CN112232682A - Method and system for determining and grading environmental risk degree of automatic driving open test road - Google Patents
Method and system for determining and grading environmental risk degree of automatic driving open test road Download PDFInfo
- Publication number
- CN112232682A CN112232682A CN202011122235.XA CN202011122235A CN112232682A CN 112232682 A CN112232682 A CN 112232682A CN 202011122235 A CN202011122235 A CN 202011122235A CN 112232682 A CN112232682 A CN 112232682A
- Authority
- CN
- China
- Prior art keywords
- accident
- risk
- road
- data
- risk degree
- 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
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000012360 testing method Methods 0.000 title claims abstract description 54
- 230000007613 environmental effect Effects 0.000 title claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 131
- 206010039203 Road traffic accident Diseases 0.000 claims description 52
- 238000000926 separation method Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 7
- 238000003064 k means clustering Methods 0.000 claims description 3
- 238000012502 risk assessment Methods 0.000 abstract description 5
- 238000011835 investigation Methods 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000013077 scoring method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06Q10/067—Enterprise or organisation modelling
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method and a system for determining and grading the risk degree of an automatic driving open test road environment, relating to the technical field of risk assessment and management of road traffic safety, and constructing various accident risk degree calculation models according to historical manual driving road accident data; correcting the calculation model of the risk degree of each type of accident according to the vehicle accident data, and calculating the risk degree of each type of accident by using the corrected calculation model of the risk degree of each type of accident according to the road environment data and the traffic flow data; and determining the safety risk degree of the road environment according to the risk degree of each type of accident. By considering risk causes of various accidents, scientific calculation of the environmental risk of the open test road for automatic driving is realized.
Description
Technical Field
The invention relates to the technical field of road traffic safety risk assessment and management, in particular to a method and a system for determining and grading an automatic driving open test road environment risk degree.
Background
At present, the rapid development of artificial intelligence, big data, communication technology and the like makes the automatic driving automobile technology highly advanced. Open road testing is an indispensable important link for developing research and development and application of automatic driving technology. In order to ensure that the vehicle can safely, reliably and efficiently run under various road traffic conditions and use scenes, the automatic driving function needs to be subjected to a large number of tests and verifications and goes through a complex evolution process. Before the automatic driving automobile is formally put on the market, the automatic driving function of the automatic driving automobile must be fully tested in a real traffic environment, and the automatic driving function of the automatic driving automobile must be comprehensively verified to coordinate with roads, facilities and other traffic participants.
However, although many developed countries have earlier conducted automated driving open road test work including various scenes, accidents in automated driving tests have been frequent in recent years, and people have been doubted about the safety of automated driving technique tests. In order to reduce the risk possibly brought by automatic driving, the most basic method is to focus on a test road on which an automatic driving automobile runs, perform safety risk assessment on the test road, and judge whether the risk is suitable for carrying out automatic driving test work.
Urban road networks in most cities are built basically, and if a large-area newly-built special road is neither economical nor practical by aiming at an automatic-driving automobile open road test, the test should be carried out after an existing road is reformed (road traffic facilities are intelligentized). The existing road facilities (such as road flatness, traffic flow, whether there is a pedestrian protection facility, etc.) of different roads are very different, and the risks generated by the road environments themselves are also different, so that it is necessary to evaluate and grade the risks based on the road facilities and the current situation of the traffic environments.
The road traffic safety risk assessment link is basically mature as a key part in traffic safety management. At present, the safety evaluation methods for roads at home and abroad mainly comprise the following methods:
(1) accident rate method. The accident rate is a relative index and has strong comparability. The accident rate method comprises a single accident rate method (such as a region accident rate, a road section accident rate, a site accident rate and the like) and a comprehensive accident rate method (such as comprehensive accident intensity, equivalent mortality and the like). The simple accident rate can only reflect one side of the road traffic safety, the false image of one side often appears, and the real result comprehensively expressed by multiple factors cannot be reflected. The equivalent weight or the conversion coefficient involved in the comprehensive accident rate method are influenced by subjective understanding and artificial judgment to different degrees, and the rationality of the method is greatly controversial. The method considers the corresponding relation between the accident number and the flow number, has reasonable indexes, and has the defect that misevaluation is easy to cause due to the sporadic nature of traffic accidents. For a long time, China and most countries in the world adopt a traffic safety evaluation system based on traffic accident statistics, such as an accident rate coefficient method, an accident probability method, a road section accident rate and the like.
(2) And (4) modeling. The method is characterized in that a quantitative functional relation model between an accident and various main influence factors is established by analyzing the relation between the traffic accident and the influence factors. It is divided into two categories, one is a statistical analysis model and the other is an empirical model. The statistical analysis model needs a large amount of statistical data, and foreign research and application are more; although the empirical model is practical, the scientific basis is not sufficient, and the empirical model is limited by regions and traffic conditions, has poor comparability and is widely applied in China.
(3) And (4) a system analysis method. The method quantitatively describes the influence of the change of the system structure and various policies on the system behavior through the simulation analysis of the system, thereby establishing a traffic system accident occurrence rule model which mainly comprises an analytic hierarchy process and a fuzzy mathematical process. The weight average in the two methods is based on human judgment as a premise, and because the recognition degree of each person on each factor is different, the judgment is easy to cause larger difference, and the influence of human factors is great.
(4) An artificial neural network evaluation method. The method selects learning samples for all factors of traffic safety, establishes a neural network model through sample training, and then judges whether an event occurs by using the model; and if the event is continuously judged to occur and the occurrence frequency reaches a preset threshold value, warning is provided. The artificial neural network evaluation method can provide a learning sample to enable the neural network to automatically learn only by selecting a learning sample data format and certain parameters during artificial neural network learning, but the artificial neural network does not clarify the practical significance of each parameter of the model, is not beneficial to control and has large requirements on data quantity.
At present, automatic driving open road tests are gradually developed in various countries, and the risk factors of accidents in the automatic driving environment are different from the traditional manual driving accidents. In addition, accident data are few at the present stage of the automatic driving, which is not enough to meet the data volume requirement of the statistical analysis of the automatic driving traffic accidents.
Disclosure of Invention
The invention aims to provide a method and a system for determining and grading the environmental risk degree of an automatic driving open test road so as to realize scientific calculation of the environmental risk of the automatic driving open test road.
In order to achieve the purpose, the invention provides the following scheme:
a method for determining the environmental risk degree of an automatic driving open test road comprises the following steps:
determining a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road accident risk cause according to historical manual driving road accident data, wherein the road traffic accident risk cause is used for determining the influence factor of the road accident type;
constructing various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the road traffic accident risk cause attribute value;
acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
according to the accident data of the automatic driving vehicle and the accident data of the manual driving vehicle, correcting the calculation models of the risk degrees of the various types of accidents to obtain the influence coefficient of the attribute value of the risk cause of the corrected accident; the method specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
updating the calculation models of the risk degrees of the various types of accidents according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected calculation models of the risk degrees of the various types of accidents, wherein the corrected calculation models of the risk degrees of the various types of accidents are calculation models of the risk degrees of the various types of accidents of the automatic driving vehicles;
calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
and determining the safety risk degree of the road environment according to the risk degrees of the various types of accidents.
Optionally, the types of accidents include: the accident of lane departure, the accident of collision against opposite motor vehicles out of control, the accident of collision against opposite motor vehicles during overtaking, the accident of intersection and the accident of road access entrance;
the calculation model of the risk degree of each type of accident is as follows:
the risk degree of each type of accident is the accident occurrence probability x the accident severity x the road section vehicle traffic speed influence coefficient x the road section traffic flow influence coefficient x the central separation zone type influence coefficient x the weather environment influence coefficient x the traffic composition influence coefficient.
Optionally, the accident occurrence probability and the accident severity are obtained by multiplying the influence coefficients of the road setting factor attribute values in the accident risk cause attribute values;
determining a vehicle passing speed influence coefficient in a road section according to the vehicle passing speed in the road section;
determining the road section traffic flow influence coefficient according to the road section flow;
determining the type influence coefficient of the central separation zone according to the type of the central separation zone;
and determining the influence coefficient of the traffic composition according to the proportion of the bus in the traffic composition.
Optionally, the determining the road environment safety risk according to the various types of accident risk specifically includes:
the road environment safety risk degree comprises a road section safety risk degree, a road channel risk degree and a road network risk degree;
calculating the safety risk degree of the road section according to the safety risk degree of the formula road section, namely the safety risk degree of each type of the accidents;
according to the formula
Calculating the road channel risk degree;
according to the formula
And calculating the risk degree of the road network.
An automatic driving open test road environment risk degree determination system, comprising:
the accident risk cause determining module is used for determining road traffic accident risk causes and influence coefficients corresponding to the attribute values of the road traffic accident risk causes according to historical manual driving road accident data, and the road traffic accident risk causes are used for determining influence factors of road accident types;
the risk degree calculation model building module is used for building various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the accident risk cause attribute value;
the vehicle accident data acquisition module is used for acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
the risk degree calculation model correction module is used for correcting the various types of accident risk degree calculation models according to the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain the influence coefficient of the corrected accident risk cause attribute value, and specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
the risk degree calculation model updating model is used for updating various types of accident risk degree calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected various types of accident risk degree calculation models, and the corrected various types of accident risk degree calculation models are various types of accident risk degree calculation models of the automatic driving vehicle;
the accident risk degree calculation module is used for calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
and the road environment safety risk degree determining module is used for determining the road environment safety risk degree according to the various types of accident risk degrees.
An automatic driving open test road environment risk grading method comprises the following steps:
determining a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road accident risk cause according to historical manual driving road accident data, wherein the road traffic accident risk cause is used for determining the influence factor of the type of the road accident;
constructing various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the road traffic accident risk cause attribute value;
acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
according to the accident data of the automatic driving vehicle and the accident data of the manual driving vehicle, correcting the calculation models of the risk degrees of the various types of accidents to obtain the influence coefficient of the attribute value of the risk cause of the corrected accident; the method specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
updating various types of accident risk calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected various types of accident risk calculation models, wherein the corrected various types of accident risk calculation models are various types of risk calculation models of the automatic driving vehicle;
calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
obtaining a road environment grade division standard by adopting a clustering method according to the accident risk degrees of all types and the preset road environment safety risk degree;
and determining the road environment safety risk degree grade according to the road environment grade division standard and the various types of accident risk degrees.
Optionally, the obtaining of the road environment grade division standard by using a clustering method according to the accident risk degrees of each type and the preset road environment safety risk degree specifically includes:
and according to the risk degrees of the various types of accidents and the preset road environment safety risk degree, taking the road environment safety risk degree value as a clustering object, and obtaining a road environment grade division standard by adopting a K-Means clustering method.
An automatic driving open test road environment risk classification system, comprising:
the accident risk cause determining module is used for determining road traffic accident risk causes and influence coefficients corresponding to the attribute values of the road traffic accident risk causes according to historical manual driving road accident data, and the road traffic accident risk causes are used for determining influence factors of road accident types;
the risk degree calculation model building module is used for building various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the accident risk cause attribute value;
the vehicle accident data acquisition module is used for acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
the risk degree calculation model correction module is used for correcting the various types of accident risk degree calculation models according to the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain the influence coefficient of the corrected accident risk cause attribute value, and specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
the risk degree calculation model updating model is used for updating various types of accident risk degree calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected various types of accident risk degree calculation models, and the corrected various types of accident risk degree calculation models are various types of accident risk degree calculation models of the automatic driving vehicle;
the accident risk degree calculation module is used for calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
the division standard determining module is used for obtaining a road environment grade division standard by adopting a clustering method according to the accident risk degrees of all types and the preset road environment safety risk degree;
and the risk degree grade determining module is used for determining the road environment safety risk degree grade according to the road environment grade division standard and the road environment safety risk degree.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a method and a system for determining the environmental risk degree of an automatic driving open test road, which are used for constructing various accident risk degree calculation models according to historical manual driving road accident data; and correcting the risk degree calculation models of the various types of accidents according to the accident data of the manually driven vehicles and the accident data of the automatically driven vehicles, and calculating the risk degree of the various types of accidents by using the corrected risk degree calculation models of the various types of accidents, so that the calculation of the safety risk degree of the road environment is realized when the data of the automatically driven accidents are less. And the scientificity of the road environment safety risk degree is improved by considering the risk cause of various accidents.
The invention also provides a grading method and a grading system for the environmental risk of the automatic driving open test road, which are used for constructing various accident risk calculation models according to the accident data of the manual driving road; and correcting the calculation models of the risk degrees of the various types of accidents according to the accident data of the manually driven vehicles and the accident data of the automatically driven vehicles, and calculating the risk degrees of the various types of accidents by using the corrected calculation models of the risk degrees of the various types of accidents. And clustering the accident risk degrees of various types and the preset road environment safety risk degree to obtain a road environment division standard, and realizing the division of the road environment safety risk degree grades.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described 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 without inventive exercise.
FIG. 1 is a flow chart of a method for determining the risk of an open test road environment for automatic driving according to the present invention;
FIG. 2 is a schematic diagram of an automatic driving open test road environment risk level determination system according to the present invention;
FIG. 3 is a flowchart of an environment risk classification method for an automatic driving open test road according to the present invention;
FIG. 4 is a schematic diagram of an automatic driving open test road environment risk classification system according to the present invention;
FIG. 5 is a schematic diagram of the method for grading the environmental risk of the open test road for automatic driving according to the present invention;
FIG. 6 is a schematic view of the research range of the open test road environment for automatic driving according to the present invention;
FIG. 7 is a schematic representation of a Bayesian network of the present invention;
FIG. 8 is a schematic diagram of a road segment division result according to the present invention;
FIG. 9 is a diagram illustrating the environment classification result according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the 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 other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a method and a system for determining and grading the environmental risk degree of an automatic driving open test road so as to realize scientific calculation of the environmental risk of the automatic driving open test road.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the present invention provides a method for determining an environmental risk level of an automatic driving open test road, including:
step 101: and determining a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road accident risk cause according to historical manual driving road accident data, wherein the road traffic accident risk cause is used for determining the influence factor of the road accident type.
Step 102: and constructing various accident risk degree calculation models according to the influence coefficients corresponding to the road traffic accident risk cause and the attribute values of the road traffic accident risk cause.
Step 103: and acquiring accident data of the automatic driving vehicle and accident data of the manual driving vehicle.
Step 104: according to the accident data of the automatic driving vehicle and the accident data of the manual driving vehicle, correcting the calculation models of the risk degrees of various types of accidents to obtain the influence coefficient of the attribute value of the risk cause of the corrected accident; the method specifically comprises the following steps: inputting the vehicle accident data of the automatic driving vehicle and the manual driving vehicle accident data into a Bayesian network, fitting the vehicle accident data of the automatic driving vehicle and the manual driving vehicle accident data by the Bayesian network to obtain a vehicle accident data fitting value, and outputting an influence coefficient of an accident risk cause attribute value after correction by the Bayesian network when the absolute value of the difference between the vehicle accident data fitting value and the actual value of the accident data is less than or equal to a set threshold value; .
Step 105: and updating the calculation models of the risk degrees of the various types of accidents according to the influence coefficients of the corrected attribute values of the risks of the accidents to obtain the corrected calculation models of the risk degrees of the various types of accidents, wherein the corrected calculation models of the risk degrees of the various types of accidents are the calculation models of the risk degrees of the various types of automatic driving vehicles.
Step 106: and calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data.
Step 107: and determining the safety risk degree of the road environment according to the risk degree of each type of accident.
Wherein, each type of accident comprises: the accident of lane departure, the accident of collision against opposite vehicles out of control, the accident of collision against opposite vehicles in overtaking, the accident of intersection and the accident of road access entrance. The calculation model of the risk degree of each type of accident is as follows:
the risk degree of each type of accident is the accident occurrence probability x the accident severity x the road section vehicle traffic speed influence coefficient x the road section traffic flow influence coefficient x the central separation zone type influence coefficient x the weather environment influence coefficient x the traffic composition influence coefficient.
The accident occurrence probability and the accident severity are determined according to the road environment data and the traffic flow data, and the risk degree of each type of accident is calculated according to the accident occurrence probability and the accident severity.
In practical application, the accident occurrence probability and the accident severity are obtained by multiplying the influence coefficients of the road setting factor attribute values in the accident risk cause attribute values.
And determining the influence coefficient of the vehicle passing speed in the road section according to the vehicle passing speed in the road section.
And determining a road section traffic flow influence coefficient according to the road section flow.
And determining the influence coefficient of the type of the central division belt according to the type of the central division belt.
And determining the influence coefficient of the traffic composition according to the proportion of the bus in the traffic composition.
Wherein, step 107 specifically comprises:
the road environment safety risk degree comprises a road section safety risk degree, a road channel risk degree and a road network risk degree.
Calculating the safety risk degree of the road section according to the safety risk degree of the formula road section, namely the safety risk degree of each type of the accidents,
according to the formula
And calculating the road channel risk degree.
According to the formula
And calculating the risk degree of the road network.
As shown in fig. 2, the present invention provides an automatic driving opening test road environment risk degree determination system, which includes:
the accident risk cause determining module 201 is configured to determine, according to historical manual driving road accident data, a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road traffic accident risk cause, where the road traffic accident risk cause is used to determine an influence factor of a road accident type.
And the risk degree calculation model construction module 202 is used for constructing various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the accident risk cause attribute value.
And the vehicle accident data acquisition module 203 is used for acquiring the automatic driving vehicle accident data and the manual driving vehicle accident data.
The risk degree calculation model modification module 204 is configured to modify each type of accident risk degree calculation model according to the vehicle accident data of the autonomous driving vehicle and the vehicle accident data of the manual driving vehicle, so as to obtain an influence coefficient of an attribute value of a corrected accident risk cause, and specifically includes: and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs the influence coefficient of the corrected accident risk cause attribute value.
And the risk degree calculation model updating model 205 is used for updating each type of accident risk degree calculation model according to the influence coefficient of the corrected accident risk cause attribute value to obtain each type of corrected accident risk degree calculation model, and each type of corrected accident risk degree calculation model is each type of risk degree calculation model of the automatic driving vehicle.
And the accident risk degree calculation module 206 is used for calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data.
And the road environment safety risk degree determining module 207 is used for determining the road environment safety risk degree according to the various types of accident risk degrees.
As shown in fig. 3, the method for classifying the environmental risk of the automatic driving open test road provided by the present invention includes:
step 301: and determining a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road accident risk cause according to historical manual driving road accident data, wherein the road traffic accident risk cause is used for determining the influence factor of the road accident type.
Step 302: and constructing various accident risk degree calculation models according to the influence coefficients corresponding to the road traffic accident risk cause and the attribute values of the road traffic accident risk cause.
Step 303: and acquiring accident data of the automatic driving vehicle and accident data of the manual driving vehicle.
Step 304: according to the accident data of the automatic driving vehicle and the accident data of the manual driving vehicle, correcting the calculation models of the risk degrees of various types of accidents to obtain the influence coefficient of the attribute value of the risk cause of the corrected accident; the method specifically comprises the following steps: and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs the influence coefficient of the corrected accident risk cause attribute value.
Step 305: and updating the calculation models of the risk degrees of the various types of accidents according to the influence coefficients of the corrected attribute values of the risks of the accidents to obtain the corrected calculation models of the risk degrees of the various types of accidents, wherein the corrected calculation models of the risk degrees of the various types of accidents are the calculation models of the risk degrees of the various types of accidents of the automatic driving vehicles.
Step 306: and calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data.
Step 307: and obtaining a road environment grade division standard by adopting a clustering method according to the accident risk degrees of various types and the preset road environment safety risk degree.
Step 308: and determining the road environment safety risk degree grade according to the road environment grade division standard and the various types of accident risk degrees.
Wherein, step 307 specifically includes:
and according to the risk degree of each type of accident and the preset road environment safety risk degree, taking the road environment safety risk degree value as a clustering object, and obtaining a road environment grade division standard by adopting a K-Means clustering method.
As shown in fig. 4, the present invention provides an automatic driving open test road environment risk classification system, which includes:
the accident risk cause determining module 401 is configured to determine, according to historical manual driving road accident data, a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road traffic accident risk cause, where the road traffic accident risk cause is used to determine an influence factor of a road accident type.
And a risk degree calculation model construction module 402, configured to construct various types of accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the accident risk cause attribute value.
A vehicle accident data acquisition module 403 for acquiring accident data of an autonomous vehicle and accident data of a manually driven vehicle
The risk degree calculation model modification module 404 is configured to modify each type of accident risk degree calculation model according to the vehicle accident data of the autonomous driving vehicle and the vehicle accident data of the manual driving vehicle, so as to obtain an influence coefficient of an attribute value of a corrected accident risk cause, and specifically includes: and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs the influence coefficient of the corrected accident risk cause attribute value.
And the risk degree calculation model updating model 405 is used for updating the various types of accident risk degree calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain the corrected various types of accident risk degree calculation models, and the corrected various types of accident risk degree calculation models are various types of risk degree calculation models of the automatic driving vehicle.
And an accident risk degree calculation module 406 for calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data.
And the division standard determining module 407 is configured to obtain a road environment grade division standard by using a clustering method according to the accident risk degrees of the various types and the preset road environment safety risk degree.
And a risk level determining module 408, configured to determine a road environment safety risk level according to the road environment level division standard and the road environment safety risk level.
In addition, the invention also provides a specific mode of the method for grading the environmental risk of the open test road for automatic driving, as shown in fig. 5, the method comprises the following steps:
1) and analyzing the types of the potential road accidents, extracting accident risk causes for the automatic driving vehicles, and constructing various accident risk degree calculation models.
11) Five types of accidents which may occur when the motor vehicle runs on the road are analyzed, which are respectively as follows: the accident of lane departure, the accident of collision against opposite vehicles in case of out of control, the accident of collision against opposite vehicles in case of overtaking, the accident of intersection and the accident of road access entrance. The risk causes are shown in table 1.
TABLE 1 Risk causation Table
And for different accident types, the related risk cause of each accident is determined by combining the accident characteristics.
If the influence factors of the lane departure risk are lane width, curvature, marking and marking conditions, road surface flatness, gradient, expected speed, suggested vehicle speed, traffic flow and the like, the influence of intersection accidents does not need to be considered when the non-intersection accidents happen.
12) Building various accident risk degree calculation models based on various risk cause attribute values, which specifically comprises the following steps:
the risk degree of each type of accident is the accident probability, the accident severity, the road section vehicle traffic speed influence coefficient, the road section traffic flow influence coefficient, the central separation zone type influence coefficient, the weather environment influence coefficient and the traffic composition influence coefficient
In the formula, the accident occurrence probability and the accident severity are obtained by carrying out cumulative multiplication on influence coefficients corresponding to attribute values of road facility factors related to accidents.
The influence coefficient of the vehicle passing speed in the road section is determined by the vehicle passing speed in the road section, when the average passing speed of the vehicles in the road section exceeds 120km/h, the influence coefficient is 1, and the influence coefficient is reduced along with the reduction of the average passing speed.
The link traffic flow influence coefficient is determined by the link flow, and the influence coefficient increases as the link flow increases.
The center separator type influence coefficient is related to the type (traversable/non-traversable) of the center separator.
The influence coefficient of the traffic composition mainly considers the influence of the ratio of the large vehicles on the risk degree, when the ratio of the large vehicles is less than 10%, the influence coefficient is 1, and the influence coefficient is increased along with the increase of the ratio of the large vehicles.
Weather influence coefficient: 1 in good weather and more than 1 in bad weather, and the influence coefficient is increased along with the increase of the bad weather; the central separation zone type influence coefficient is only used for calculating part of the accident risk degree.
And determining the environmental risk degree of the automatic driving open test road through calculating the risk degree of each type of accident.
2) Checking and correcting the influence coefficients of the attribute values of the risk factors in the model based on the traditional manual driving vehicle accident data and the limited automatic driving accident data; and constructing a Bayesian network according to the correlation of the risk causes. And inputting automatic driving accident data and manual driving accident data into the network for model training and learning, wherein the accident data comprises accident types of all accidents, corresponding risk cause attribute values and influence coefficients of the risk cause attribute values. Checking and correcting the influence coefficients of all risk causes in the model according to the Bayesian network parameter learning result, and outputting the corrected influence coefficients of all risk causes by the Bayesian network;
3) calculating road environment safety risk degree based on road environment data and traffic flow data collected by field investigation;
31) investigation of road infrastructure factors: the conditions of all the influence factors are recorded in detail by investigating main static factors influencing road safety on the spot, such as the number of lanes, the road flatness, the type of a middle isolation belt, the curvature, the gradient, the road access point, the intersection condition and the like, and the calculation of subsequent risk degree is facilitated by compiling an influence factor recording table.
32) Road section division: dividing the roads into the same road section by using the principle of continuous roads with the same or similar factors investigated in 31) and numbering the roads.
33) Traffic factor survey: the vehicle passing speed and the traffic flow are important dynamic factors influencing the road safety risk, the vehicle passing speed can be determined according to the speed limit condition of the road section, and the traffic flow needs to investigate each divided road section to obtain the average daily traffic volume and the traffic vehicle type proportion of each road section.
The data quality should be strictly controlled in the three stages of design, implementation and result statistics in the above investigation process. The control content comprises data integrity, data authenticity, human errors, data quality evaluation indexes and the like. Wherein, the traffic survey and traffic survey of road users should ensure that the survey time covers at least one hour each of the early peak, the late peak and the average peak; the road facility survey should ensure that the survey coverage is both sides of the road and can not be interrupted or omitted; road environment surveys should ensure the simultaneity of the survey. Data obtained by automatic acquisition means with reliable quality, such as videos, coils and the like, are preferentially adopted for data investigation, corresponding data are lacked, and field investigation needs to be carried out. All on-site data acquisition must be carried out by more than three investigators trained correspondingly, so as to ensure the reliability of investigation results. The survey results should be averaged.
34) The road environment safety risk degree calculation formula is as follows:
road section safety risk degree ═ sigma various accident risk degree (2)
The method for calculating the road channel risk degree is shown in formula (3):
wherein n is the number of road segments in the road channel.
The road network risk degree calculation method is shown in formula (4):
wherein n is the number of road segments in the road network, and m is the number of road channels in the road network.
4) And collecting data by adopting an expert scoring method, determining a road environment grade division standard through clustering, and determining a road environment grade based on the road environment safety risk degree.
41) The method comprises the steps that a questionnaire survey collection expert is designed, the rating condition of the questionnaire survey collection expert on the road environment is used for determining a road environment rating threshold, the questionnaire content comprises multiple groups of pictures and text descriptions of the road environment, the expert performs rating by a four-level rating method, and a rating result (low risk/general risk/higher risk/high risk) is given to each group of road environment by combining the road environment information described by the questionnaire.
42) Calculating the risk degree of the road environment in the questionnaire according to the model updated in the step 2), determining a risk degree grading threshold value by taking the obtained risk degree value as a clustering object by combining the grading results of all experts on the road environment and adopting a K-Means method, and further determining the risk degree range corresponding to each road environment grade.
43) And matching the calculated safety risk degree of each road into a road environment grade division standard to obtain the road environment risk grade corresponding to each road section. The road environment is divided into four levels, namely low risk (road environment type I), general risk (road environment type II), higher risk (road environment type III), high risk (road environment type IV).
Road environment classification is carried out by taking 11.1 kilometers in total of jayu south road-ann way-north ann de road-ann intelligence road-Boyuan road-ann rainbow road-ann topology road in Jiading area of Shanghai city under a clear weather state as research objects, the research range is shown in figure 6, and the specific implementation steps are as follows:
the method comprises the following steps: based on the five accident types, the clear risk causes are shown in table 1.
Step two: and (6) correcting the model. A bayesian network is constructed according to the correlation of various risk causes as shown in fig. 7, wherein the automatic driving accident data is from an automatic driving automobile operation accident Report database established by the united states california motor vehicle administration (DMV), and the manual driving accident data is from an accident Report Sampling system established by the united states highway traffic safety administration, which is abbreviated as crss (the blast Report Sampling system). And determining the influence coefficient of each risk cause by referring to the existing traditional manual driving accident literature. The results of the correction based on the results of the bayesian network parameter learning are shown in table 2.
TABLE 2 correction of risk cause influence coefficient results table
Step three: and (5) carrying out investigation according to the factors determined in the step one, dividing the evaluation road into 53 road sections according to the investigation result, and displaying the road section result in fig. 8, wherein each road section is alternately represented by two colors of black and gray. And after the investigation is finished, calculating the road environmental safety risk degree of each road section according to the model parameters corrected in the step two.
Step four: based on questionnaire data, clustering risk values of road environments of all levels, determining that critical values between two adjacent levels are 3.5, 12.5 and 22.5 respectively, and determining the critical values as index grading points to grade road traffic safety risks. The environmental rating criteria defined based on the traffic safety risk assessment indicators are shown in table 3.
Table 3 environment registration division standard table
As shown in fig. 9, 40 road environment class i road segments, which account for 75%, exist in all 53 road segments to be evaluated; 10 road environment II-type road sections accounting for 19 percent; 3 road environment III road sections, accounting for 6%; and the road-free environment is IV-type road sections. Wherein, the horizontal line in the figure represents the road environment class I section, the dotted line represents the road environment class II section, the double horizontal line represents the road environment class III section, the unequal thick line type double horizontal line represents the road environment class IV section
The method and the system for determining and grading the environmental risk of the automatic driving open test road have the following obvious advantages that:
1. originality: the selection and management of the automatic driving open test road play an important role in safely and orderly developing the automatic driving open test, and the significance in researching the environment classification of the automatic driving open test road is great. The existing automatic driving test results show that the cause of the automatic driving vehicle accident is greatly different from the traditional manual driving vehicle accident, and the road facilities under the automatic driving environment are also greatly changed, so that a calculation method for more comprehensively analyzing the risk cause and the risk degree of the automatic driving accident is needed.
2. Integrity: the invention relates to a complete road environment risk degree grading method for an automatic driving open test, which comprises accident type determination, influence factor selection, factor investigation mode, risk degree calculation method and risk grading method.
3. Scientifically: the method fully considers the characteristics of the road environment required by automatic driving and the automatic driving accidents, determines the potential accident types and the accident risk causes, and checks and corrects the influence coefficients of the causes through the Bayesian network based on the actual automatic driving opening test result. And a reasonable road environment grade division standard facing the automatic driving opening test is determined by adopting an expert scoring method.
4. The practicability is as follows: the method is comprehensive in consideration and universal, and is suitable for environment classification of various types of roads including expressways, urban roads, rural roads and the like.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
1. A method for determining the environmental risk degree of an automatic driving open test road is characterized by comprising the following steps:
determining a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road accident risk cause according to historical manual driving road accident data, wherein the road traffic accident risk cause is used for determining the influence factor of the type of the road accident;
constructing various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the road traffic accident risk cause attribute value;
acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
according to the accident data of the automatic driving vehicle and the accident data of the manual driving vehicle, correcting the calculation models of the risk degrees of the various types of accidents to obtain the influence coefficient of the attribute value of the risk cause of the corrected accident; the method specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
updating the calculation models of the risk degrees of the various types of accidents according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected calculation models of the risk degrees of the various types of accidents, wherein the corrected calculation models of the risk degrees of the various types of accidents are calculation models of the risk degrees of the various types of accidents of the automatic driving vehicles;
calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
and determining the safety risk degree of the road environment according to the risk degrees of the various types of accidents.
2. The automated driving open test road environment determination method according to claim 1, wherein each type of accident includes: the accident of lane departure, the accident of collision against opposite motor vehicles out of control, the accident of collision against opposite motor vehicles during overtaking, the accident of intersection and the accident of road access entrance;
the calculation model of the risk degree of each type of accident is as follows:
the risk degree of each type of accident is the accident occurrence probability x the accident severity x the road section vehicle traffic speed influence coefficient x the road section traffic flow influence coefficient x the central separation zone type influence coefficient x the weather environment influence coefficient x the traffic composition influence coefficient.
3. The automated driving open test road environment determination method according to claim 2,
accumulating and multiplying according to influence coefficients of the attribute values of the road setting factors in the attribute values of the accident risk cause to obtain the accident occurrence probability and the accident severity;
determining a vehicle passing speed influence coefficient in a road section according to the vehicle passing speed in the road section;
determining the road section traffic flow influence coefficient according to the road section flow;
determining the type influence coefficient of the central separation zone according to the type of the central separation zone;
and determining the influence coefficient of the traffic composition according to the proportion of the bus in the traffic composition.
4. The method for determining the road environment for the automatic driving opening test according to claim 1, wherein the determining the road environment safety risk degree according to the various types of accident risk degrees specifically comprises:
the road environment safety risk degree comprises a road section safety risk degree, a road channel risk degree and a road network risk degree;
calculating the safety risk degree of the road section according to the safety risk degree of the formula road section, namely the safety risk degree of each type of the accidents;
according to the formula
Calculating the road channel risk degree;
according to the formula
And calculating the risk degree of the road network.
5. An automatic driving open test road environment risk degree determination system, characterized by comprising:
the accident risk cause determining module is used for determining road traffic accident risk causes and influence coefficients corresponding to the attribute values of the road traffic accident risk causes according to historical manual driving road accident data, and the road traffic accident risk causes are used for determining influence factors of road accident types;
the risk degree calculation model building module is used for building various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the accident risk cause attribute value;
the vehicle accident data acquisition module is used for acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
the risk degree calculation model correction module is used for correcting the various types of accident risk degree calculation models according to the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain the influence coefficient of the corrected accident risk cause attribute value, and specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
the risk degree calculation model updating model is used for updating various types of accident risk degree calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected various types of accident risk degree calculation models, and the corrected various types of accident risk degree calculation models are various types of accident risk degree calculation models of the automatic driving vehicle;
the accident risk degree calculation module is used for calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
and the road environment safety risk degree determining module is used for determining the road environment safety risk degree according to the various types of accident risk degrees.
6. The method for grading the environmental risk of the automatic driving open test road is characterized by comprising the following steps of:
determining a road traffic accident risk cause and an influence coefficient corresponding to the attribute value of the road accident risk cause according to historical manual driving road accident data, wherein the road traffic accident risk cause is used for determining the influence factor of the type of the road accident;
constructing various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the road traffic accident risk cause attribute value;
acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
according to the accident data of the automatic driving vehicle and the accident data of the manual driving vehicle, correcting the calculation models of the risk degrees of the various types of accidents to obtain the influence coefficient of the attribute value of the risk cause of the corrected accident; the method specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
updating various types of accident risk calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected various types of accident risk calculation models, wherein the corrected various types of accident risk calculation models are various types of risk calculation models of the automatic driving vehicle;
calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
obtaining a road environment grade division standard by adopting a clustering method according to the accident risk degrees of all types and the preset road environment safety risk degree;
and determining the road environment safety risk degree grade according to the road environment grade division standard and the various types of accident risk degrees.
7. The method for grading the road environment risk level for the automatic driving opening test according to claim 6, wherein the step of obtaining the road environment grade division standard by using a clustering method according to the risk level of each type of accident and the preset road environment safety risk level specifically comprises the steps of:
and according to the risk degrees of the various types of accidents and the preset road environment safety risk degree, taking the road environment safety risk degree value as a clustering object, and obtaining a road environment grade division standard by adopting a K-Means clustering method.
8. An automatic driving open test road environment risk classification system is characterized by comprising:
the accident risk cause determining module is used for determining road traffic accident risk causes and influence coefficients corresponding to the attribute values of the road traffic accident risk causes according to historical manual driving road accident data, and the road traffic accident risk causes are used for determining influence factors of road accident types;
the risk degree calculation model building module is used for building various accident risk degree calculation models according to the road traffic accident risk cause and the influence coefficient corresponding to the accident risk cause attribute value;
the vehicle accident data acquisition module is used for acquiring accident data of an automatic driving vehicle and accident data of a manual driving vehicle;
the risk degree calculation model correction module is used for correcting the various types of accident risk degree calculation models according to the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain the influence coefficient of the corrected accident risk cause attribute value, and specifically comprises the following steps: inputting the automatic driving vehicle accident data and the manual driving vehicle accident data into a Bayesian network, wherein the Bayesian network fits the automatic driving vehicle accident data and the manual driving vehicle accident data to obtain a vehicle accident data fitting value, and when the absolute value of the difference between the vehicle accident data fitting value and the accident data actual value is less than or equal to a set threshold value, the Bayesian network outputs an influence coefficient of a corrected accident risk cause attribute value;
the risk degree calculation model updating model is used for updating various types of accident risk degree calculation models according to the influence coefficients of the corrected accident risk cause attribute values to obtain corrected various types of accident risk degree calculation models, and the corrected various types of accident risk degree calculation models are various types of accident risk degree calculation models of the automatic driving vehicle;
the accident risk degree calculation module is used for calculating the risk degree of each type of accident by using the corrected risk degree calculation model of each type of accident according to the road environment data and the traffic flow data;
the division standard determining module is used for obtaining a road environment grade division standard by adopting a clustering method according to the accident risk degrees of all types and the preset road environment safety risk degree;
and the risk degree grade determining module is used for determining the road environment safety risk degree grade according to the road environment grade division standard and the road environment safety risk degree.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2020107087228 | 2020-07-22 | ||
| CN202010708722.8A CN111667204A (en) | 2020-07-22 | 2020-07-22 | Method and system for determining and grading environmental risk degree of automatic driving open test road |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112232682A true CN112232682A (en) | 2021-01-15 |
Family
ID=72392887
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010708722.8A Pending CN111667204A (en) | 2020-07-22 | 2020-07-22 | Method and system for determining and grading environmental risk degree of automatic driving open test road |
| CN202011122235.XA Pending CN112232682A (en) | 2020-07-22 | 2020-10-20 | Method and system for determining and grading environmental risk degree of automatic driving open test road |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010708722.8A Pending CN111667204A (en) | 2020-07-22 | 2020-07-22 | Method and system for determining and grading environmental risk degree of automatic driving open test road |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN111667204A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114529088A (en) * | 2022-02-21 | 2022-05-24 | 山东大学 | Driving path planning method and system based on accident risk cost |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112419719B (en) * | 2020-11-18 | 2022-06-07 | 济南北方交通工程咨询监理有限公司 | Method and system for evaluating traffic operation safety of highway |
| CN113487863A (en) * | 2021-06-30 | 2021-10-08 | 中国科学技术大学先进技术研究院 | Multi-factor traffic risk real-time quantitative evaluation system and method |
| CN113887904A (en) * | 2021-09-23 | 2022-01-04 | 同济大学 | Freight vehicle driver risk grade assessment method based on analytic hierarchy process |
| CN114530040A (en) * | 2022-02-08 | 2022-05-24 | 东风汽车集团股份有限公司 | Method and system for handling and rescuing vehicle accidents on driverless road |
| CN114898557B (en) * | 2022-05-20 | 2023-05-23 | 中国电信股份有限公司 | Automatic driving road section testing method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020797A (en) * | 2019-03-27 | 2019-07-16 | 清华大学苏州汽车研究院（吴江） | The evaluation method of automatic Pilot test scene based on perception defect |
| CN110782125A (en) * | 2019-09-23 | 2020-02-11 | 同济大学 | Road safety risk degree evaluation method for automatically driving automobile |
| CN110956335A (en) * | 2019-12-10 | 2020-04-03 | 黑龙江省公路勘察设计院 | Road safety evaluation system and method based on forecast accident rate grading |
-
2020
- 2020-07-22 CN CN202010708722.8A patent/CN111667204A/en active Pending
- 2020-10-20 CN CN202011122235.XA patent/CN112232682A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020797A (en) * | 2019-03-27 | 2019-07-16 | 清华大学苏州汽车研究院（吴江） | The evaluation method of automatic Pilot test scene based on perception defect |
| CN110782125A (en) * | 2019-09-23 | 2020-02-11 | 同济大学 | Road safety risk degree evaluation method for automatically driving automobile |
| CN110956335A (en) * | 2019-12-10 | 2020-04-03 | 黑龙江省公路勘察设计院 | Road safety evaluation system and method based on forecast accident rate grading |
Non-Patent Citations (2)
| Title |
|---|
| 木又青: ""自动驾驶开放测试道路安全风险评估方法(SRAAV)，评估流程与调研清单、评估清单、相关名"", 《道客巴巴》 * |
| 资料之王的店: ""《自动驾驶开放测试道路环境划分标准》编制说明"", 《百度文库》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114529088A (en) * | 2022-02-21 | 2022-05-24 | 山东大学 | Driving path planning method and system based on accident risk cost |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111667204A (en) | 2020-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111667204A (en) | Method and system for determining and grading environmental risk degree of automatic driving open test road | |
| CN109977812B (en) | Vehicle-mounted video target detection method based on deep learning | |
| CN111311918B (en) | Traffic management method and device based on visual analysis | |
| CN114783183A (en) | Monitoring method and system based on traffic situation algorithm | |
| CN110675626B (en) | Traffic accident black point prediction method, device and medium based on multidimensional data | |
| CN116631186B (en) | Expressway traffic accident risk assessment method and system based on dangerous driving event data | |
| CN110782125A (en) | Road safety risk degree evaluation method for automatically driving automobile | |
| CN110674887A (en) | End-to-end road congestion detection algorithm based on video classification | |
| CN113450573A (en) | Traffic monitoring method and traffic monitoring system based on unmanned aerial vehicle image recognition | |
| CN115205965A (en) | Helmet detection method based on combination of improved YOLO _ X algorithm and attention module | |
| Zhao et al. | Factors affecting traffic risks on bridge sections of freeways based on partial dependence plots | |
| CN117238126A (en) | Traffic accident risk assessment method under continuous flow road scene | |
| CN112669596A (en) | Traffic safety situation distinguishing system and method based on big data | |
| CN116168356A (en) | Vehicle damage judging method based on computer vision | |
| CN111797188A (en) | Urban functional area quantitative identification method based on open-source geographic space vector data | |
| CN113642162B (en) | Urban road traffic emergency plan simulation comprehensive analysis method | |
| CN113392885B (en) | Traffic accident space-time hot spot distinguishing method based on random forest theory | |
| CN115587463A (en) | Automatic driving test scene design method based on scene classification | |
| CN117456342A (en) | Traffic accident risk assessment method and system based on street view image | |
| CN116946183A (en) | Commercial vehicle driving behavior prediction method considering driving capability and vehicle equipment | |
| CN110765900A (en) | DSSD-based automatic illegal building detection method and system | |
| CN114880852A (en) | Modeling analysis method and system based on social perception data | |
| CN115170783A (en) | Expressway pavement crack detection method using street view image | |
| CN115130895A (en) | Intersection risk studying and judging early warning system under urban traffic environment | |
| CN112801482B (en) | Data-driven non-motor vehicle riding suitability evaluation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210115 |