CN108304986A - A kind of assessment method of automatic driving vehicle behavior safety - Google Patents

A kind of assessment method of automatic driving vehicle behavior safety Download PDF

Info

Publication number
CN108304986A
CN108304986A CN201711348568.2A CN201711348568A CN108304986A CN 108304986 A CN108304986 A CN 108304986A CN 201711348568 A CN201711348568 A CN 201711348568A CN 108304986 A CN108304986 A CN 108304986A
Authority
CN
China
Prior art keywords
test
parameter
appraisal
automatic driving
operating mode
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201711348568.2A
Other languages
Chinese (zh)
Other versions
CN108304986B (en
Inventor
宋娟
何承坤
王荣
薛晓卿
刘大鹏
丁文龙
朱科屹
李京泰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Research Center Of Computer And Microelectronics Development Ministry Of Industry And Information Technology (china Software Testing Center)
Original Assignee
Research Center Of Computer And Microelectronics Development Ministry Of Industry And Information Technology (china Software Testing Center)
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Research Center Of Computer And Microelectronics Development Ministry Of Industry And Information Technology (china Software Testing Center) filed Critical Research Center Of Computer And Microelectronics Development Ministry Of Industry And Information Technology (china Software Testing Center)
Priority to CN201711348568.2A priority Critical patent/CN108304986B/en
Publication of CN108304986A publication Critical patent/CN108304986A/en
Application granted granted Critical
Publication of CN108304986B publication Critical patent/CN108304986B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M17/00—Testing of vehicles
    • G01M17/007—Wheeled or endless-tracked vehicles
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
    • G05B23/0213—Modular or universal configuration of the monitoring system, e.g. monitoring system having modules that may be combined to build monitoring program; monitoring system that can be applied to legacy systems; adaptable monitoring system; using different communication protocols

Abstract

The invention discloses a kind of assessment methods of automatic driving vehicle behavior safety, including：It obtains the environment sensing parameter of automatic driving vehicle, network link parameter, judge that decision parameters, control execute parameter and net connection degree coefficient；Parameter is executed according to the environment sensing parameter, the network link parameter, the judgement decision parameters, the control and the net joins degree coefficient, obtains the test and appraisal score of automatic driving vehicle behavior safety；According to the test and appraisal score of the automatic driving vehicle behavior safety, the evaluating result of the automatic driving vehicle behavior safety is obtained.The assessment method of the automatic driving vehicle behavior safety of the present invention can consider the behavior safety of automatic driving vehicle from multi-angle comprehensive, keep evaluating result accurate, reliable, a kind of possible ways are provided to test and evaluate for automatic driving vehicle behavior safety, keep associated test standards perfect, promotes the development of China's automatic driving vehicle industry.

Description

A kind of assessment method of automatic driving vehicle behavior safety
Technical field
The present invention relates to the driving behavior security evaluation technical fields of intelligent network connection automobile, autonomous driving vehicle, especially relate to And a kind of assessment method of automatic driving vehicle behavior safety.
Background technology
Intelligent network connection automobile refers to carrying the devices such as advanced onboard sensor, controller, actuator, and merge modern logical It is shared to realize that vehicle is exchanged with people, Che Yuche, Che Yulu, vehicle with intelligent information such as backstages, has complicated ring for letter and network technology The functions such as border perception, intelligent decision, Collaborative Control and execution can finally be replaced, it can be achieved that safe, comfortable, energy saving, efficient traveling The young mobile operated for people.
The intelligent network connection automotive engineering of current industrialization is mainly DAS (Driver Assistant System).Existing DAS (Driver Assistant System) Including：Before hit alarm system, lane-changing assistance system, automatic emergency brake system etc..It tests and assesses to these DAS (Driver Assistant System)s Core be individual event function standard compliance test and appraisal, relatively be easy carry out.And further, intelligent network joins automobile certainly It moves the complexity of control loop and no less than the complexity of automobile itself, the possibility of the person and property safety is endangered It is endangered caused by possible no less than automobile itself.Existing Te st grogram and the test and appraisal demand phase for ensureing upper road vehicle safety Than, integrity and systematicness be not sufficient to ensure and support civilian automated driving system can real roads environment (especially Under the operating mode of city) safe and reliable use, cannot function as the security baseline requirement of automated driving system " upper road access ".
Domestic and international automobile industry also recognizes this problem, although to how to provide automated driving system " road in allowance " Security baseline requires still during research and discussion, but also reaches a basic Consensus to the test and appraisal of automated driving system, I.e. automated driving system test and appraisal need the support in compbined test verification place.For example, the U.S., Europe, Japan and China etc. all exist Establish the verification experimental verification demonstration area of automated driving system or intelligent network connection automobile.This respect U.S. is on the forefront, many state difference Laws and regulations are established to the popularization of automatic Pilot technology, relevant access system is established to the automobile of the test that needs to reach the standard grade, operation, And unified examination point is set.Examination can be generally divided into three phases：(1) data verification；(2) road examination；(3) synthesis is commented Estimate, issue licence plate, allows specified section test.
But at present how system evaluation automatic driving vehicle behavior safety, check and approve autonomous driving vehicle it is final In terms of upper road traveling, countries in the world are still at an early stage, and there is no the assessment methods for forming system.Therefore, test and appraisal side is carried out The research of method is conducive to the development of China's automatic driving vehicle industry, promotes associated test standards perfect.
Invention content
The technical problem to be solved by the present invention is to：At present how the safety of system evaluation automatic driving vehicle behavior Property, check and approve autonomous driving vehicle finally in terms of upper road traveling, countries in the world are still at an early stage, and there is no the surveys for forming system Method is commented, causes associated test standards and not perfect, is unfavorable for the development of China's automatic driving vehicle industry.
In order to solve the above technical problem, the present invention provides a kind of assessment methods of automatic driving vehicle behavior safety.
According to an embodiment of the invention, a kind of assessment method of automatic driving vehicle behavior safety is provided, including：
It obtains the environment sensing parameter of automatic driving vehicle, network link parameter, judge that decision parameters, control execute parameter Join degree coefficient with net；
Ginseng is executed according to the environment sensing parameter, the network link parameter, the judgement decision parameters, the control Number and the net join degree coefficient, obtain the test and appraisal score of automatic driving vehicle behavior safety；
According to the test and appraisal score of the automatic driving vehicle behavior safety, the automatic driving vehicle behavior safety is obtained Evaluating result.
Preferably, the test and appraisal score is obtained according to following formula：
Wherein, Z is the test and appraisal score, and H is the environment sensing parameter, and L is the network link parameter, and P is described Judge that decision parameters, K are that the control executes parameter, a, b are that the net joins degree coefficient, and a+b=1.
Preferably, according to the magnitude relationship of the test and appraisal score and preset score threshold, the automatic Pilot vehicle is obtained The evaluating result of behavior safety.
Preferably, the environment sensing parameter H of the automatic driving vehicle is obtained according to following formula：
H=HN × 0.4+HWW×0.4+HWA×0.2
Wherein, HN is environment sensing accuracy test and appraisal parameter, HWWFor environment sensing stability test and appraisal parameter, HWAFor environment Perceive promptness test and appraisal parameter.
Preferably, environment sensing accuracy test and appraisal parameter, environment sensing stability test and appraisal parameter and environment sensing and When property test and appraisal parameter perceives test and appraisal parameter by corresponding static environment respectively and dynamic environment perception test and appraisal parameter determines, Described in dynamic environment perception test and appraisal parameter include：There is operating mode test and appraisal parameter and danger in normal driving operating mode test and appraisal parameter, danger Danger release operating mode test and appraisal parameter.
Preferably, the network link parameter L of the automatic driving vehicle is obtained according to following formula：
L=LN × 0.4+LWW×0.4+LWA×0.2
Wherein, LN is network link accuracy test and appraisal parameter, LWWFor network link stability test and appraisal parameter, LWAFor network Unicom promptness test and appraisal parameter.
Preferably, network link accuracy test and appraisal parameter, network link stability test and appraisal parameter and network link and When property test and appraisal parameter respectively by corresponding static network unicom test and assess parameter and dynamic network unicom test and appraisal parameter determine, Described in dynamic network unicom test and appraisal parameter include：There is operating mode test and appraisal parameter and danger in normal driving operating mode test and appraisal parameter, danger Danger release operating mode test and appraisal parameter.
Preferably, the judgement decision parameters P of the automatic driving vehicle is obtained according to following formula：
P=PN × 0.2+PWW×0.6+PWA×0.2
Wherein, PN is to judge decision accuracy test and appraisal parameter, PWWTo judge decision stability test and appraisal parameter, PWATo judge Decision promptness test and appraisal parameter.
Preferably, judgement decision accuracy test and appraisal parameter, judge decision stability test and assess parameter and judge decision and When property test and appraisal parameter judges that decision test and appraisal parameter and dynamic judge that decision test and appraisal parameter determines by corresponding static state respectively, Described in dynamically judge decision test and appraisal parameter include：There is operating mode test and appraisal parameter and danger in normal driving operating mode test and appraisal parameter, danger Danger release operating mode test and appraisal parameter.
Preferably, the control for the automatic driving vehicle being obtained according to following formula executes parameter K：
K=KN × 0.4+KWW×0.4+KWA×0.2
Wherein, KN executes accuracy test and appraisal parameter, KW in order to controlWStability test and appraisal parameter, KW are executed in order to controlAIn order to control Execute promptness test and appraisal parameter.
Preferably, the control executes accuracy test and appraisal parameter, control executes stability test and appraisal parameter and control executes and When property test and appraisal parameter executes test and appraisal parameter by corresponding static cost control respectively and dynamic control executes test and appraisal parameter and determines, Described in dynamic control execute test and appraisal parameter include：There is operating mode test and appraisal parameter and danger in normal driving operating mode test and appraisal parameter, danger Danger release operating mode test and appraisal parameter.
Compared with prior art, one or more of said program embodiment can have the following advantages that or beneficial to effect Fruit：
Using the assessment method of automatic driving vehicle behavior safety provided in an embodiment of the present invention, by the influence of acquisition from The dynamic environment sensing parameter for driving vehicle behavior safety, network link parameter judge decision parameters, control execution parameter and automatically The net for driving vehicle joins degree coefficient, obtains the test and appraisal score of automatic driving vehicle behavior safety, is obtained according to the test and appraisal score The evaluating result of automatic driving vehicle behavior safety.Therefore, the assessment method energy of automatic driving vehicle behavior safety of the invention Enough behavior safeties that automatic driving vehicle is considered from multi-angle comprehensive keep evaluating result accurate, reliable, to be automatic Pilot Vehicle behavior safety test and evaluation provide a kind of possible ways, keep associated test standards perfect, and China is promoted to drive automatically Sail the development of Vehicle Industry.
Also, in the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention, the environment sensing of acquisition Parameter, network link parameter judge that decision parameters and control execution parameter are respectively by combining corresponding static test and appraisal The characteristic of parameter and dynamic measuring and evaluating parameter and obtain.Therefore, the present invention can consider automatic driving vehicle from multi-angle comprehensive Behavior safety is effectively improved the accuracy and reliability of evaluating result.
Other features and advantages of the present invention will be illustrated in the following description, and partly becomes from specification It is clear that understand through the implementation of the invention.The purpose of the present invention and other advantages can be by wanting in specification, right Specifically noted structure is sought in book and attached drawing to realize and obtain.
Description of the drawings
Attached drawing is used to provide further understanding of the present invention, and a part for constitution instruction, the reality with the present invention It applies example and is used together to explain the present invention, be not construed as limiting the invention.In the accompanying drawings：
Fig. 1 shows the flow diagram of the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention.
Specific implementation mode
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and examples, how to be applied to the present invention whereby Technological means solves technical problem, and the realization process for reaching technique effect can fully understand and implement.It needs to illustrate As long as not constituting conflict, each embodiment in the present invention and each feature in each embodiment can be combined with each other, It is formed by technical solution within protection scope of the present invention.
For solve in the prior art how system evaluation automatic driving vehicle behavior safety, check and approve automatic Pilot vapour For vehicle finally in terms of upper road traveling, countries in the world are still at an early stage, and there is no the assessment methods for forming system, cause correlative measurement Test-object is accurate simultaneously not perfect, the technical issues of being unfavorable for the development of China's automatic driving vehicle industry, an embodiment of the present invention provides A kind of assessment method of automatic driving vehicle behavior safety.
The assessment method of the automatic driving vehicle behavior safety of the present invention is its safety of being tested and assessed by the way of score of testing and assessing Property.
The test and appraisal score of automatic driving vehicle behavior safety depend primarily on automatic driving vehicle environment sensing parameter H, Network link parameter L, judge that decision parameters P, control execute parameter K and net connection degree coefficient a, b.
Fig. 1 shows the flow diagram of the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention.
As shown in Figure 1, the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention, includes mainly following step Rapid S101 to step S103.
Step S101, obtain the environment sensing parameter H of automatic driving vehicle, network link parameter L, judge decision parameters P, Control executes parameter K and net connection degree coefficient a, b.
Step S102 according to environment sensing parameter H, network link parameter L, judges that decision parameters P, control execute parameter K Join degree coefficient a, b with net, obtains the test and appraisal score of automatic driving vehicle behavior safety.
Step S103 obtains automatic driving vehicle behavior safety according to the test and appraisal score of automatic driving vehicle behavior safety Evaluating result.
Specifically, step S101 is first carried out, obtains the environment sensing parameter H of automatic driving vehicle.
Here, the environment sensing parameter of automatic driving vehicle is for evaluating automatic driving vehicle in automatic Pilot action process In include to the perception degree of related physical environment, the physical environment：Pavement conditions, traffic information, road markings etc..
Before determining the environment sensing parameter H of automatic driving vehicle, it is thus necessary to determine that following parameter：Environment sensing accuracy Parameter of testing and assessing HN, environment sensing stability test and appraisal parameter HWWWith environment sensing promptness test and appraisal parameter HWA.Wherein, environment sensing Accuracy test and appraisal parameter HN, environment sensing stability test and appraisal parameter HWWWith environment sensing promptness test and appraisal parameter HWARespectively by each Self-corresponding static environment perception test and appraisal parameter and dynamic environment perception test and appraisal parameter determine.Here, wherein dynamic environment perceives Test and appraisal parameter include：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger. Specifically, static environment perception test and appraisal parameter includes：Static environment perceives accuracy HJC, static environment perception stability HJRWith Static environment perceives promptness HJT.Dynamic environment perceives test and appraisal parameter：Normal driving operating mode dynamic environment perceives accuracy HDCN, normal driving operating mode dynamic environment perception stability HDRN, normal driving operating mode dynamic environment perception promptness HDTN；It is dangerous There is operating mode dynamic environment perception accuracy HDCWW, danger there is operating mode dynamic environment perception stability HDRWW, danger there is work Condition dynamic environment perceives promptness HDTWW；Danger release operating mode dynamic environment perception accuracy HDCWA, dangerous release operating mode dynamic Environment sensing stability HDRWA, dangerous release operating mode dynamic environment perception promptness HDTWA。
As it can be seen that the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention is by combining static environment to perceive Test and appraisal parameter and dynamic environment perception test and appraisal parameter calculate the environment sensing parameter H of automatic driving vehicle, can be from multi-angle The behavior safety for considering automatic driving vehicle is effectively improved the accuracy and reliability of evaluating result.
After obtaining above-mentioned parameter, the environment sensing parameter H of automatic driving vehicle is obtained according to following formula：
H=HN × 0.4+HWW×0.4+HWA×0.2 (1)
Utilize each parameter in following various acquisition formula (1)：
HN=HCN×0.5+HRN×0.3+HTN×0.2
HWW=HCWW×0.5+HRWW×0.3+HTWW×0.2
HWA=HCWA×0.5+HRWA×0.3+HTWA×0.2
Wherein, HCNAccuracy test and appraisal parameter, HR are perceived for normal driving work condition environmentNThere is work condition environment perception for danger Accuracy test and appraisal parameter, HTNAccuracy test and appraisal parameter, HC are perceived for danger release work condition environmentWWFor normal driving work condition environment Perceive stability test and appraisal parameter, HRWWThere is work condition environment for danger and perceives stability test and appraisal parameter, HTWWOperating mode is discharged for danger Environment sensing stability test and appraisal parameter, HCWAPromptness test and appraisal parameter, HR are perceived for normal driving work condition environmentWAOccur for danger Work condition environment perceives promptness test and appraisal parameter, HTWAPromptness test and appraisal parameter is perceived for danger release work condition environment.
Utilize the following various each parameter obtained in above-mentioned expression formula：
The specific acquisition process of above-mentioned parameter is as follows：
By static environment perception accuracy HJC, static environment perception stability HJR, static environment perception promptness HJT's Initial value is respectively set to 100 points.
Static environment perceives accuracy：The order of accuarcy of static automatic driving vehicle identification object.Static environment Perceive accuracy HJCScoring rule：Static evaluation and test vehicle front certain distance (20 meters) occurs respectively：3 kinds of type signal lamps (conventional natural light irradiation), 100 kinds of traffic marks (positive, grey), 20 kinds of automobiles (forward direction) and 21 pedestrians's (forward direction) etc. 150 A perceptive object, it is desirable that vehicle is identified before the deadline, each Object identifying mistake or unidentified out 1 point of button, Minimum 0 point, output character result after the completion of identification.
Static environment perceives stability：When identification image parameter is (for example, signal lamp light intensity, traffic mark angle Degree, color etc.) when changing, static automatic driving vehicle accurately identifies the degree of stability of object.Static environment perception is steady Surely HJ is spentRScoring rule：Static evaluation and test vehicle front certain distance (20 meters) occurs respectively：3 kinds of type signal lamp (intense light irradiations Penetrate), 100 kinds of traffic marks (15 degree of angles, white or blue), 20 kinds of automobiles (lateral and backwards to) and 21 pedestrians it is (lateral and carry on the back To) etc. 291 perceptive objects, it is desirable that vehicle is identified before the deadline, each Object identifying mistake or it is unidentified go out Detain 0.5 point, minimum 0 point, output character result after the completion of identification.
It should be noted that in the present invention, automobile or pedestrian " lateral " refer to the vehicle with side static (or dynamic) Or meeting occurs by pedestrian or behavior of meeting, angle are unlimited.
Static environment perceives promptness：Static automatic driving vehicle accurately identifies the timely degree of object.It is static Environment sensing promptness HJTScoring rule：Static evaluation and test vehicle front certain distance (20 meters) occurs respectively：3 kinds of typical letters Signal lamp (conventional natural light irradiation), 100 kinds of traffic marks (positive, grey), 20 kinds of automobiles (forward direction) and 21 pedestrians's (forward direction) Deng 150 perceptive objects, it is desirable that vehicle was identified within the defined time (100ms), each Object identifying mistake or did not knew It Chu Lai not detain 1 point, minimum 0 point, output character result after the completion of identification.
By normal driving operating mode dynamic environment perception accuracy HDCN, normal driving operating mode dynamic environment perceive stability HDRN, normal driving operating mode dynamic environment perception promptness HDTNInitial value be respectively set to 100 points.
Normal driving operating mode dynamic environment perceives accuracy：Automatic driving vehicle identifies pair in normal driving process As the order of accuarcy of (for example, traffic lights, static traffic mark, other traffic participants).Normal driving operating mode dynamic environment sense Know accuracy HDCNScoring rule：The evaluation and test vehicle front certain distance (20 meters) of normally travel occurs respectively：9 major class are normally driven It sails operating mode and amounts to 45 perceptive objects, perceptive object is in normal condition, it is desirable that vehicle is identified before the deadline, often A Object identifying mistake or unidentified out 3 points of button, minimum 0 point, output character result after the completion of identification.
It should be noted that in the present invention, traffic mark can have that there are two types of versions：Angle (positive and 15 degree of angles) With the variations such as background (white, grey and blue)；Test automobile and pedestrian can have a kind of version：Direction variation (positive, Laterally and backwards)；Test signal lamp can have a kind of version：Intensity of illumination (conventional natural light irradiation, strong illumination). " normal condition " refers to 3 kinds of type signal lamps (conventional natural light irradiation), 100 kinds of traffic marks (positive, grey), 20 kinds of automobiles (forward direction) and 21 pedestrians's (forward direction) etc..Under normal condition, automatic driving vehicle does not face a danger scene in the process of moving. Normal condition is removed, other is abnormality.
Normal driving operating mode dynamic environment perceives stability：Automatic driving vehicle identifies pair in normal driving process As when identification image parameter (for example, signal lamp light intensity, traffic mark angle, color etc.) changes or occurs abnormal When, automatic driving vehicle can still accurately identify the degree of stability of object.Normal driving operating mode dynamic environment perceives stability HDRN Scoring rule：The evaluation and test vehicle front certain distance (20 meters) of normally travel occurs respectively：9 major class normal driving operating modes total 45 A perceptive object, perceptive object are in abnormality (illumination, direction, background etc. and normal condition are variant), it is desirable that vehicle exists It is identified in the defined time, each Object identifying mistake or unidentified out 3 points of button are minimum 0 point, defeated after the completion of identification Go out character result.
Normal driving operating mode dynamic environment perceives promptness：Automatic driving vehicle is accurately known in normal driving process The timely degree of other object.Normal driving operating mode dynamic environment perceives promptness HDTNScoring rule：The evaluation and test vehicle of normally travel Front certain distance (20 meters) occurs respectively：9 major class normal driving operating modes amount to 45 perceptive objects, it is desirable that vehicle is defined It is identified in time (0.5s), each Object identifying mistake or unidentified out 3 points of button are minimum 0 point, defeated after the completion of identification Go out character result.
There is into operating mode dynamic environment perception accuracy HD in dangerCWW, danger occur operating mode dynamic environment perception stability HDRWW, danger there is operating mode dynamic environment perception promptness HDTWWInitial value be respectively set to 100 points.
There is the perception accuracy of operating mode dynamic environment in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, the order of accuarcy of identification object (for example, traffic lights, static traffic mark, other traffic participants).It is dangerous There is operating mode dynamic environment perception accuracy HDCWWScoring rule：(20 meters) points of the evaluation and test vehicle front certain distance of normally travel Do not occur：There is operating mode and amount to 230 perceptive objects in 32 major class danger, and perceptive object is in normal condition, it is desirable that vehicle is being advised It is identified in the fixed time, each Object identifying mistake or unidentified out 0.5 point of button are minimum 0 point, defeated after the completion of identification Go out character result.
There is operating mode dynamic environment perception stability in danger：Automatic driving vehicle identifies pair in normal driving process As in the case where operating mode occurs in danger, when identification image parameter (for example, signal lamp light intensity, traffic mark angle, color etc.) is sent out When changing or abnormal appearance, automatic driving vehicle can still accurately identify the degree of stability of object.There is operating mode dynamic in danger Environment sensing stability HDRWWScoring rule：The evaluation and test vehicle front certain distance (20 meters) of normally travel occurs respectively：32 is big There is operating mode and amounts to 230 perceptive objects in class danger, and perceptive object is in abnormality (illumination, direction, background etc. and normal shape State is variant), it is desirable that vehicle is identified before the deadline, each Object identifying mistake or unidentified out 0.5 point of button, Minimum 0 point, output character result after the completion of identification.
In order to make it easy to understand, being illustrated so that non-motor vehicle crosses as an example here.
There is operating mode in danger：Non-motor vehicle crosses.
Environment：It is daytime, fine, in addition to Primary Actor, appropriate mark etc. is supplemented, in total 5 perception points.
Road scene：There are large amount of building in urban district T-shaped road junction or crossroad at crossing.
Vehicle-state：Main vehicle is turned right in urban district T-shaped road junction or crossroad, and right front occurs crossing highway suddenly Cyclist, main vehicle camera and pilot's line of vision are blocked by building, if two vehicles keep motion state constant at crossing It can collide.
Assessment method：Scene section grows 10 meters, has curtain to block other than 50 meters of scene, and vehicle is with the speed of 40km/h to field Scape direction running starts timing, the number of the correct marker lamp of record vehicle, mark, other traffic participants etc. by curtain Amount.
There is operating mode dynamic environment perception promptness in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, accurately identifies the timely degree of object (for example, traffic lights, static traffic mark, other traffic participants). There is operating mode dynamic environment perception promptness HD in dangerTWWScoring rule：The evaluation and test vehicle front certain distance (20 of normally travel Rice) occur respectively：There is operating mode and amount to 230 perceptive objects in 32 major class danger, it is desirable that vehicle is within the defined time (0.5s) It is identified, each Object identifying mistake or unidentified out 0.5 point of button, minimum 0 point, output character knot after the completion of identification Fruit.
By danger release operating mode dynamic environment perception accuracy HDCWA, dangerous release operating mode dynamic environment perceive stability HDRWA, dangerous release operating mode dynamic environment perception promptness HDTWAInitial value be respectively set to 100 points.
Danger release operating mode dynamic environment perceives accuracy and refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, the order of accuarcy of identification object (for example, traffic lights, static traffic mark, other traffic participants).It is dangerous Discharge operating mode dynamic environment perception accuracy HDCWAScoring rule：(20 meters) points of the evaluation and test vehicle front certain distance of normally travel Do not occur：5 major class danger discharge operating mode and amount to 25 perceptive objects, and perceptive object is in normal condition, it is desirable that vehicle is providing Time in be identified, each Object identifying mistake or it is unidentified out button 4 points, minimum 0 point, output word after the completion of identification Accord with result.
It should be noted that in the present invention, " danger release operating mode " refers to the dangerous situation of release, totally 5 major class, respectively For：Collide with pedestrian collision, with animal, collide with straight traffic, colliding at crossing and vehicle, with cycle People collides.
In order to make it easy to understand, here by with illustrated for pedestrian collision.
Danger release operating mode：With pedestrian collision.
Environment：It is daytime, fine, in addition to Primary Actor, appropriate mark etc. is supplemented, in total 5 perception points.
Road scene：All kinds of roads.
Vehicle-state：Main vehicle is kept straight on certain speed (V) on road, and there is stationary vehicle in front crossroad roadside, drives The person's of sailing sight is blocked by roadside static background vehicle, and main vehicle continues to keep straight on.
Pedestrian's state：By road, main vehicle is collided with pedestrian at crossing pedestrian when main vehicle reaches crossroad.
Assessment method：Scene section grows 10 meters, has curtain to block other than 50 meters of scene, and vehicle is with the speed of 40km/h to field Scape direction running starts timing, the quantity of the correct marker lamp of record vehicle, mark, other traffic participants etc. after collision.
Danger release operating mode dynamic environment perceives stability and refers to：Automatic driving vehicle identifies pair in normal driving process As in the case where discharging dangerous situation, when identification image parameter (for example, signal lamp light intensity, traffic mark angle, color etc.) is sent out When changing or abnormal appearance, automatic driving vehicle can still accurately identify the degree of stability of object.Danger release operating mode dynamic Environment sensing stability HDRWAScoring rule：The evaluation and test vehicle front certain distance (20 meters) of normally travel occurs respectively：5 major class Danger release operating mode amounts to 25 perceptive objects, and perceptive object is in abnormality (illumination, direction, background etc. and normal condition It is variant), it is desirable that vehicle is identified before the deadline, each Object identifying mistake or unidentified out 4 points of button, minimum It is 0 point, output character result after the completion of identification.
Danger release operating mode dynamic environment perceives promptness and refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, accurately identifies the timely degree of object (for example, traffic lights, static traffic mark, other traffic participants). Danger release operating mode dynamic environment perception promptness HDTWAScoring rule：The evaluation and test vehicle front certain distance (20 of normally travel Rice) occur respectively：5 major class danger discharge operating mode and amount to 25 perceptive objects, it is desirable that vehicle within the defined time (0.5s) into Row identification, each Object identifying mistake or unidentified out 4 points of button, minimum 0 point, output character result after the completion of identification.
Obtain the network link parameter L of automatic driving vehicle.
In order to make automatic driving vehicle preferably be communicated with other vehicles and roadside device, control centre can be more preferable Ground provides service for vehicle, and intelligent network connection automobile should have good network link ability.The network link of automatic driving vehicle Parameter is for evaluating the ability that automatic driving vehicle is communicated with other vehicles and roadside device.
The network link parameter of automatic driving vehicle includes static network unicom test and appraisal parameter and the test and appraisal of dynamic network unicom Parameter.
Static network unicom test and assess parameter mainly for the ad hoc capability of automatic driving vehicle, communication range, bandwidth and The adaptability of signal strength tests and assesses to network link.When test and appraisal, the row model for the national games for this four direction system is needed It encloses and respectively generates 1000 test response inputs and ensure without other interference signals, in addition generate 1000 test response inputs simultaneously It is combined with interference signal, is evaluated and tested according to the communications reception situation of system.
The dynamic network unicom ability of dynamic network unicom test and appraisal parameter emphasis evaluation and test intelligent network connection automobile.When test and appraisal, lead to Vehicle is crossed under motion state, to different operating modes (for example, the communications status under continually changing weather, road is (for example, signal Intensity, distance etc.)) network link ability evaluated and tested.
Before determining the network link parameter L of automatic driving vehicle, it is thus necessary to determine that following parameter：Network link accuracy Parameter of testing and assessing LN, network link stability test and appraisal parameter LWWWith network link promptness test and appraisal parameter LWA.Wherein, network link Accuracy test and appraisal parameter LN, network link stability test and appraisal parameter LWWWith network link promptness test and appraisal parameter LWARespectively by each Self-corresponding static network unicom test and appraisal parameter and dynamic network unicom test and appraisal parameter determine.Here, wherein dynamic network unicom Test and appraisal parameter include：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger. Specifically, static network unicom test and appraisal parameter includes：Static network unicom accuracy LJC, static network unicom stability LJRWith Static network unicom promptness LJT.Dynamic network unicom test and appraisal parameter include：Normal driving operating mode dynamic network unicom accuracy LDCN, normal driving operating mode dynamic network unicom stability LDRN, normal driving operating mode dynamic network unicom promptness LDTN；It is dangerous There is operating mode dynamic network unicom accuracy LDCWW, danger there is operating mode dynamic network unicom stability LDRWW, danger there is work Condition dynamic network unicom promptness LDTWW；Danger release operating mode dynamic network unicom accuracy LDCWA, dangerous release operating mode dynamic Network link stability LDRWA, dangerous release operating mode dynamic network unicom promptness LDTWA。
As it can be seen that the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention is by combining static network unicom Test and appraisal parameter and dynamic network unicom test and appraisal parameter calculate the network link parameter L of automatic driving vehicle, can be from multi-angle The behavior safety for considering automatic driving vehicle is effectively improved the accuracy and reliability of evaluating result.
After obtaining above-mentioned parameter, the network link parameter L of automatic driving vehicle is obtained according to following formula：
L=LN × 0.4+LWW×0.4+LWA×0.2 (2)
Utilize each parameter in following various acquisition formula (2)：
LN=LCN×0.4+LRN×0.3+LTN×0.3
LWW=LCWW×0.4+LRWW×0.3+LTWW×0.3
LWA=LCWA×0.4+LRWA×0.3+LTWA×0.3
Wherein, LCNFor normal driving operating mode network link accuracy test and appraisal parameter, LRNThere is operating mode network link for danger Accuracy test and appraisal parameter, LTNFor danger release operating mode network link accuracy test and appraisal parameter, LCWWFor normal driving operating mode network Unicom stability test and appraisal parameter, LRWWThere is operating mode network link stability test and appraisal parameter, LT for dangerWWOperating mode is discharged for danger Network link stability test and appraisal parameter, LCWAFor normal driving operating mode network link promptness test and appraisal parameter, LRWAOccur for danger Operating mode network link promptness test and appraisal parameter, LTWAFor danger release operating mode network link promptness test and appraisal parameter.
Utilize the following various each parameter obtained in above-mentioned expression formula：
The specific acquisition process of above-mentioned parameter is as follows：
By static network unicom accuracy LJC, static network unicom stability LJR, static network unicom promptness LJT's Initial value is respectively set to 100 points.
Static network unicom accuracy refers to：Static automatic driving vehicle receives the order of accuarcy of communication.Static network Unicom accuracy LJCScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance：1000 test response inputs And without the scene of other interference signals, it is desirable that vehicle receives before the deadline, and each scene receives information errors or do not receive Divide to information button 0.1, minimum 0 point, finishes receiving rear output character result.
Static network unicom stability refers to：Static automatic driving vehicle accurately receives the degree of stability of communication.It is static Network link stability LJRScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance：1000 test responses Input and have the scene of interference signal, it is desirable that vehicle receives before the deadline, and each scene receives information errors or do not receive Divide to information button 0.1, minimum 0 point, finishes receiving rear output character result.
Static network unicom promptness refers to：Static automatic driving vehicle accurately receives the timely degree of communication.It is static Network link promptness LJTScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance：1000 test responses It inputs and without the scene of other interference signals, it is desirable that vehicle receives within the defined time (100ms), and each scene receives information Mistake does not receive information button 0.1 and divides, and minimum 0 point, finishes receiving rear output character result.
By normal driving operating mode dynamic network unicom accuracy LDCN, normal driving operating mode dynamic network unicom stability LDRN, normal driving operating mode dynamic network unicom promptness LDTNInitial value be respectively set to 100 points.
Normal driving operating mode dynamic network unicom accuracy refers to：Automatic driving vehicle is in normal driving process, not With (for example, communications status (for example, signal strength, distance etc.) under continually changing weather, road) under operating mode, vehicle receives The order of accuarcy of communication.Normal driving operating mode dynamic network unicom accuracy LDCNScoring rule：Before the evaluation and test vehicle of normally travel (20 meters) appearance of square certain distance：9 major class normal driving operating modes amount to 9 network link scenes, no signal interference, it is desirable that vehicle It receives before the deadline, each scene, which receives information errors or do not receive information button 15, divides, and minimum 0 point, finishes receiving Output character result afterwards.
In order to make it easy to understand, being illustrated by taking section of keeping straight on as an example here.
Normal driving operating mode：Straight trip section.
Environment explanation：The linear section in two-way two track, roadside have traffic mark and other traffic participants (to have communication energy Power), there is communications facility in roadside, can configure jamming equipment.
Assessment method：Scene section grows 10 meters, has curtain to block other than 50 meters of scene, and vehicle is with the speed of 40km/h to field The case where scape direction running starts timing by curtain, and record vehicle receives information.
Normal driving operating mode dynamic network unicom stability refers to：Automatic driving vehicle is in normal driving process, not With (for example, communications status (for example, signal strength, distance etc.) under continually changing weather, road) under operating mode, vehicle is accurate Receive the degree of stability of communication.Normal driving operating mode dynamic network unicom stability LDRNScoring rule：The evaluation and test vehicle of normally travel (20 meters) appearance of certain distance in front of：9 major class normal driving operating modes amount to 9 network link scenes, there is signal interference, it is desirable that Vehicle receives before the deadline, and each scene, which receives information errors or do not receive information button 15, divides, and minimum 0 point, receives Output character result after the completion.
Normal driving operating mode dynamic network unicom promptness refers to：Automatic driving vehicle is in normal driving process, not With (for example, communications status (for example, signal strength, distance etc.) under continually changing weather, road) under operating mode, vehicle is accurate Receive the timely degree of communication.Normal driving operating mode dynamic network unicom promptness LDTNScoring rule：The evaluation and test vehicle of normally travel (20 meters) appearance of certain distance in front of：9 major class normal driving operating modes amount to 9 network link scenes, no signal interference, it is desirable that Vehicle receives within the defined time (0.5s), and each scene, which receives information errors or do not receive information button 15, divides, and minimum 0 Point, finish receiving rear output character result.
There is into operating mode dynamic network unicom accuracy LD in dangerCWW, danger there is operating mode dynamic network unicom stability LDRWW, danger there is operating mode dynamic network unicom promptness LDTWWInitial value be respectively set to 100 points.
There is operating mode dynamic network unicom accuracy in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, and has signal interference, and vehicle receives the order of accuarcy of communication.It is accurate that operating mode dynamic network unicom occurs in danger Spend LDCWWScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：It is total that operating mode occur in 32 major class danger Count 46 network link scenes, no signal interference, it is desirable that vehicle receives before the deadline, and each scene receives information errors Or do not receive information button 3 and divide, minimum 0 point, finish receiving rear output character result.
There is operating mode dynamic network unicom stability in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, and has signal interference, and vehicle accurately receives the degree of stability of communication.There is operating mode dynamic network unicom in danger Stability LDRWWScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There is work in 32 major class danger Condition amounts to 46 network link scenes, there is signal interference, it is desirable that vehicle receives before the deadline, and each scene receives information Mistake does not receive information button 3 and divides, and minimum 0 point, finishes receiving rear output character result.
There is operating mode dynamic network unicom promptness in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, and has signal interference, and vehicle accurately receives the timely degree of communication.There is operating mode dynamic network unicom in danger Promptness LDTWWScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There is work in 32 major class danger Condition amounts to 46 network link scenes, no signal interference, it is desirable that vehicle receives within the defined time (0.5s), and each scene connects It receives information errors or does not receive information button 3 and divide, minimum 0 point, finish receiving rear output character result.
By danger release operating mode dynamic network unicom accuracy LDCWA, dangerous release operating mode dynamic network unicom stability LDRWA, dangerous release operating mode dynamic network unicom promptness LDTWAInitial value be respectively set to 100 points.
Danger discharges operating mode dynamic network unicom accuracy：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, and has signal interference, vehicle receives the order of accuarcy of communication.Danger release operating mode dynamic network unicom is accurate Spend LDCWAScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：It is total that 5 major class danger discharge operating mode 5 network link scenes, no signal interference, it is desirable that vehicle receives before the deadline, each scene receive information errors or Information button 20 is not received to divide, minimum 0 point, finish receiving rear output character result.
Danger discharges operating mode dynamic network unicom stability：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, and has signal interference, vehicle accurately receives the degree of stability of communication.Danger release operating mode dynamic network unicom Stability LDRWAScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：5 major class danger discharge operating mode Amount to 25 network link scenes, there is signal interference, it is desirable that vehicle receives before the deadline, and it is wrong that each scene receives information It misses or does not receive information button 20 and divide, minimum 0 point, finish receiving rear output character result.
Danger discharges operating mode dynamic network unicom promptness：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, and has signal interference, vehicle accurately receives the timely degree of communication.Danger release operating mode dynamic network unicom Promptness LDTWAScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：5 major class danger discharge operating mode Amount to 5 network link scenes, no signal interference, it is desirable that vehicle receives within the defined time (0.5s), and each scene receives Information errors do not receive information button 20 and divide, and minimum 0 point, finish receiving rear output character result.
Obtain the judgement decision parameters P of automatic driving vehicle.
Automatic driving vehicle is in the case where having environment sensing ability and network link ability, it is necessary to which it is accurate to have The ability for judging and making decisions.The judgement decision parameters of automatic driving vehicle are for evaluating automatic driving vehicle in automatic Pilot In action process, different operating modes are judged with the correct degree of decision.
The judgement decision parameters of automatic driving vehicle include that static state judges that decision test and appraisal parameter and dynamic judge that decision is tested and assessed Parameter.
Static state judges decision test and appraisal parameter emphasis evaluation and test traveling state of vehicle judgement, condition of road surface judgement and determines Plan ability.By evaluate and test vehicle judge should take control ability (accelerate, slow down, at the uniform velocity, turn to etc.), judge vehicle stop position The ability set, judges the acquisitions such as the ability what situation road is at the ability for judging vehicle movement track.When test and appraisal, in reality It tests room environmental and generates scene and 1000 kinds of other traffic participants that 1000 kinds of other traffic participants keep original driving intention Whether the scene that can change original driving intention carries out judgement and decision for automobile, safe or correct according to Vehicle Decision Method result It is evaluated and tested.
Dynamic judges that the dynamic of decision test and appraisal parameter emphasis evaluation and test intelligent network connection automobile judges decision-making capability, is existed by vehicle Under motion state, the judgement decision-making capability of different operating modes is evaluated and tested.
Before determining the judgement decision parameters P of automatic driving vehicle, it is thus necessary to determine that following parameter：Judge decision accuracy Test and appraisal parameter PN, judge decision stability test and appraisal parameter PWWWith judge decision promptness test and appraisal parameter PWA.Judge decision accuracy Test and appraisal parameter PN, judge decision stability test and appraisal parameter PWWWith judge decision promptness test and appraisal parameter PWARespectively by respectively corresponding to Static state judge decision test and appraisal parameter and dynamic judge decision test and appraisal parameter determine, wherein dynamically judge decision test and appraisal parameter packet It includes：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger.Specifically, quiet State judges that decision test and appraisal parameter includes：Static state judges decision accuracy PJC, static judge decision stability PJRIt determines with static state judgement Plan promptness PJT.Dynamic judges that decision test and appraisal parameter includes：Normal driving operating mode dynamic judges decision accuracy PDCN, normally drive It sails operating mode dynamic and judges decision stability PDRN, normal driving operating mode dynamic judge decision promptness PDTN；It is dynamic that operating mode occurs in danger State judges decision accuracy PDCWW, danger occur operating mode dynamic judge decision stability PDRWW, danger occur operating mode dynamic judge Decision promptness PDTWW；Danger release operating mode dynamic judges decision accuracy PDCWA, dangerous release operating mode dynamic judge that decision is steady Surely PD is spentRWA, dangerous release operating mode dynamic judge decision promptness PDTWA。
As it can be seen that the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention in conjunction with static state by judging decision Test and appraisal parameter calculates the judgement decision parameters P of automatic driving vehicle with decision test and appraisal parameter is dynamically judged, can be from multi-angle The behavior safety for considering automatic driving vehicle is effectively improved the accuracy and reliability of evaluating result.
After obtaining above-mentioned parameter, the judgement decision parameters P of automatic driving vehicle is obtained according to following formula：
P=PN × 0.2+PWW×0.6+PWA×0.2 (3)
Utilize each parameter in following various acquisition formula (3)：
PN=PCN×0.5+PRN×0.3+PTN×0.2
PWW=PCWW×0.5+PRWW×0.3+PTWW×0.2
PWA=PCWA×0.5+PRWA×0.3+PTWA×0.2
Wherein, PCNJudge decision accuracy test and appraisal parameter, PR for normal driving operating modeNThere is operating mode for danger and judges decision Accuracy test and appraisal parameter, PTNJudge decision accuracy test and appraisal parameter, PC for danger release operating modeWWJudge for normal driving operating mode Decision stability test and appraisal parameter, PRWWThere is operating mode for danger and judges decision stability test and appraisal parameter, PTWWOperating mode is discharged for danger Judge decision stability test and appraisal parameter, PCWAJudge decision promptness test and appraisal parameter, PR for normal driving operating modeWAOccur for danger Operating mode judges decision promptness test and appraisal parameter, PTWAJudge decision promptness test and appraisal parameter for danger release operating mode.
Utilize the following various each parameter obtained in above-mentioned expression formula：
The specific acquisition process of above-mentioned parameter is as follows：
Static state is judged into decision accuracy PJC, static judge decision stability PJR, static judge decision promptness PJT's Initial value is respectively set to 100 points.
Static state judges that decision accuracy refers to：Static automatic driving vehicle judges the accurate of decision to different operating modes Degree.Static state judges decision accuracy PJCScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance：1000 The other traffic participants of kind keep the scene of original driving intention, it is desirable that vehicle carries out judgement decision before the deadline, often A scene judges that decision error or non-decision out detain 0.1 point, minimum 0 point, judges output character result after decision.
It should be noted that setting other traffic participants here keeps original driving intention, it is in order to avoid traffic is joined The influence of decision is judged with person's oneself factor automatic driving vehicle.
Static state judges that decision stability refers to：When driving intention scene changes, static automatic driving vehicle is done Go out the degree of stability of accurate judgement decision.Static state judges decision stability PJRScoring rule：Static evaluation and test vehicle front is certain (20 meters) appearance of distance：1000 kinds of other traffic participants change the scene of original driving intention, it is desirable that vehicle is when defined Interior to carry out judgement decision, each scene judges that decision error or non-decision out detain 0.1 point, minimum 0 point, judges certainly Output character result after plan.
It should be noted that setting other traffic participants here changes original driving intention, it is in order to avoid traffic is joined The influence of decision is judged with person's oneself factor automatic driving vehicle.
Static state judges that decision promptness refers to：Static automatic driving vehicle makes accurate judgement decision to different operating modes Timely degree.Static state judges decision promptness PJTScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance： 1000 kinds of other traffic participants keep the scene of original driving intention, it is desirable that vehicle carries out within the defined time (100ms) Judge that decision, each scene judge that decision error or non-decision out detain 0.1 point, minimum 0 point, exported after judging decision Character result.
It should be noted that setting other traffic participants here keeps original driving intention, it is in order to avoid traffic is joined The influence of decision is judged with person's oneself factor automatic driving vehicle.
Normal driving operating mode dynamic is judged into decision accuracy PDCN, normal driving operating mode dynamic judge decision stability PDRN, normal driving operating mode dynamic judge decision promptness PDTNInitial value be respectively set to 100 points.
Normal driving operating mode dynamic judges that decision accuracy refers to：Automatic driving vehicle is in normal driving process, to not The order of accuarcy of decision is judged with scene.Normal driving operating mode dynamic judges decision accuracy PDCNScoring rule：Normal row (20 meters) appearance of the evaluation and test vehicle front certain distance sailed：Total 9 of 9 major class normal driving operating modes judge decision scene, other Traffic participant keeps original driving intention, it is desirable that vehicle carries out judgement decision before the deadline, and each scene judgement is determined Plan mistake or non-decision out detain 15 points, minimum 0 point, judge output character result after decision.
Normal driving operating mode dynamic judges that decision stability refers to：Automatic driving vehicle is in normal driving process, on the spot When scape changes, automatic driving vehicle makes the degree of stability of accurate judgement decision.Normal driving operating mode dynamic judges decision Stability PDRNScoring rule：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：9 major class normal driving operating modes Total 9 judge that decision scene, other traffic participants change original driving intention, it is desirable that vehicle carries out before the deadline Judge that decision, each scene judge that decision error or non-decision out detain 15 points, minimum 0 point, exported after judging decision Character result.
Normal driving operating mode dynamic judges that decision promptness refers to：Automatic driving vehicle is in normal driving process, to not The timely degree of accurate judgement decision is made with scene.Normal driving operating mode dynamic judges decision promptness PDTNScoring rule：Just (20 meters) appearance of the evaluation and test vehicle front certain distance often travelled：Total 9 of 9 major class normal driving operating modes judge decision scene, Other traffic participants keep original driving intention, it is desirable that vehicle carries out judgement decision within the defined time (0.5s), each Scene judges that decision error or non-decision out detain 15 points, minimum 0 point, judges output character result after decision.
Is there is into operating mode dynamic in danger and judges decision accuracy PDCWW, danger occur operating mode dynamic judge decision stability PDRWW, danger occur operating mode dynamic judge decision promptness PDTWWInitial value be respectively set to 100 points.
There is operating mode dynamic and judges that decision accuracy refers in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, judges the order of accuarcy of decision.There is operating mode dynamic and judges decision accuracy PD in dangerCWWScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There are total 46 judgements of operating mode and determine in 32 major class danger Plan scene, other traffic participants keep original driving intention, it is desirable that vehicle carries out judgement decision before the deadline, each Scene judges that decision error or non-decision out detain 3 points, minimum 0 point, judges output character result after decision.
There is operating mode dynamic and judges that decision stability refers in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, makes the degree of stability of accurate judgement decision.There is operating mode dynamic and judges decision stability PD in dangerRWWMeter Divider is then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There is total 46 of operating mode and sentence in 32 major class danger Disconnected decision scene, other traffic participants change original driving intention, it is desirable that and vehicle carries out judgement decision before the deadline, Each scene judges that decision error or non-decision out detain 3 points, minimum 0 point, judges output character result after decision.
There is operating mode dynamic and judges that decision promptness refers in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, makes the timely degree of accurate judgement decision.There is operating mode dynamic and judges decision promptness PD in dangerTWWMeter Divider is then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There is total 46 of operating mode and sentence in 32 major class danger Disconnected decision scene, other traffic participants keep original driving intention, it is desirable that vehicle is sentenced within the defined time (0.5s) Disconnected decision, each scene judge that decision error or non-decision out detain 3 points, minimum 0 point, judge output character after decision As a result.
Danger release operating mode dynamic is judged into decision accuracy PDCWA, dangerous release operating mode dynamic judge decision stability PDRWA, dangerous release operating mode dynamic judge decision promptness PDTWAInitial value be respectively set to 100 points.
Danger release operating mode dynamic judges that decision accuracy refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, judges the order of accuarcy of decision.Danger release operating mode dynamic judges decision accuracy PDCWAScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：Total 5 of 5 major class danger release operating mode judges decision Scene, other traffic participants keep original driving intention, it is desirable that vehicle carries out judgement decision, each field before the deadline Scape judges that decision error or non-decision out detain 20 points, minimum 0 point, judges output character result after decision.
Danger release operating mode dynamic judges that decision stability refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, makes the degree of stability of accurate judgement decision.Danger release operating mode dynamic judges decision stability PDRWAMeter Divider is then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：5 major class danger discharge total 5 judgements of operating mode Decision scene, other traffic participants change original driving intention, it is desirable that vehicle carries out judgement decision before the deadline, often A scene judges that decision error or non-decision out detain 20 points, minimum 0 point, judges output character result after decision.
Danger release operating mode dynamic judges that decision promptness refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, makes the timely degree of accurate judgement decision.Danger release operating mode dynamic judges decision promptness PDTWAMeter Divider is then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：5 major class danger discharge total 5 judgements of operating mode Decision scene, other traffic participants keep original driving intention, it is desirable that vehicle is judged within the defined time (0.5s) Decision, each scene judge that decision error or non-decision out detain 20 points, minimum 0 point, judge output character after decision As a result.
The control for obtaining automatic driving vehicle executes parameter K.
Automatic driving vehicle must have accurately control executive capability, by environment sensing, network link, judge decision It is converted into the driving behavior of vehicle accurate safety on road.The control of automatic driving vehicle executes parameter and is driven automatically for evaluating Sail the control execution performance of vehicle.
It includes that static cost control execution test and appraisal parameter and dynamic control execute test and appraisal that the control of automatic driving vehicle, which executes parameter, Parameter.
Static cost control executes test and appraisal parameter：The reaction of the response speed, precision and system totality of executing agency Time is surveyed mainly for the control execution performance of the brake-by-wire of intelligent network connection automobile, drive-by-wire, wire-controlled steering system It comments.When test and appraisal, need for respective system line range for the national games divide equally generate 1000 test response input and keep loading it is constant, In addition 1000 test responses are generated and input and coordinate the variation of load, it is whether suitable or just according to the control implementing result of system Really evaluated and tested.
Dynamic control executes the dynamic control executive capability of test and appraisal parameter emphasis evaluation and test intelligent network connection automobile, is existed by vehicle Under motion state, the control executive capability of different operating modes is evaluated and tested.By evaluate and test state of motion of vehicle response time and Precision (for example, stop position and the time, the time of acceleration, steering precision etc.), track keep precision, manipulation task it is complete It is obtained at rate, from the time etc. perceived spent by execution.
Before the control for determining automatic driving vehicle executes parameter K, it is thus necessary to determine that following parameter：Control executes accuracy Parameter of testing and assessing KN, control execute stability test and appraisal parameter KWWPromptness test and appraisal parameter KW is executed with controlA.Control executes accuracy Parameter of testing and assessing KN, control execute stability test and appraisal parameter KWWPromptness test and appraisal parameter KW is executed with controlARespectively by respectively corresponding to Static cost control execute test and appraisal parameter and dynamic control execute test and appraisal parameter determine, wherein dynamic control execute test and appraisal parameter packet It includes：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger.Specifically, quiet State control executes test and appraisal parameter：Static cost control executes accuracy KJC, static cost control execute stability KJRIt is held with static cost control Row promptness KJT.Dynamic control executes test and appraisal parameter：Normal driving operating mode dynamic control executes accuracy KDCN, normally drive It sails operating mode dynamic control and executes stability KDRN, normal driving operating mode dynamic control execute promptness KDTN；It is dynamic that operating mode occurs in danger State control executes accuracy KDCWW, there is operating mode dynamic control and executes stability KD in dangerRWW, danger there is operating mode dynamic control Execute promptness KDTWW；Danger release operating mode dynamic control executes accuracy KDCWA, dangerous release operating mode dynamic control execute it is steady Surely KD is spentRWA, dangerous release operating mode dynamic control execute promptness KDTWA。
As it can be seen that the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention is by combining static cost control to execute Test and appraisal parameter and dynamic control execute test and appraisal parameter and execute parameter K to calculate the control of automatic driving vehicle, can be from multi-angle The behavior safety for considering automatic driving vehicle is effectively improved the accuracy and reliability of evaluating result.
After obtaining above-mentioned parameter, the control that automatic driving vehicle is obtained according to following formula executes parameter K：
K=KN × 0.4+KWW×0.4+KWA×0.2 (4)
Utilize each parameter in following various acquisition formula (4)：
KN=KCN×0.4+KRN×0.3+KTN×0.3
KWW=KCWW×0.4+KRWW×0.3+KTWW×0.3
KWA=KCWA×0.4+KRWA×0.3+KTWA×0.3
Wherein, KCNAccuracy test and appraisal parameter, KR are executed for normal driving operating conditionsNThere is operating conditions execution for danger Accuracy test and appraisal parameter, KTNAccuracy test and appraisal parameter, KC are executed for danger release operating conditionsWWFor normal driving operating conditions Execute stability test and appraisal parameter, KRWWThere are operating conditions for danger and executes stability test and appraisal parameter, KTWWOperating mode is discharged for danger Control executes stability test and appraisal parameter, KCWAPromptness test and appraisal parameter, KR are executed for normal driving operating conditionsWAOccur for danger Operating conditions execute promptness test and appraisal parameter, KTWAPromptness test and appraisal parameter is executed for danger release operating conditions.
Utilize the following various each parameter obtained in above-mentioned expression formula：
The specific acquisition process of above-mentioned parameter is as follows：
Static cost control is executed into accuracy KJC, static cost control execute stability KJR, static cost control execute promptness KJT's Initial value is respectively set to 100 points.
Static cost control executes accuracy：The order of accuarcy that static automatic driving vehicle executes scenery control.It is quiet State control executes accuracy KJCScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance：1000 tests are rung It should input and keep loading constant scene, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake It accidentally or is not carried out out 0.1 point of button, minimum 0 point, controls output character result after executing.
Static cost control executes stability：When scene changes, static automatic driving vehicle is to scenery control Execute correct degree of stability.Static cost control executes stability KJRScoring rule：Static evaluation and test vehicle front certain distance (20 meters) appearance：1000 test responses input and coordinate the variation of load, it is desirable that and vehicle controls execution before the deadline, Each scenery control executes mistake or is not carried out out 0.1 point of button, minimum 0 point, output character result after control executes.
Static cost control executes promptness：Static automatic driving vehicle executes correct and time-histories to scenery control Degree.Static cost control executes promptness KJTScoring rule：Static (20 meters) appearance of evaluation and test vehicle front certain distance：1000 Test response inputs and keeps loading constant scene, it is desirable that vehicle control within the defined time (100ms) executes, each field Scape control executes mistake or is not carried out out 0.1 point of button, minimum 0 point, output character result after control executes.
Normal driving operating mode dynamic control is executed into accuracy KDCN, normal driving operating mode dynamic control execute stability KDRN, normal driving operating mode dynamic control execute promptness KDTNInitial value be respectively set to 100 points.
Normal driving operating mode dynamic control executes accuracy：Automatic driving vehicle is in normal driving process, to not The order of accuarcy executed with scenery control.Normal driving operating mode dynamic control executes accuracy KDCNScoring rule：Normally travel Evaluate and test (20 meters) appearance of vehicle front certain distance：Total 9 control of 9 major class normal driving operating modes executes scene, pavement conditions Not changing, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake or is not carried out out 15 points of button, Minimum 0 point, output character result after control executes.
Normal driving operating mode dynamic control executes stability：Automatic driving vehicle is in normal driving process, on the spot When scape changes, control executes correct degree of stability.Normal driving operating mode dynamic control executes stability KDRNScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：Total 9 control of 9 major class normal driving operating modes executes Scene, pavement conditions have mutation, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake or do not hold Row out detains 15 points, minimum 0 point, controls output character result after executing.
Normal driving operating mode dynamic control executes promptness：Automatic driving vehicle is in normal driving process, to not Correctly timely degree is executed with scenery control.Normal driving operating mode dynamic control executes promptness KDTNScoring rule：Normal row (20 meters) appearance of the evaluation and test vehicle front certain distance sailed：Total 9 control of 9 major class normal driving operating modes executes scene, road surface Condition does not change, it is desirable that vehicle control within the defined time (0.5s) executes, and each scenery control executes mistake or is not carried out 15 points are out detained, minimum 0 point, controls output character result after executing.
Is there is into operating mode dynamic control in danger and executes accuracy KDCWW, danger occur operating mode dynamic control execute stability KDRWW, there is operating mode dynamic control and executes promptness KD in dangerTWWInitial value be respectively set to 100 points.
There is the execution accuracy of operating mode dynamic control in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, controls the order of accuarcy of execution.There is operating mode dynamic control and executes accuracy KD in dangerCWWScoring rule：Just (20 meters) appearance of the evaluation and test vehicle front certain distance often travelled：There is total 46 control of operating mode and execute field in 32 major class danger Scape, pavement conditions do not change, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake or is not carried out 3 points are out detained, minimum 0 point, controls output character result after executing.
There is operating mode dynamic control execution stability in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, and control executes correct degree of stability.There is operating mode dynamic control and executes stability KD in dangerRWWScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There are total 46 controls of operating mode and hold in 32 major class danger Row scene, pavement conditions have mutation, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake or not Out 3 points of button is executed, minimum 0 point, controls output character result after executing.
There is operating mode dynamic control execution promptness in danger：Automatic driving vehicle is being endangered in normal driving process Danger occurs under operating mode, and control executes correctly timely degree.There is operating mode dynamic control and executes promptness KD in dangerTWWScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：There are total 46 controls of operating mode and hold in 32 major class danger Row scene, pavement conditions do not change, it is desirable that vehicle control within the defined time (0.5s) executes, and each scenery control executes mistake It accidentally or is not carried out out 3 points of button, minimum 0 point, controls output character result after executing.
Danger release operating mode dynamic control is executed into accuracy KDCWA, dangerous release operating mode dynamic control execute stability KDRWA, dangerous release operating mode dynamic control execute promptness KDTWAInitial value be respectively set to 100 points.
Danger release operating mode dynamic control executes accuracy and refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, controls the order of accuarcy of execution.Danger release operating mode dynamic control executes accuracy KDCWAScoring rule：Just (20 meters) appearance of the evaluation and test vehicle front certain distance often travelled：5 major class danger discharge total 5 control of operating mode and execute scene, Pavement conditions do not change, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake or is not carried out out Detain 20 points, minimum 0 point, output character result after control executes.
Danger release operating mode dynamic control executes stability and refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, control executes correct degree of stability.Danger release operating mode dynamic control executes stability KDRWAScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：5 major class danger discharge total 5 control of operating mode and execute Scene, pavement conditions have mutation, it is desirable that vehicle controls execution before the deadline, and each scenery control executes mistake or do not hold Row out detains 20 points, minimum 0 point, controls output character result after executing.
Danger release operating mode dynamic control executes promptness and refers to：Automatic driving vehicle is being released in normal driving process It puts under dangerous situation, control executes correctly timely degree.Danger release operating mode dynamic control executes promptness KDTWAScore rule Then：(20 meters) appearance of evaluation and test vehicle front certain distance of normally travel：5 major class danger discharge total 5 control of operating mode and execute Scene, pavement conditions do not change, it is desirable that vehicle control within the defined time (0.2s) executes, and each scenery control executes mistake Or it is not carried out out 20 points of button, and minimum 0 point, output character result after control executes.
The net connection degree for joining automobile according to intelligent network determines net connection degree coefficient a, b.
Then, step S102 is executed.Based on environment sensing parameter H, network link parameter L, judge decision parameters P, control Parameter K and net connection degree coefficient a, b are executed, the test and appraisal score Z of automatic driving vehicle behavior safety is obtained according to following formula：
In expression formula (5), the concrete numerical value of net connection degree coefficient a is determined by the net connection degree of intelligent network connection automobile, and a+b =1.
Finally, step S103 is executed.According to the test and appraisal score of automatic driving vehicle behavior safety, automatic driving vehicle is obtained The evaluating result of behavior safety.
Preferably, according to the magnitude relationship of the test and appraisal score and preset score threshold of automatic driving vehicle behavior safety, Obtain the evaluating result of automatic driving vehicle behavior safety.
Specifically, when the test and appraisal score Z of automatic driving vehicle behavior safety be more than or equal to A1 timesharing, be determined as automatic Pilot Vehicle behavior has preferable safety；When the test and appraisal score Z of automatic driving vehicle behavior safety is more than or equal to B1 points and is less than A1 timesharing is determined as that automatic driving vehicle behavior has certain safety；When testing and assessing for automatic driving vehicle behavior safety Divide Z to be less than B1 timesharing, is determined as that automatic driving vehicle behavior has poor safety.
Using the assessment method of automatic driving vehicle behavior safety provided in an embodiment of the present invention, by the influence of acquisition from The dynamic environment sensing parameter for driving vehicle behavior safety, network link parameter judge decision parameters, control execution parameter and automatically The net for driving vehicle joins degree coefficient, obtains the test and appraisal score of automatic driving vehicle behavior safety, is obtained according to the test and appraisal score The evaluating result of automatic driving vehicle behavior safety.Therefore, the assessment method energy of automatic driving vehicle behavior safety of the invention Enough behavior safeties that automatic driving vehicle is considered from multi-angle comprehensive keep evaluating result accurate, reliable, to be automatic Pilot Vehicle behavior safety test and evaluation provide a kind of possible ways, keep associated test standards perfect, and China is promoted to drive automatically Sail the development of Vehicle Industry.
Also, in the assessment method of the automatic driving vehicle behavior safety of the embodiment of the present invention, the environment sensing of acquisition Parameter, network link parameter judge that decision parameters and control execution parameter are respectively by combining corresponding static test and appraisal The characteristic of parameter and dynamic measuring and evaluating parameter and obtain.Therefore, the present invention can consider automatic driving vehicle from multi-angle comprehensive Behavior safety is effectively improved the accuracy and reliability of evaluating result.
While it is disclosed that embodiment content as above but described only to facilitate understanding the present invention and adopting Embodiment is not limited to the present invention.Any those skilled in the art to which this invention pertains are not departing from this Under the premise of the disclosed spirit and scope of invention, any modification and change can be made in the implementing form and in details, But protection scope of the present invention still should be subject to the scope of the claims as defined in the appended claims.

Claims (11)

1. a kind of assessment method of automatic driving vehicle behavior safety, which is characterized in that including：
It obtains the environment sensing parameter of automatic driving vehicle, network link parameter, judge that decision parameters, control execute parameter and net Join degree coefficient；
According to the environment sensing parameter, the network link parameter, the judgement decision parameters, the control execute parameter and The net joins degree coefficient, obtains the test and appraisal score of automatic driving vehicle behavior safety；
According to the test and appraisal score of the automatic driving vehicle behavior safety, the test and appraisal of the automatic driving vehicle behavior safety are obtained As a result.
2. the assessment method of automatic driving vehicle behavior safety according to claim 1, which is characterized in that according to following table The test and appraisal score is obtained up to formula：
Wherein, Z is the test and appraisal score, and H is the environment sensing parameter, and L is the network link parameter, and P is the judgement Decision parameters, K are that the control executes parameter, and a, b are that the net joins degree coefficient, and a+b=1.
3. the assessment method of automatic driving vehicle behavior safety according to claim 2, which is characterized in that according to the survey The magnitude relationship for commenting score and preset score threshold obtains the evaluating result of the automatic driving vehicle behavior safety.
4. the assessment method of automatic driving vehicle behavior safety according to any one of claim 1 to 3, feature exist In obtaining the environment sensing parameter H of the automatic driving vehicle according to following formula：
H=HN × 0.4+HWW×0.4+HWA×0.2
Wherein, HN is environment sensing accuracy test and appraisal parameter, HWWFor environment sensing stability test and appraisal parameter, HWAFor environment sensing Promptness test and appraisal parameter.
5. the assessment method of automatic driving vehicle behavior safety according to claim 4, which is characterized in that the environment sense Know that accuracy test and appraisal parameter, environment sensing stability test and appraisal parameter and environment sensing promptness test and assess parameter respectively by respectively corresponding to Static environment perception test and appraisal parameter and dynamic environment perception test and appraisal parameter determine, wherein the dynamic environment perception test and appraisal parameter Including：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger.
6. the assessment method of automatic driving vehicle behavior safety according to any one of claim 1 to 3, feature exist In obtaining the network link parameter L of the automatic driving vehicle according to following formula：
L=LN × 0.4+LWW×0.4+LWA×0.2
Wherein, LN is network link accuracy test and appraisal parameter, LWWFor network link stability test and appraisal parameter, LWAFor network link Promptness test and appraisal parameter.
7. the assessment method of automatic driving vehicle behavior safety according to claim 6, which is characterized in that the network connection Logical accuracy test and appraisal parameter, network link stability test and appraisal parameter and network link promptness test and assess parameter respectively by respectively corresponding to Static network unicom test and appraisal parameter and dynamic network unicom test and appraisal parameter determine, wherein the dynamic network unicom test and assess parameter Including：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger.
8. the assessment method of automatic driving vehicle behavior safety according to any one of claim 1 to 3, feature exist In obtaining the judgement decision parameters P of the automatic driving vehicle according to following formula：
P=PN × 0.2+PWW×0.6+PWA×0.2
Wherein, PN is to judge decision accuracy test and appraisal parameter, PWWTo judge decision stability test and appraisal parameter, PWATo judge decision Promptness test and appraisal parameter.
9. the assessment method of automatic driving vehicle behavior safety according to claim 8, which is characterized in that the judgement is determined Plan accuracy test and appraisal parameter judges decision stability test and appraisal parameter and judges that decision promptness tests and assesses parameter respectively by respectively corresponding to Static state judge decision test and appraisal parameter and dynamic judge decision test and appraisal parameter determination, wherein the dynamic judge decision test and appraisal parameter Including：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger.
10. the assessment method of automatic driving vehicle behavior safety according to any one of claim 1 to 3, feature exist In the control for obtaining the automatic driving vehicle according to following formula executes parameter K：
K=KN × 0.4+KWW×0.4+KWA×0.2
Wherein, KN executes accuracy test and appraisal parameter, KW in order to controlWStability test and appraisal parameter, KW are executed in order to controlAIt executes in order to control Promptness test and appraisal parameter.
11. the assessment method of automatic driving vehicle behavior safety according to claim 10, which is characterized in that the control Execute accuracy test and appraisal parameter, control executes stability test and appraisal parameter and control executes promptness test and appraisal parameter respectively by respectively right The static cost control answered executes test and appraisal parameter and dynamic control executes test and appraisal parameter and determines, wherein the dynamic control executes test and appraisal ginseng Number includes：There is operating mode test and appraisal parameter and dangerous release operating mode test and appraisal parameter in normal driving operating mode test and appraisal parameter, danger.
CN201711348568.2A 2017-12-15 2017-12-15 Evaluation method for behavior safety of automatic driving vehicle Active CN108304986B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711348568.2A CN108304986B (en) 2017-12-15 2017-12-15 Evaluation method for behavior safety of automatic driving vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711348568.2A CN108304986B (en) 2017-12-15 2017-12-15 Evaluation method for behavior safety of automatic driving vehicle

Publications (2)

Publication Number Publication Date
CN108304986A true CN108304986A (en) 2018-07-20
CN108304986B CN108304986B (en) 2021-06-29

Family

ID=62870060

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711348568.2A Active CN108304986B (en) 2017-12-15 2017-12-15 Evaluation method for behavior safety of automatic driving vehicle

Country Status (1)

Country Link
CN (1) CN108304986B (en)

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109064763A (en) * 2018-10-23 2018-12-21 百度在线网络技术（北京）有限公司 Test method, device, test equipment and the storage medium of automatic driving vehicle
CN109242251A (en) * 2018-08-03 2019-01-18 百度在线网络技术（北京）有限公司 Vehicular behavior safety detecting method, device, equipment and storage medium
CN109614886A (en) * 2018-11-23 2019-04-12 中国人民解放军63966部队 It is a kind of nobody/performance evaluation methodology and device of autonomous driving system
CN109753065A (en) * 2018-12-29 2019-05-14 百度在线网络技术（北京）有限公司 The environmental modeling capability assessment method and system of automatic driving vehicle
CN110108505A (en) * 2019-06-06 2019-08-09 中国汽车工程研究院股份有限公司 Automatic Pilot assists lamp stand test fixture vehicle and its working method
CN110245837A (en) * 2019-05-20 2019-09-17 西安交通大学 Driving behavior normalization evaluation system and evaluation method when a kind of vehicle crosses deceleration strip
CN110414803A (en) * 2019-07-08 2019-11-05 清华大学 The assessment method and device of automated driving system level of intelligence under different net connection degree
CN110488671A (en) * 2019-05-13 2019-11-22 中国汽车工程研究院股份有限公司 A kind of automatic Pilot data processing method and its system
CN111178735A (en) * 2019-12-24 2020-05-19 国汽(北京)智能网联汽车研究院有限公司 Test evaluation method, device and system for automatic driving function
CN111175055A (en) * 2018-11-09 2020-05-19 百度在线网络技术（北京）有限公司 Automatic driving distributed collaborative simulation method and device and terminal
CN111186446A (en) * 2018-10-29 2020-05-22 中国电信股份有限公司 Control method, apparatus and computer-readable storage medium for automatic driving
WO2021171112A1 (en) * 2020-02-28 2021-09-02 International Business Machines Corporation Autonomous driving evaluation using data analysis
CN116056034A (en) * 2023-04-03 2023-05-02 鹰驾科技（深圳）有限公司 Vehicle-mounted wireless communication analysis regulation and control system based on in-vehicle environment monitoring
US11644331B2 (en) 2020-02-28 2023-05-09 International Business Machines Corporation Probe data generating system for simulator
US11702101B2 (en) 2020-02-28 2023-07-18 International Business Machines Corporation Automatic scenario generator using a computer for autonomous driving

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102829980A (en) * 2012-03-29 2012-12-19 中国科学院自动化研究所 Intelligent degree evaluation method of intelligent vehicle
US20130002415A1 (en) * 2011-06-28 2013-01-03 Geotab Inc. Vehicular telematics device with voltage sensor-predicated GPS reciever activation
CN202892854U (en) * 2012-06-14 2013-04-24 北京理工大学 Automatic analysis processing and automatic grading device for unmanned vehicle race data
CN103234763A (en) * 2013-04-09 2013-08-07 北京理工大学 System and method for quantitatively evaluating unmanned vehicles
US8612135B1 (en) * 2012-02-14 2013-12-17 Google Inc. Method and apparatus to localize an autonomous vehicle using convolution

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130002415A1 (en) * 2011-06-28 2013-01-03 Geotab Inc. Vehicular telematics device with voltage sensor-predicated GPS reciever activation
US8612135B1 (en) * 2012-02-14 2013-12-17 Google Inc. Method and apparatus to localize an autonomous vehicle using convolution
CN102829980A (en) * 2012-03-29 2012-12-19 中国科学院自动化研究所 Intelligent degree evaluation method of intelligent vehicle
CN202892854U (en) * 2012-06-14 2013-04-24 北京理工大学 Automatic analysis processing and automatic grading device for unmanned vehicle race data
CN103234763A (en) * 2013-04-09 2013-08-07 北京理工大学 System and method for quantitatively evaluating unmanned vehicles

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109242251A (en) * 2018-08-03 2019-01-18 百度在线网络技术（北京）有限公司 Vehicular behavior safety detecting method, device, equipment and storage medium
US11084494B2 (en) 2018-08-03 2021-08-10 Baidu Online Network Technology (Beijing) Co., Ltd. Method for detecting safety of driving behavior, apparatus, device and storage medium
CN109064763A (en) * 2018-10-23 2018-12-21 百度在线网络技术（北京）有限公司 Test method, device, test equipment and the storage medium of automatic driving vehicle
CN111186446B (en) * 2018-10-29 2022-02-01 中国电信股份有限公司 Control method, apparatus and computer-readable storage medium for automatic driving
CN111186446A (en) * 2018-10-29 2020-05-22 中国电信股份有限公司 Control method, apparatus and computer-readable storage medium for automatic driving
CN111175055A (en) * 2018-11-09 2020-05-19 百度在线网络技术（北京）有限公司 Automatic driving distributed collaborative simulation method and device and terminal
CN111175055B (en) * 2018-11-09 2022-08-09 百度在线网络技术（北京）有限公司 Automatic driving distributed collaborative simulation method and device and terminal
CN109614886A (en) * 2018-11-23 2019-04-12 中国人民解放军63966部队 It is a kind of nobody/performance evaluation methodology and device of autonomous driving system
CN109753065B (en) * 2018-12-29 2022-06-17 百度在线网络技术（北京）有限公司 Environment modeling capability evaluation method and system for automatic driving vehicle
CN109753065A (en) * 2018-12-29 2019-05-14 百度在线网络技术（北京）有限公司 The environmental modeling capability assessment method and system of automatic driving vehicle
CN110488671A (en) * 2019-05-13 2019-11-22 中国汽车工程研究院股份有限公司 A kind of automatic Pilot data processing method and its system
CN110488671B (en) * 2019-05-13 2020-08-07 中国汽车工程研究院股份有限公司 Automatic driving data processing method and system
CN110245837A (en) * 2019-05-20 2019-09-17 西安交通大学 Driving behavior normalization evaluation system and evaluation method when a kind of vehicle crosses deceleration strip
CN110108505A (en) * 2019-06-06 2019-08-09 中国汽车工程研究院股份有限公司 Automatic Pilot assists lamp stand test fixture vehicle and its working method
CN110414803A (en) * 2019-07-08 2019-11-05 清华大学 The assessment method and device of automated driving system level of intelligence under different net connection degree
CN111178735A (en) * 2019-12-24 2020-05-19 国汽(北京)智能网联汽车研究院有限公司 Test evaluation method, device and system for automatic driving function
CN111178735B (en) * 2019-12-24 2024-02-02 国汽(北京)智能网联汽车研究院有限公司 Test evaluation method, device and system for automatic driving function
WO2021171112A1 (en) * 2020-02-28 2021-09-02 International Business Machines Corporation Autonomous driving evaluation using data analysis
GB2611632A (en) * 2020-02-28 2023-04-12 Ibm Autonomous driving evaluation using data analysis
US11644331B2 (en) 2020-02-28 2023-05-09 International Business Machines Corporation Probe data generating system for simulator
US11702101B2 (en) 2020-02-28 2023-07-18 International Business Machines Corporation Automatic scenario generator using a computer for autonomous driving
US11814080B2 (en) 2020-02-28 2023-11-14 International Business Machines Corporation Autonomous driving evaluation using data analysis
CN116056034A (en) * 2023-04-03 2023-05-02 鹰驾科技（深圳）有限公司 Vehicle-mounted wireless communication analysis regulation and control system based on in-vehicle environment monitoring

Also Published As

Publication number Publication date
CN108304986B (en) 2021-06-29

Similar Documents

Publication Publication Date Title
CN108304986A (en) A kind of assessment method of automatic driving vehicle behavior safety
CN109213126B (en) Automatic driving automobile test system and method
CN110248311B (en) V2I application function testing method based on intelligent networking platform
CN111061167B (en) Intelligent network connection demonstration area-oriented mixed reality automatic driving test method and virtual test platform
CN111781855B (en) Traffic is automatic driving simulation system in ring
CN109100155A (en) A kind of unmanned vehicle is in ring high-speed simulation test macro and method
CN103335853B (en) A kind of automatic driving vehicle Cognitive Aptitude Test system and method
CN106781493B (en) Vehicle queuing length simulation system, method and device
Zhou et al. A framework for virtual testing of ADAS
CN106598053B (en) A kind of automatic driving vehicle transverse movement control object system of selection
CN107543726B (en) A kind of test method and device of unmanned vehicle Ride Control System manipulation precision
CN108645628A (en) Autonomous driving vehicle based on driving path technical ability tests system
CN206601241U (en) The checkout area of automatic driving vehicle lane-change ability
CN106873397A (en) Intelligent network joins automobile " hardware in loop " accelerated loading emulation test system
CN108137050A (en) Steering control device, driving control method and program
CN108137052A (en) Steering control device, driving control method and program
CN106706334A (en) Test method and test field for capacity of autonomous driving vehicle in avoiding approaching vehicle
CN109064763A (en) Test method, device, test equipment and the storage medium of automatic driving vehicle
CN113340615B (en) Automobile automatic driving function simulation test system and method
CN109975033A (en) A kind of automatic Pilot dynamic testing method and device
CN113887037B (en) Automatic driving system evaluation method under network environment with different permeability
Thielen et al. A feasibility study on a cooperative safety application for cyclists crossing intersections
Gelbal et al. Hardware-in-the-Loop and Road Testing of RLVW and GLOSA Connected Vehicle Applications
CN108447306B (en) Simulation method for real-time position sharing collision avoidance early warning among collision vehicles at uncontrolled intersection
Belz et al. Event-based framework for noncompliant driver behavior at single-lane roundabouts

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant