023-11-07 | 2023-12-08 | 陕西交通电子工程科技有限公司 | Tunnel road section safety monitoring method based on vehicle driving data |
| CN117351708A (en) * | 2023-10-08 | 2024-01-05 | 北京迈道科技有限公司 | Expressway safety operation management early warning method, system and storage medium |
| CN117727183A (en) * | 2024-02-18 | 2024-03-19 | 南京淼瀛科技有限公司 | Automatic driving safety early warning method and system combining vehicle-road cooperation |
| CN117727183B (en) * | 2024-02-18 | 2024-05-17 | 南京淼瀛科技有限公司 | Automatic driving safety early warning method and system combining vehicle-road cooperation |
-
2021
- 2021-11-26 CN CN202111417767.0A patent/CN114118795A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116432448A (en) * | 2023-04-06 | 2023-07-14 | 西南交通大学 | Variable speed limit optimization method based on intelligent network coupling and driver compliance |
| CN116432448B (en) * | 2023-04-06 | 2024-02-23 | 西南交通大学 | Variable speed limit optimization method based on intelligent network coupling and driver compliance |
| CN117351708A (en) * | 2023-10-08 | 2024-01-05 | 北京迈道科技有限公司 | Expressway safety operation management early warning method, system and storage medium |
| CN117172424A (en) * | 2023-10-31 | 2023-12-05 | 江苏科运智慧交通科技有限公司 | Method for enhancing transmission effectiveness of road safety warning information |
| CN117172424B (en) * | 2023-10-31 | 2024-04-09 | 苏交科集团股份有限公司 | Method for enhancing transmission effectiveness of road safety warning information |
| CN117194930A (en) * | 2023-11-07 | 2023-12-08 | 陕西交通电子工程科技有限公司 | Tunnel road section safety monitoring method based on vehicle driving data |
| CN117194930B (en) * | 2023-11-07 | 2024-01-19 | 陕西交通电子工程科技有限公司 | Tunnel road section safety monitoring method based on vehicle driving data |
| CN117727183A (en) * | 2024-02-18 | 2024-03-19 | 南京淼瀛科技有限公司 | Automatic driving safety early warning method and system combining vehicle-road cooperation |
| CN117727183B (en) * | 2024-02-18 | 2024-05-17 | 南京淼瀛科技有限公司 | Automatic driving safety early warning method and system combining vehicle-road cooperation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114118795A (en) | Safety risk degree evaluation grading and dynamic early warning method for intelligent heavy-load expressway | |
| CN109415043B (en) | Automatic driving planning method based on comfort level | |
| CN105699095A (en) | A test method for a remotely piloted vehicle (PRV) | |
| Jung et al. | Modeling highway safety and simulation in rainy weather | |
| CN113065685B (en) | High-precision map traffic rule model based on automatic driving application scene and implementation method thereof | |
| Savolainen et al. | Safety impacts at intersections on curved segments | |
| Al-Ahmadi et al. | Calibrating the highway safety manual predictive models for multilane rural highway segments in Saudi Arabia | |
| Koloushani et al. | Mobile mapping system-based methodology to perform automated road safety audits to improve horizontal curve safety on rural roadways | |
| CN116597690B (en) | Highway test scene generation method, equipment and medium for intelligent network-connected automobile | |
| Meocci et al. | Black boxes data for road pavement conditions monitoring: A case study in Florence | |
| Pike et al. | Pavement Markings—Wet Retroreflectivity Standards | |
| Chrysler et al. | Driver response to delineation treatments on horizontal curves on two-lane roads. | |
| Subotić et al. | Models of Analysis of Credible Deviation from Speed Limits on Two-Lane Roads of Bosnia and Herzegovina | |
| CN117238139B (en) | Real-time road condition early warning system based on meteorological data | |
| Najm et al. | Definition of the light vehicle off-roadway crash problem for the intelligent vehicle initiative | |
| Storsæter | Designing and Maintaining Roads to Facilitate Automated Driving | |
| CN115223118B (en) | High-precision map confidence judging method, system and vehicle | |
| Pasias et al. | Vehicle Data Collection for Predicting Driving Behavior on Interchanges | |
| CN117831286A (en) | Road curve road safety evaluation and prediction method based on gyroscope data | |
| Kuciemba | Connected Roadway Classification System Development | |
| Abufares et al. | Sustainable Civil Infrastructures Utilization and Regulations of Innovative Technology to Improve Road Safety via Drivers’ Warnings and Enforcement | |
| Bartlett | The Impact of Inclement Weather on Transportation Systems and the Future of Connected and Autonomous Vehicles | |
| Stewart | The ability of road infrastructure in New Brunswick to support Lane Keep Assist technology | |
| Theiss et al. | Evaluation of warning lights on maintenance of traffic devices and development of possible alternatives. | |
| Ahmed | Effect of Improving the Visibility of the Traffic Path on Vehicles Speeds and Safety at Work Zones |
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 |