CN110182217B - Running task complexity quantitative evaluation method oriented to complex overtaking scene - Google Patents
Running task complexity quantitative evaluation method oriented to complex overtaking scene Download PDFInfo
- Publication number
- CN110182217B CN110182217B CN201910328180.9A CN201910328180A CN110182217B CN 110182217 B CN110182217 B CN 110182217B CN 201910328180 A CN201910328180 A CN 201910328180A CN 110182217 B CN110182217 B CN 110182217B
- Authority
- CN
- China
- Prior art keywords
- overtaking
- probability
- decision
- vehicle
- scene
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000011158 quantitative evaluation Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000006399 behavior Effects 0.000 claims description 24
- 238000012360 testing method Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000001133 acceleration Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000007619 statistical method Methods 0.000 claims description 4
- 230000002457 bidirectional effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 10
- 238000011160 research Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/0097—Predicting future conditions
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
The invention discloses a running task complexity quantitative evaluation method facing a complex overtaking scene in the technical field of automatic driving, through the combination of the implementation flow of the running task complexity quantitative evaluation method of the complex overtaking scene and the implementation formula of the running task complexity quantitative evaluation method facing the complex overtaking scene, the problem that the existing environment task complexity quantitative evaluation method does not pay attention to the mutual influence of the road user and the running task of the main vehicle is solved, and a quantitative evaluation method for the complexity of the 'overtaking' driving task in a complex conflict scene is not provided, so that the problem that the influence of the existing algorithm on the complexity of the primary form task of the 'overtaking' decision of the main vehicle can not be quantitatively calculated, therefore, the influence of road users on the task complexity of the primary form of the 'overtaking' decision of the main vehicle is quantitatively calculated.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a running task complexity quantitative evaluation method for a complex overtaking scene.
Background
The automobile industry is an important component of national economy in China, outstanding contribution is made to steady and rapid increase of national economy, along with rapid development of the intelligent automobile industry, the market scale is increasingly large, developed countries are in a dispute and are in a leading position, the intelligent automobile industry in China starts relatively late, but the industry is rapidly developed, at present, the establishment and specification of intelligent automobile standards are bottlenecks for restricting the development of intelligent automobiles, and the research of the environment task complexity quantitative evaluation method can provide scientific basis and foundation for the establishment of the intelligent automobile standards.
The environment task complexity quantitative evaluation research is divided into two aspects of running environment complexity research and running task complexity research, wherein the running task complexity is difficult to carry out relatively comprehensive and objective quantitative evaluation on the vehicle running task due to the uncertainty of road users, road types, driving targets, constraints and the like and the uncertainty of related traffic rules.
In the current quantitative evaluation of the running task complexity, a traditional statistical method is still used, the elements such as the road attribute, the road user, the traffic control and the meteorological condition of the current automobile running need to be deconstructed and classified manually at the background, the prior knowledge of each element in the database on the running task complexity is contrasted, the current automobile running task complexity is quantitatively evaluated through weighting summation, a large amount of manpower is consumed to classify the running scenes and the elements, the influence of the road user on the running task complexity of the main automobile is not considered, the complexity quantization precision is extremely low, and the automatic driving technical standard is not easy to make and standardize.
The existing running task complexity quantitative evaluation method cannot describe the task complexity in 'overtaking' running, and the core functional modules required by the normal running of the intelligent automobile on the road are as follows: the system comprises a perception module, a decision-making module and a control module, wherein a behavior decision-making technology in the decision-making module is a key technology for restricting the development of automatic driving, and a overtaking driving task relates to typical complex scene series-parallel decision, and comprises the following steps: if a traditional quantitative evaluation method for the complexity of the driving task based on scene deconstruction is adopted, the complexity of the 'overtaking' driving task cannot be calculated quantitatively, and the influence of a road user on the 'overtaking' decision of a main vehicle and the complexity of the driving task cannot be calculated quantitatively by the conventional algorithm.
Disclosure of Invention
The invention aims to provide a running task complexity quantitative evaluation method for a complex overtaking scene, and aims to solve the problems that the existing environmental task complexity quantitative evaluation method proposed in the background technology does not pay attention to the mutual influence of a road user and a main vehicle running task, and the existing algorithm cannot quantitatively calculate the influence of the road user on the main vehicle 'overtaking' decision-making primary form task complexity due to the fact that the existing algorithm does not propose the quantitative evaluation method for the 'overtaking' running task complexity under the complex conflict scene.
In order to achieve the purpose, the invention provides the following technical scheme: a running task complexity quantitative evaluation method for a complex overtaking scene is implemented by the following flows:
1) acquisition of "overtaking" scene prior knowledge
Using SCANER studio to establish overtaking scene, and obstacle vehicle B is in front of main vehicle A by XabThe longitudinal distance between a conflicted vehicle C in the opposite traffic flow and an obstacle vehicle B is XbcMeter, setting the initial speed of the obstacle vehicle B to Vbkm/h, initial speed of subject vehicle A Vckm/h, initial velocity of colliding vehicle C is Vakm/h, with a scene function of S (X)ab2Xbc2Va2Vb2Vc) Indicating a passing scene, V, set from the above initial statea、Vb、VcThe speed selection depends on the traffic regulation speed limit value of the corresponding road, x drivers with the driving ages of 3-5 years are selected, three driving simulators are used for carrying out combined simulation in the built scene at the same time, and in order to utilize human resources to the maximum degree, the x drivers can design A by utilizing the permutation and combination principlex xGroup tests, each group of tests was repeated y times, totaling y Ax xThe test is performed by recording the running states of the host vehicle A, the obstacle vehicle B and the collision vehicle C in each test, and using Nb1、Nb2、Nb3、Nc1、Nc2、Nc3Respectively represents three driving behaviors of accelerating, uniform speed and decelerating of the vehicle B, C, Ma1、Ma2、Ma3、Ma4、Ma5、Ma6Respectively representing six driving behaviors of accelerating non-lane change, accelerating lane change, constant-speed non-lane change, constant-speed lane change, decelerating non-lane change and decelerating lane change of the main vehicle A;
let the scene function be S (X)ab,Xbc,Va,Vb,Vc) The decision probability P of the acceleration of the vehicle Bb,1Uniform speed running decision probability Pb,2And a deceleration driving decision probability Pb,3Acceleration decision probability P of conflicting vehicle Cc,1Uniform speed running decision probability Pc,2And a deceleration driving decision probability Pc,3Setting the driving behaviors of the obstacle vehicle B and the conflict vehicle C as N respectivelybi、NcjUnder the scene, the driving behavior of the main vehicle A is MakHas a decision probability of Pk|ijThe calculation formula is as follows:
changing the scene function S (X)ab,Xbc,Va,Vb,Vc) Initial values of variables, S (X) being obtained using the above calculation methodab,Xbc,Va,Vb,Vc) And Pb,i、Pc,j、Pk|ijThe mapping table of (1), namely the overtaking scene prior knowledge;
2) "overtaking" decision probability extraction
Reading in a specific overtaking driving case aiming at a 'overtaking' driving task in a bidirectional two-lane conflict scene, respectively calculating an accelerated driving decision probability, a constant speed driving decision probability and a decelerated driving decision probability of a front obstacle vehicle B and a lane change conflict vehicle C at the current moment through priori knowledge, establishing a mixed decision joint probability of the overtaking scene of a main vehicle A at the current moment based on a statistical analysis theory, extracting a conditional probability of each decision of the main vehicle A by utilizing the priori knowledge, and calculating an accelerated non-lane-changing decision probability, an accelerated lane-changing decision probability, a constant speed non-lane-changing decision probability, a constant speed lane-changing decision probability, a decelerated non-lane-changing decision probability and a decelerated lane-changing decision probability of the main vehicle A under each decision of the overtaking scene at the current moment;
3) quantitative evaluation of 'overtaking' driving task complexity
The method comprises the steps of counting the occurrence probability of each behavior decision of a main vehicle A in a overtaking scene, calculating the average decision probability of overtaking, selecting a reasonable task complexity normalization parameter, establishing a task complexity function of the main vehicle A in the overtaking scene at the current time by utilizing the principle that the larger the covariance of the actual behavior decision probability and the average decision probability of the main vehicle A is, the higher the complexity of the overtaking driving task at the current time is, popularizing the task complexity quantitative evaluation method to the complex overtaking scene, and selecting the maximum value of the task complexity function for quantitatively evaluating the overall complexity of the overtaking driving task.
Preferably, the implementation formula of the running task complexity quantitative evaluation method for the complex overtaking scene is as follows:
reading tpScene of overtaking at any moment
(1) Determining t from a priori knowledgepAcceleration driving decision probability P of preceding obstacle vehicle B at timeb,1(tp) Uniform speed running decision probability Pb,2(tp) And a deceleration driving decision probability Pb,3(tp);
(2) Determining t from a priori knowledgepAcceleration driving decision probability P of vehicle C with conflict of lane change at any momentc,1(tp) Uniform speed running decision probability Pc,2(tp) And a deceleration driving decision probability Pc,3(tp);
(3) Calculating t according to the accelerated driving decision probability, the uniform speed driving decision probability and the decelerated driving decision probability of the front obstacle vehicle B and the lane change conflict vehicle CpDecision joint probability P of time overtaking scenebc,ij(tp) Where i is {1,2,3}, and j is {1,2,3}, the following formula:
Pbc,ij(tp)=Pc,i(tp)×Pc,j(tp) (4);
(4) known calculation of tpLower master of each decision joint probability in time overtaking sceneConditional probability of the body vehicle A, i.e. the subject vehicle acceleration lane-change-free conditional probability Pt,1|ij(tp) Accelerated lane change probability Pt,2|ij(tp) Constant velocity non-changing track probability Pt,3|ij(tp) Uniform speed lane changing probability Pt,4|ij(tp) Deceleration non-lane-changing probability Pt,5|ij(tp) And deceleration lane change probability Pt,6|ij(tp);
(5) According to tpJoint probability P of each decision in moment overtaking scenebc,ij(tp) I ═ 1,2,3, and j ═ 1,2,3, and the conditional probability P of the subject vehicle at,k|ij(tp) And k is {1,2,3,4,5,6}, and the probability P of occurrence of each decision-making behavior of the subject vehicle A is calculatedk(tp) The formula is as follows:
(6) knowing tpThe average decision probability of the time overtaking scene is Pa(tp) 1/n, n-6 represents the actual decision number of the main vehicle A, and a task normalization parameter W is selectedc>0, quantitative calculation of tpThe overtaking task complexity of the main vehicle A in the overtaking scene at any moment is as follows:
(7) reading the overtaking scene at the next moment, wherein p + tau, tau is 1, …, N, repeating the steps (1) - (6) to calculate the overtaking task complexity (formula (3)) at each moment, and selecting the maximum value as a quantitative evaluation index of the overtaking task complexity in the overtaking process, wherein the formula is as follows:
compared with the prior art, the invention has the beneficial effects that: the invention has practical significance for the formulation and specification of the automatic driving technical standard of the intelligent automobile, and definitely provides a quantitative evaluation method of the running task complexity facing to the 'overtaking' scene.
Drawings
FIG. 1 is a decision probability extraction diagram of the present invention;
FIG. 2 is a diagram illustrating a driving complexity quantitative evaluation according to the present invention;
FIG. 3 is a cut-in view of the present invention;
FIG. 4 shows the present invention tp+1、tp+2、tp+3And (5) a moment overtaking scene schematic diagram.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the protection scope of the present invention.
The invention provides the following technical scheme: a running task complexity quantitative evaluation method for a complex overtaking scene is used for establishing a new running task complexity quantitative evaluation standard for quantitative evaluation of 'overtaking' running tasks in a two-way two-lane conflict scene, and the implementation flow of the running task complexity quantitative evaluation method for the complex overtaking scene is as follows:
1) acquisition of "overtaking" scene prior knowledge
Using SCANER studio to create a cut-in scenario As shown in FIG. 3, an obstacle vehicle B is in front of a subject vehicle A by XabThe longitudinal distance between a conflicted vehicle C in the opposite traffic flow and an obstacle vehicle B is XbcRice, setting upInitial speed of obstacle vehicle B is Vbkm/h, initial speed of subject vehicle A Vckm/h, initial velocity of colliding vehicle C is Vakm/h, with a scene function of S (X)ab2Xbc2Va2Vb2Vc) Indicating a passing scene, V, set from the above initial statea、Vb、VcThe speed selection depends on the traffic regulation speed limit value of the corresponding road, x drivers with the driving ages of 3-5 years are selected, three driving simulators are used for carrying out combined simulation in the built scene at the same time, and in order to utilize human resources to the maximum degree, the x drivers can design A by utilizing the permutation and combination principlex xGroup tests, each group of tests was repeated y times, totaling y Ax xIn the secondary test, the running states of the host vehicle a, the obstacle vehicle B, and the collision vehicle C in each test are recorded, and N is used as shown in tables 1 and 2b1、Nb2、Nb3、Nc1、Nc2、Nc3Respectively represents three driving behaviors of the vehicle B, C, namely acceleration driving, constant speed driving and deceleration driving; ma1、Ma2、Ma3、Ma4、Ma5、Ma6Respectively representing six driving behaviors of accelerating non-lane change, accelerating lane change, constant-speed non-lane change, constant-speed lane change, decelerating non-lane change and decelerating lane change of the main vehicle A;
TABLE 1 test chart
TABLE 2 summary of tests for subject vehicle A
Ma1 | Ma2 | Ma3 | Ma4 | Ma5 | Ma6 | |
Nb1Nc1 | ||||||
… | ||||||
NbiNcj | ak|ij | |||||
… | ||||||
Nb3Nc3 |
Note: a isk|ijN represents the driving behaviors of the obstacle vehicle B and the conflict vehicle Cbi、NcjUnder the scene, the driving behavior M of the subject vehicle AakThe number of hours;
let the scene function be S (X)ab,Xbc,Va,Vb,Vc) The decision probability P of the acceleration of the vehicle Bb,1Uniform speed running decision probability Pb,2And a deceleration driving decision probability Pb,3Acceleration decision probability P of conflicting vehicle Cc,1Uniform speed running decision probability Pc,2And a deceleration driving decision probability Pc,3Setting the driving behaviors of the obstacle vehicle B and the conflict vehicle C as N respectivelybi、NcjUnder the scene, the driving behavior of the main vehicle A is MakHas a decision probability of Pk|ijThe calculation formula is as follows:
changing the scene function S (X)ab,Xbc,Va,Vb,Vc) Initial values of variables, S (X) being obtained using the above calculation methodab,Xbc,Va,Vb,Vc) And Pb,i、Pc,j、Pk|ijThe mapping table of (1), namely the overtaking scene prior knowledge;
2) "overtaking" decision probability extraction
The overtaking decision probability extraction process is as shown in fig. 1, a specific overtaking driving case is read in aiming at an overtaking driving task in a bidirectional two-lane conflict scene, the accelerating driving decision probability, the constant-speed driving decision probability and the decelerating driving decision probability of a front obstacle vehicle B and a lane-changing conflict vehicle C at the current moment are respectively calculated through priori knowledge, the mixed decision joint probability of the overtaking scene of a main vehicle A at the current moment is established based on a statistical analysis theory, the conditional probability of each decision of the main vehicle A is extracted by utilizing the priori knowledge, and the accelerating non-lane-changing decision probability, the accelerating lane-changing decision probability, the constant-speed non-lane-changing decision probability, the decelerating non-lane-changing decision probability and the decelerating lane-changing decision probability of the main vehicle A at each decision of the overtaking scene at the current moment are calculated;
3) quantitative evaluation of 'overtaking' driving task complexity
The process of quantitative evaluation of the complexity of the overtaking driving task is shown in fig. 2, the occurrence probability of each behavior decision of the main vehicle A under the overtaking scene is counted, the average decision probability of overtaking is calculated, reasonable task complexity normalization parameters are selected, a task complexity function of the main vehicle A under the overtaking scene at the current moment is established by utilizing the principle that the covariance of the actual behavior decision probability and the average decision probability of the main vehicle A is larger and the complexity of the overtaking driving task at the current moment is higher, the quantitative evaluation method of the task complexity is popularized to the complex overtaking scene, and the maximum value of the task complexity function is selected to be used for quantitatively evaluating the overall complexity of the overtaking driving task.
Examples
Reading tpThe time-of-day cut-in scenario, as shown in figure 3,
determining t from a priori knowledgepAcceleration driving decision probability P of preceding obstacle vehicle B at timeb,1(tp) 20% constant speed driving decision probability Pb,2(tp) 80% and a deceleration driving decision probability Pb,3(tp)=0%;
Determining t from a priori knowledgepAcceleration driving decision probability P of vehicle C with conflict of lane change at any momentc,1(tp) 10% constant speed driving decision probability Pc,2(tp) 80% and a deceleration driving decision probability Pc,3(tp)=10%;
According to the accelerated driving decision probability, the uniform speed driving decision probability and the decelerated driving decision probability of the front obstacle vehicle B and the lane change conflict vehicle C, t is calculated according to a formula (1)pDecision joint probability P of time overtaking scenebc,ij(tp) I ═ 1,2,3, and j ═ 1,2,3, the results are as follows:
Pbc,11(tp)=2%,Pbc,12(tp)=16%,Pbc,13(tp)=2%,
Pbc,21(tp)=8%,Pbc,22(tp)=64%,Pbc,23(tp)=8%, (8)
Pbc,31(tp)=0%,Pbc,32(tp)=0%,Pbc,33(tp)=0%.
known calculation of tpConditional probability of the subject vehicle A under each decision joint probability in the time overtaking scene, namely the acceleration lane-unchanging conditional probability P of the subject vehiclet,1|11(tp)=40%、Pt,1|12(tp)=50%、Pt,1|13(tp)=25%、Pt,1|21(tp)=5%、Pt,1|22(tp)=4%、Pt,1|23(tp)=3%、Pt,1|31(tp)=Pt,1|32(tp)=Pt,1|33(tp) 0%, accelerated lane change probability Pt,2|11(tp)=10%、Pt,2|12(tp)=20%、Pt,2|13(tp)=25%、Pt,2|21(tp)=5%、Pt,2|22(tp)=6%、Pt,2|23(tp)=7%、Pt,2|31(tp)=Pt,2|32(tp)=Pt,2|33(tp) 0%, constant speed non-changing track probability Pt,3|11(tp)=40%、Pt,3|12(tp)=30%、Pt,3|13(tp)=25%、Pt,3|21(tp)=40%、Pt,3|22(tp)=32%、Pt,3|23(tp)=24%、Pt,3|31(tp)=Pt,3|32(tp)=Pt,3|33(tp) 0%, constant speed lane change probability Pt,4|11(tp)=10%、Pt,4|12(tp)=20%、Pt,4|13(tp)=25%、Pt,4|21(tp)=40%、Pt,4|22(tp)=48%、Pt,4|23(tp)=56%、Pt,4|31(tp)=Pt,4|32(tp)=Pt,4|33(tp) 0%, deceleration non-lane-change probability Pt,5|11(tp)=0%、Pt,5|12(tp)=0%、Pt,5|13(tp)=0%、Pt,5|21(tp)=5%、Pt,5|22(tp)=4%、Pt,5|23(tp)=3%、Pt,5|31(tp)=Pt,5|32(tp)=Pt,5|33(tp) 0% and deceleration lane change probability Pt,6|11(tp)=0%、Pt,6|12(tp)=0%、Pt,6|13(tp)=0%、Pt,6|21(tp)=5%、Pt,6|22(tp)=6%、Pt,6|23(tp)=7%、Pt,6|31(tp)=Pt,6|32(tp)=Pt,6|33(tp)=0%;
Knowing tpJoint probability P of each decision in moment overtaking scenebc,ij(tp) I ═ 1,2,3, and j ═ 1,2,3, and the conditional probability P of the subject vehicle at,k|ij(tp) And k is {1,2,3,4,5,6}, and the probability P of occurrence of each decision-making behavior of the subject vehicle a is calculated according to the formula (2)k(tp) The results are as follows:
knowing tpThe average decision probability of the time overtaking scene is Pa(tp) Representing the actual decision number of the main vehicle A by 16.7%, and selecting a task normalization parameter Wc=1>0, quantitative calculation of tpThe overtaking task complexity of the main vehicle A in the overtaking scene at any moment is as follows:
Jjb(tp)=2.76% (10)
reading the overtaking scenes at other moments, repeating the steps (1) to (6) to calculate the overtaking task complexity (J) at each moment as shown in figure 4jb(tp+1)=32.48%,Jjb(tp+1)=15.74%,Jjb(tp+1) 3.13%), selecting the maximum value according to the formula (4) as the quantitative evaluation index of the complexity of the overtaking process driving task, and obtaining the following results:
Jjb=32.48% (11)
although the invention has been described above with reference to some embodiments, various modifications can be made and equivalents can be substituted for elements thereof without departing from the scope of the invention, the examples merely provide an implementation description for the algorithm of the invention, and therefore, the number of samples and the driving decision prior probability involved in the actual quantitative evaluation result of the complexity of the overtaking task are not in the scope of the discussion of the embodiment case, and therefore, the invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims (2)
1. A running task complexity quantitative evaluation method for a complex overtaking scene is characterized by comprising the following steps: the implementation flow of the running task complexity quantitative evaluation method for the complex overtaking scene is as follows:
1) acquisition of "overtaking" scene prior knowledge
Using SCANER studio to establish overtaking scene, and obstacle vehicle B is in front of main vehicle A by XabThe longitudinal distance between a conflicted vehicle C in the opposite traffic flow and an obstacle vehicle B is XbcMeter, setting the initial speed of the obstacle vehicle B to Vbkm/h, initial speed of subject vehicle A Vckm/h, initial velocity of colliding vehicle C is Vakm/h, with a scene function of S (X)ab,Xbc,Va,Vb,Vc) Indicating a passing scene, V, set from the above initial statea、Vb、VcThe speed of is selected depending onSelecting x drivers with the driving ages of 3-5 years according to the traffic regulation speed limit value of the corresponding road, performing joint simulation in the built scene by using three driving simulators simultaneously, and designing the x drivers by using a permutation and combination principle to maximally utilize human resourcesGroup tests, each group test was repeated y times, totalingThe test is performed by recording the running states of the host vehicle A, the obstacle vehicle B and the collision vehicle C in each test, and using Nb1、Nb2、Nb3、Nc1、Nc2、Nc3Respectively represents three driving behaviors of the vehicle B, C, namely acceleration driving, constant speed driving and deceleration driving; ma1、Ma2、Ma3、Ma4、Ma5、Ma6Respectively representing six driving behaviors of accelerating non-lane change, accelerating lane change, constant-speed non-lane change, constant-speed lane change, decelerating non-lane change and decelerating lane change of the main vehicle A;
let the scene function be S (X)ab,Xbc,Va,Vb,Vc) The decision probability P of the acceleration of the vehicle Bb,1Uniform speed running decision probability Pb,2And a deceleration driving decision probability Pb,3Acceleration decision probability P of conflicting vehicle Cc,1Uniform speed running decision probability Pc,2And a deceleration driving decision probability Pc,3Setting the driving behaviors of the obstacle vehicle B and the conflict vehicle C as N respectivelybi、NcjUnder the scene, the driving behavior of the main vehicle A is MakHas a decision probability of Pk|ijThe calculation formula is as follows:
changing the scene function S (X)ab,Xbc,Va,Vb,Vc) Initial values of variables, S (X) being obtained using the above calculation methodab,Xbc,Va,Vb,Vc) And Pb,i、Pc,j、Pk|ijThe mapping table of (1), namely the overtaking scene prior knowledge;
2) "overtaking" decision probability extraction
Reading in a specific overtaking driving case aiming at a 'overtaking' driving task in a bidirectional two-lane conflict scene, respectively calculating an accelerated driving decision probability, a constant speed driving decision probability and a decelerated driving decision probability of a front obstacle vehicle B and a lane change conflict vehicle C at the current moment through priori knowledge, establishing a mixed decision joint probability of the overtaking scene of a main vehicle A at the current moment based on a statistical analysis theory, extracting a conditional probability of each decision of the main vehicle A by utilizing the priori knowledge, and calculating an accelerated non-lane-changing decision probability, an accelerated lane-changing decision probability, a constant speed non-lane-changing decision probability, a constant speed lane-changing decision probability, a decelerated non-lane-changing decision probability and a decelerated lane-changing decision probability of the main vehicle A under each decision of the overtaking scene at the current moment;
3) quantitative evaluation of 'overtaking' driving task complexity
The method comprises the steps of counting the occurrence probability of each behavior decision of a main vehicle A in a overtaking scene, calculating the average decision probability of overtaking, selecting a reasonable task complexity normalization parameter, establishing a task complexity function of the main vehicle A in the overtaking scene at the current time by utilizing the principle that the larger the covariance of the actual behavior decision probability and the average decision probability of the main vehicle A is, the higher the complexity of the overtaking driving task at the current time is, popularizing the task complexity quantitative evaluation method to the complex overtaking scene, and selecting the maximum value of the task complexity function for quantitatively evaluating the overall complexity of the overtaking driving task.
2. The complex overtaking scene oriented running task complexity quantitative evaluation method as recited in claim 1, characterized in that: the realization formula of the driving task complexity quantitative evaluation method facing the complex overtaking scene is as follows:
reading tpScene of overtaking at any moment
(1) Determining t from a priori knowledgepAcceleration driving decision probability P of preceding obstacle vehicle B at timeb,1(tp) Uniform speed running decision probability Pb,2(tp) And a deceleration driving decision probability Pb,3(tp);
(2) Determining t from a priori knowledgepAcceleration driving decision probability P of vehicle C with conflict of lane change at any momentc,1(tp) Uniform speed running decision probability Pc,2(tp) And a deceleration driving decision probability Pc,3(tp);
(3) Calculating t according to the accelerated driving decision probability, the uniform speed driving decision probability and the decelerated driving decision probability of the front obstacle vehicle B and the lane change conflict vehicle CpDecision joint probability P of time overtaking scenebc,ij(tp) Where i is {1,2,3}, and j is {1,2,3}, the following formula:
Pbc,ij(tp)=Pc,i(tp)×Pc,j(tp) (4);
(4) known calculation of tpConditional probability of the subject vehicle A under each decision joint probability in the time overtaking scene, namely the acceleration lane-unchanging conditional probability P of the subject vehiclet,1|ij(tp) Accelerated lane change probability Pt,2|ij(tp) Constant velocity non-changing track probability Pt,3|ij(tp) Uniform speed lane changing probability Pt,4|ij(tp) Deceleration non-lane-changing probability Pt,5|ij(tp) And deceleration lane change probability Pt,6|ij(tp);
(5) According to tpJoint probability P of each decision in moment overtaking scenebc,ij(tp),i={1,2,3},j={1,2,3} and conditional probability P of subject vehicle At,k|ij(tp) And k is {1,2,3,4,5,6}, and the probability P of occurrence of each decision-making behavior of the subject vehicle A is calculatedk(tp) The formula is as follows:
(6) knowing tpThe average decision probability of the time overtaking scene is Pa(tp) 1/n, n-6 represents the actual decision number of the main vehicle A, and a task normalization parameter W is selectedc>0, quantitative calculation of tpThe overtaking task complexity of the main vehicle A in the overtaking scene at any moment is shown in the following formula;
(7) reading the overtaking scene at the next moment, wherein p + tau, tau is 1, …, N, repeating the steps (1) - (6) to calculate the overtaking task complexity (formula (3)) at each moment, and selecting the maximum value as a quantitative evaluation index of the overtaking task complexity in the overtaking process, wherein the formula is as follows:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910328180.9A CN110182217B (en) | 2019-04-23 | 2019-04-23 | Running task complexity quantitative evaluation method oriented to complex overtaking scene |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910328180.9A CN110182217B (en) | 2019-04-23 | 2019-04-23 | Running task complexity quantitative evaluation method oriented to complex overtaking scene |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110182217A CN110182217A (en) | 2019-08-30 |
CN110182217B true CN110182217B (en) | 2020-11-10 |
Family
ID=67714950
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910328180.9A Expired - Fee Related CN110182217B (en) | 2019-04-23 | 2019-04-23 | Running task complexity quantitative evaluation method oriented to complex overtaking scene |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110182217B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110764494A (en) * | 2019-11-26 | 2020-02-07 | 畅加风行(苏州)智能科技有限公司 | Human-computer hybrid driving automatic driving virtual test platform based on SCANeR |
CN111797000A (en) * | 2020-05-27 | 2020-10-20 | 中汽数据有限公司 | Scene complexity evaluation method based on gradient lifting decision tree model |
CN111651712B (en) * | 2020-05-27 | 2023-03-28 | 北京航空航天大学 | Method and system for evaluating complexity of test scene of intelligent automobile |
CN111784142B (en) * | 2020-06-24 | 2022-08-26 | 吉林大学 | Method for establishing task complexity quantitative model of advanced driving assistance system |
CN111813083B (en) * | 2020-07-09 | 2022-11-01 | 吉林大学 | Scene task complexity quantitative evaluation method |
CN111882870B (en) * | 2020-07-14 | 2022-07-01 | 吉林大学 | Quantification method of road traffic environment |
CN112373482B (en) * | 2020-11-23 | 2021-11-05 | 浙江天行健智能科技有限公司 | Driving habit modeling method based on driving simulator |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2405383A1 (en) * | 2010-07-06 | 2012-01-11 | BAE Systems PLC | Assisting with guiding a vehicle over terrain |
CN103077280A (en) * | 2013-01-12 | 2013-05-01 | 天津市市政工程设计研究院 | Traffic design method for branch and confluent areas of underground road for safety of large vehicles |
CN103359013A (en) * | 2012-04-01 | 2013-10-23 | 同济大学 | Lane changing and overtaking assisting method and system based on inter-vehicle communication |
DE102013009995A1 (en) * | 2013-06-14 | 2013-12-19 | Daimler Ag | Method for distance regulation of vehicle, involves reducing spacing between vehicle and preceding vehicle at desired distance within time duration and delaying vehicle based on delay strategies |
CN104269073A (en) * | 2014-09-11 | 2015-01-07 | 同济大学 | Reverse lane overtaking early warning method based on vehicle-to-vehicle communication |
CN104951764A (en) * | 2015-06-17 | 2015-09-30 | 浙江工业大学 | Identification method for behaviors of high-speed vehicle based on secondary spectrum clustering and HMM (Hidden Markov Model)-RF (Random Forest) hybrid model |
CN109353338A (en) * | 2018-08-10 | 2019-02-19 | 重庆大数据研究院有限公司 | A kind of intelligent vehicle fast collision probability safety predicting method |
-
2019
- 2019-04-23 CN CN201910328180.9A patent/CN110182217B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2405383A1 (en) * | 2010-07-06 | 2012-01-11 | BAE Systems PLC | Assisting with guiding a vehicle over terrain |
CN103359013A (en) * | 2012-04-01 | 2013-10-23 | 同济大学 | Lane changing and overtaking assisting method and system based on inter-vehicle communication |
CN103077280A (en) * | 2013-01-12 | 2013-05-01 | 天津市市政工程设计研究院 | Traffic design method for branch and confluent areas of underground road for safety of large vehicles |
DE102013009995A1 (en) * | 2013-06-14 | 2013-12-19 | Daimler Ag | Method for distance regulation of vehicle, involves reducing spacing between vehicle and preceding vehicle at desired distance within time duration and delaying vehicle based on delay strategies |
CN104269073A (en) * | 2014-09-11 | 2015-01-07 | 同济大学 | Reverse lane overtaking early warning method based on vehicle-to-vehicle communication |
CN104951764A (en) * | 2015-06-17 | 2015-09-30 | 浙江工业大学 | Identification method for behaviors of high-speed vehicle based on secondary spectrum clustering and HMM (Hidden Markov Model)-RF (Random Forest) hybrid model |
CN109353338A (en) * | 2018-08-10 | 2019-02-19 | 重庆大数据研究院有限公司 | A kind of intelligent vehicle fast collision probability safety predicting method |
Also Published As
Publication number | Publication date |
---|---|
CN110182217A (en) | 2019-08-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110182217B (en) | Running task complexity quantitative evaluation method oriented to complex overtaking scene | |
CN105868787A (en) | Electric car driving range evaluation method based on working condition identification and fuzzy energy consumption | |
CN108198425A (en) | A kind of construction method of Electric Vehicles Driving Cycle | |
CN110321954A (en) | The driving style classification and recognition methods of suitable domestic people and system | |
CN103971097B (en) | Vehicle license plate recognition method and system based on multiscale stroke models | |
CN103337161A (en) | Optimization method of intersection dynamic comprehensive evaluation and signal control system based on real-time simulation model | |
CN103679214B (en) | Vehicle checking method based on online Class area estimation and multiple features Decision fusion | |
CN105279966A (en) | Jam recognition method for campus traffic | |
CN109190444A (en) | A kind of implementation method of the lane in which the drivers should pay fees vehicle feature recognition system based on video | |
Song et al. | Comparative analysis of car-following models for emissions estimation | |
CN106326581A (en) | Determination method and device for driving range and automobile | |
CN107705635B (en) | Method for judging traffic conflict of electric bicycles at signalized intersection | |
CN113419942B (en) | Automatic driving safety evaluation method based on natural driving data | |
CN111243338A (en) | Vehicle acceleration-based collision risk evaluation method | |
Alshayeb et al. | Investigating impacts of various operational conditions on fuel consumption and stop penalty at signalized intersections | |
CN106960189A (en) | A kind of driving intention decision method based on hidden Markov model | |
CN106203296A (en) | The video actions recognition methods of one attribute auxiliary | |
CN110459053A (en) | Highway Speed of Mixed Traffic Flow estimation method based on intelligent network connection vehicle information | |
Mao et al. | Driving conditions of new energy logistics vehicles using big data technology | |
CN108830414B (en) | Load prediction method for commercial charging area of electric automobile | |
CN115100847B (en) | Queuing service time estimation method for low-permeability network-connected track data | |
CN106056923A (en) | Road traffic state discrimination method based on traffic scene radar | |
CN103164531A (en) | Two-stage instance layer data integration approach based on fuzzy priority | |
Li et al. | Driving type categorizing and microscopic simulation model calibration | |
CN113807669A (en) | Scheme selecting method for modifying urban road intersection passenger-cargo separation system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201110 |
|
CF01 | Termination of patent right due to non-payment of annual fee |