CN111985092B - Intelligent automobile simulation test matrix generation method - Google Patents
Intelligent automobile simulation test matrix generation method Download PDFInfo
- Publication number
- CN111985092B CN111985092B CN202010750334.6A CN202010750334A CN111985092B CN 111985092 B CN111985092 B CN 111985092B CN 202010750334 A CN202010750334 A CN 202010750334A CN 111985092 B CN111985092 B CN 111985092B
- Authority
- CN
- China
- Prior art keywords
- simulation test
- lstm
- matrix
- test matrix
- cstm
- 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.)
- Active
Links
- 239000011159 matrix material Substances 0.000 title claims abstract description 86
- 238000004088 simulation Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012360 testing method Methods 0.000 title claims abstract description 51
- 238000005070 sampling Methods 0.000 claims abstract description 19
- 238000005457 optimization Methods 0.000 claims abstract description 16
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 7
- 230000003068 static effect Effects 0.000 claims description 7
- 230000006399 behavior Effects 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 abstract description 27
- 230000001133 acceleration Effects 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 230000002068 genetic effect Effects 0.000 description 4
- 239000002245 particle Substances 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000002922 simulated annealing Methods 0.000 description 3
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000011164 primary particle Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Mathematics (AREA)
- Molecular Biology (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method for generating an intelligent automobile simulation test matrix, and belongs to the technical field of intelligent automobile simulation experiments. The generating method comprises the following steps: step one, extracting a simulation test matrix CSTM based on natural driving data; and step two, generating a new simulation test matrix LSTM cor based on the simulation test matrix CSTM by using COLHS. The intelligent automobile simulation test matrix is generated by adopting an optimized Latin hypercube algorithm based on correlation control. The method can ensure that the simulation test matrix has good space coverage under any case number, and can also reserve the correlation among variables in the CSTM. COLHS used in the present invention is an algorithm that combines Cholesky decomposition with combinatorial optimization. The method can complete sampling at non-positive timing of the correlation coefficient matrix and can rapidly and accurately control the correlation of the simulation test matrix at positive timing of the correlation coefficient matrix.
Description
Technical Field
The invention relates to a method for generating an intelligent automobile simulation test matrix, and belongs to the technical field of intelligent automobile simulation experiments.
Background
In order to link intelligent automobile simulation experiments with real traffic scenes, many students have studied simulation test matrix methods based on natural driving data. The invention refers to a method for extracting a simulation test matrix by using a Markov random process (the simulation test matrix extracted directly by natural driving data is marked as CSTM, each row of the CSTM corresponds to one test case, each column corresponds to one variable), and the method is widely researched because the method accords with the randomness of traffic participants and has high reusability. However, in the early stage of development of intelligent driving automobile control algorithm, direct use of CSTM is time-consuming and unnecessary. In addition, in some extreme cases the extraction of natural driving data is difficult (such as extreme weather, extreme environment, extreme dangerous situations, etc.), and cases in CSTM are not sufficient. To solve these two problems, a method for generating a simulation test matrix by extracting the simulation test matrix by means of a Markov random process and utilizing optimized Latin hypercube sampling is disclosed. The invention can generate the simulation test matrix with good space coverage and very similar correlation coefficient matrix with CSTM according to the user requirement. The field of direct application of the method belongs to the field of intelligent automobile simulation test.
Disclosure of Invention
The invention aims to provide a method for generating an intelligent automobile simulation test matrix, which aims to solve the problems in the existing simulation test.
A method for generating an intelligent automobile simulation test matrix, the method comprising:
Step one, extracting a simulation test matrix CSTM based on natural driving data;
And step two, generating a new simulation test matrix LSTM cor based on the simulation test matrix CSTM by using COLHS.
Further, in the first step, the method specifically includes the following steps:
Step one, extracting a specified scene event from natural driving data;
step one, extracting a variable of interest in simulation from a specified scene event;
Step one, directly storing the static variable into the CSTM, regarding the dynamic variable as a Markov random process, fitting the dynamic variable into a Markov random model, and then storing model parameters into the CSTM.
Further, the scenario events include a following event, a lane change event, a pre-crash event, a rider event, and a pedestrian event.
Further, in step one, the conditions specifying the scene event include weather conditions, lighting conditions, traffic participant behavior, driving environment conditions, and vehicle driving data.
Further, in step one three, the static variables include a vehicle status, a road status, an environmental status, and other traffic participant status in the scene event.
Further, in step one and three, the dynamic variables include a vehicle dynamic parameter, a dynamic parameter of the vehicle relative to other traffic participants, and a self dynamic parameter of the other traffic participants.
Further, in the second step, the method specifically includes the following steps:
step two, LHS sampling is carried out on the CSTM to obtain a pull Ding Chao cube simulation test matrix before correlation control, and the matrix is recorded as LSTM;
step two, extracting a correlation coefficient matrix of the CSTM, and marking the correlation coefficient matrix as Maarix cor;
Step two, judging whether Matrix cor is positive or not and whether n s is larger than m or not, if positive, and if n s is larger than m, executing step two, and if n is positive, executing step four; otherwise, executing the second step;
step two, executing a cholesky decomposition method on the LSTM to obtain LSTM 'subjected to correlation control, judging whether the correlation error of the LSTM' meets the requirement, and executing step two five if the correlation error does not meet the requirement; if yes, the LSTM cor=LSTM′,LSTMcor is the required simulation test matrix;
And step two, performing a combination optimization method on the LSTM (or LSTM') to obtain the final LSTM cor.
The invention has the main advantages that: the invention provides a method for generating an intelligent automobile simulation test matrix, which adopts an optimized Latin hypercube algorithm (COLHS) based on correlation control to generate the intelligent automobile simulation test matrix. The method can ensure that the simulation test matrix has good space coverage under any case number, and can also reserve the correlation among variables in the CSTM. COLHS used in the present invention is an algorithm that combines Cholesky decomposition with combinatorial optimization. The method can complete sampling at non-positive timing of the correlation coefficient matrix and can rapidly and accurately control the correlation of the simulation test matrix at positive timing of the correlation coefficient matrix.
Drawings
FIG. 1 is a schematic view of LHS sampling;
FIG. 2 is a flow chart of a method for generating an intelligent automobile simulation test matrix according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Latin hypercube sampling technology (Latin hypercube sampling, hereinafter abbreviated as LHS) is widely applied in the field of simulation experiment design because the Latin hypercube sampling technology can uniformly cover the distribution range of experimental variables through layered sampling. When experimental variables related to simulation experiments are independent of each other, in order to make sample points uniformly fill the whole sample space, an optimized latin square sampling technology (Optimized Latin hypercube sampling, hereinafter abbreviated as OLHS) based on a space filling criterion is mainly adopted at present. And when the experimental variables have correlation, correlation control is needed for the sampling result of the LHS.
The invention provides an intelligent automobile simulation test matrix generated by adopting an optimized Latin hypercube algorithm (COLHS) based on correlation control. The method can ensure that the simulation test matrix has good space coverage under any case number, and can also reserve the correlation among variables in the CSTM. COLHS used in the present invention is an algorithm that combines Cholesky decomposition with combinatorial optimization. The method can complete sampling at non-positive timing of the correlation coefficient matrix and can rapidly and accurately control the correlation of the simulation test matrix at positive timing of the correlation coefficient matrix.
A method for generating an intelligent automobile simulation test matrix, the method comprising:
Step one, extracting a simulation test matrix CSTM based on natural driving data;
And step two, generating a new simulation test matrix LSTM cor based on the simulation test matrix CSTM by using COLHS.
Further, in the first step, the method specifically includes the following steps:
Step one, extracting a specified scene event from natural driving data;
step two, extracting variables (which can be divided into dynamic variables and static variables) of interest in simulation from specified scene events;
Step one, directly storing the static variable into the CSTM, regarding the dynamic variable as a Markov random process, fitting the dynamic variable into a Markov random model, and then storing model parameters into the CSTM.
Further, in step one, the specified scene event includes weather conditions (such as sunny days, rainy days, cloudy days, etc.), lighting conditions (such as illumination intensity, etc.), traffic participant behaviors (such as forward vehicles going straight, lane changing, turning, etc.), driving environment conditions (such as rural areas, cities, suburban areas, etc.), and vehicle driving data (such as lateral/longitudinal/vertical speeds, lateral/longitudinal/vertical accelerations, etc.).
Further, in step one, the static variables include a state of the vehicle at the beginning of the event (such as a driver's operation behavior, a self-initial speed, a distance of the vehicle from other traffic participants, position information, etc.), a road state (such as the number of lanes, a road surface level, whether guardrails are present), an environmental state (such as a weather condition, an illumination condition, etc.), and other traffic participant states.
Further, in step one three, the dynamic variables include a vehicle dynamic parameter (such as a vehicle speed, a vehicle acceleration, etc. in the entire scene event), a vehicle dynamic parameter with respect to other traffic participants (such as a vehicle lateral/longitudinal distance with respect to a preceding vehicle in the entire scene event, etc.), and a vehicle dynamic parameter of other traffic participants (such as a preceding vehicle speed, acceleration, etc. in the entire scene event).
Further, in the second step, the method specifically includes the following steps:
step two, LHS sampling is carried out on the CSTM to obtain a pull Ding Chao cube simulation test matrix before correlation control, and the matrix is recorded as LSTM;
step two, extracting a correlation coefficient Matrix of CSTM, and marking the correlation coefficient Matrix as Matrix cor;
judging whether Maarix cor is positive or not and whether n s is larger than m or not, if yes, executing the second step if n s is larger than m; otherwise, executing the second step;
Step two, performing a cholesky decomposition method on Matrix cor to obtain LSTM 'subjected to correlation control, judging whether the correlation error of the LSTM' meets the requirement, and if not, executing step two fifth; if yes, the LSTM cor=LSTM′,LSTMcor is the required simulation test matrix;
and step two, performing a combination optimization method on Maarix cor to obtain the final LSTM cor.
Specifically, S1, carrying out LHS sampling on the CSTM by using Algorithm 1. Because the cumulative probability distribution function of each variable in the CSTM is unknown, a cumulative empirical distribution function is used instead. A schematic of the sampling is shown in fig. 1.
Algorithm 1
S2, extracting a correlation coefficient Matrix (marked as Matrix cor) of the CSTM. The correlation coefficient matrix includes, but is not limited to, spearman, kendall, pearson correlation coefficients.
S3, judging whether Matrix cor is positive or not and whether n s is larger than m or not. If positive, and n s > m, then S4 follows, otherwise S5 follows.
S4, executing a cholesky decomposition method, wherein the specific steps are shown as Algorithm 2
Algorithm 2
Judging whether the correlation error is required or not, and if the correlation error meets LSTM cor =LSTM', obtaining the required simulation test matrix. If not, S5 follows.
S5, executing a combination optimization method to obtain the final LSTM cor. It should be noted that if the algorithm is executed S4, LSTM described in the following step is replaced by LSTM'.
The combined optimization method is a common method for processing discrete problems by using an intelligent optimization algorithm, and is more common in a simulated annealing algorithm, a particle swarm algorithm, a genetic algorithm and the like. When applied to LHS dependency control, they are mainly characterized in that the dependency relationship between columns is changed by "swapping" the positions of several elements of each column in the LSTM matrix. The "change" is then made to "approach" the correlation coefficient matrix of the LSTM gradually towards the correlation coefficient matrix of the CSTM, according to a defined objective function.
Specifically, the 'exchange', 'change', 'approximation' is realized by a defined 'disturbance' operator, an acceptance criterion and the like in a simulated annealing algorithm; the genetic algorithm is realized by crossing, mutation and selecting operators; the particle swarm algorithm is realized by defined particle 'movement' and 'disturbance' operators.
Specifically, if Matrix cor is a positive definite Matrix, the initial solution of the combinatorial optimization algorithm is generated by S4 (the initial solution in the simulated annealing algorithm, the primary particles in the particle swarm algorithm, the initial population in the genetic algorithm). If Matrix cor is a non-positive definite Matrix, then the initial solution for the combined optimization Algorithm is randomly generated (call 3-6 steps in Algorithm 2, randomly ordering each column in LSTM).
One specific embodiment is set forth below:
the effectiveness of the invention is illustrated by taking the extraction of the following scene simulation test cases based on natural driving data as an example.
1. Extracting a following event
First, we extracted 2917 car following events from 30 km natural driving data provided by the national institute of automotive engineering. The extraction criteria were as follows:
(1)vr≥0m/s
(2)vl≥0m/s
(3)Rl(t)∈(0.1m,90m)
(4) No other vehicles cut in between the front vehicle and the rear vehicle
(5) The front vehicle and the rear vehicle do not have lane change
(6) The length of the following car is more than or equal to 20s
V l and v r represent the front vehicle speed and the rear vehicle speed, respectively, and R l is the relative distance between the front vehicle and the rear vehicle.
2. Extraction of CSTM
And then establishing a front vehicle random model under the following scene according to the following formula.
rh~N(μr,σr 2)
Wherein h= [ H 0,h1,h2]T ] is a model parameter vector, r h is a random number following a normal distribution, the average value of the normal distribution is μ r, the variance is σ r,al is the acceleration of the front vehicle, and t=0.04 s is the sampling period. a lerror is a l and when r h =0Errors between them. H can be calculated from a l using a robust regression model (robust regression model. Mu. r and σ r can be calculated from a lerror using a gaussian mixture model (Gaussian mixture model).
CSTM in the following scene is defined as follows:
Wherein v rstart,vLstart,aLstart,Rlstart represents the initial speed of the rear vehicle, the initial speed of the front vehicle, the initial acceleration of the front vehicle, and the relative distance between the front vehicle and the rear vehicle at the initial time, respectively. T represents the number of iterations of equation 1, which is also the sampling frequency for each following event. n represents the number of following events.
3. Acquisition of LSTM cor
To further illustrate the advantages of the present invention, we have generated LSTM's at different n s with the present invention, respectively. Wherein, n s has the value shown in table 1. The combination optimization method used in this example is a genetic algorithm-based combination optimization method (GA), the parameter settings are shown in table 2, and the correlation coefficient matrices used are spearman correlation coefficients. In this example, the correlation coefficient matrix corresponding to CSTM is a positive definite matrix.
TABLE 1 Small sample n s values
Variable name | Value taking |
pnum | 10 |
mutation | 0.1 |
itmax | 1000 |
Table 2. Objective function of ga parameter combination optimization method is shown in formula 2:
Where F represents the fitness value of the objective function, the optimization objective is to minimize F. i represents the ith row of the matrix and j represents the jth column of the matrix. W is a weight matrix that can be adjusted when the user of the present invention places some emphasis on the correlation between variables during the optimization process for some reason. A is the correlation coefficient matrix of LSTM (or LSTM'), and T is the correlation coefficient matrix of CSTM.
Each experiment number is run for 10 times, a group of LSTM is obtained by one run, the fitness value average value of the objective function is obtained, and the average calculation time is shown in table 3.
TABLE 3 fitness values and calculated time-to-time averages
Advantageous effects of lstm cor.
(1) As shown in table 3, the correlation coefficient matrix of LSTM cor is very close to CSTM. LSTM cor thus preserves to a maximum extent the degree of correlation between variables in the CSTM. This is of positive interest for simulation testing.
(2) This allows LSTM cor to cover the distribution space of each variable well due to the uniformity of the latin hypercube samples themselves.
(3) Since the number of cases in LSTM cor is flexible. Therefore, in the early stage of intelligent driving automobile control algorithm development, LSTM cor with a small sample can be used for replacing CSTM, so that the coverage of a variable interval is ensured, and the simulation efficiency is improved. In addition, extraction of natural driving data is difficult in some extreme cases (such as extreme weather, extreme environments, extreme dangerous situations, etc.), when LSTM cor can be used to augment cases in CSTM.
Claims (5)
1. The intelligent automobile simulation test matrix generation method is characterized by comprising the following steps of:
Step one, extracting a simulation test matrix CSTM based on natural driving data;
Step two, generating a new simulation test matrix LSTM cor based on the simulation test matrix CSTM by COLHS;
In the second step, the method specifically comprises the following steps:
step two, LHS sampling is carried out on the CSTM to obtain a pull Ding Chao cube simulation test matrix before correlation control, and the matrix is recorded as LSTM;
step two, extracting a correlation coefficient Matrix of CSTM, and marking the correlation coefficient Matrix as Matrix cor;
Step two, judging whether Matrix cor is positive or not and whether n s is larger than m or not, if positive, and if n s>m,ns is the number of samples required by LSTM designated by a user, executing step two and four; otherwise, executing the second step;
Step two, executing a cholesky decomposition method on the LSTM to obtain LSTM 'subjected to correlation control, judging whether the correlation error of the LSTM' meets the requirement, and executing step two five if the correlation error does not meet the requirement; if yes, the LSTM cor=LSTM',LSTMcor is the required simulation test matrix;
Step two, a combination optimization method is carried out on the LSTM or the LSTM' to obtain a final LSTM cor;
in the first step, the method specifically comprises the following steps:
Step one, extracting a specified scene event from natural driving data;
step one, extracting a variable of interest in simulation from a specified scene event;
Step one, directly storing the static variable into the CSTM, regarding the dynamic variable as a Markov random process, fitting the dynamic variable into a Markov random model, and then storing model parameters into the CSTM.
2. The intelligent automobile simulation test matrix generation method of claim 1, wherein the specified scene events include a following event, a lane change event, a pre-crash event, a cyclist event and a pedestrian event.
3. The method for generating a simulation test matrix for an intelligent automobile according to claim 1, wherein in step one, the conditions for specifying the scene event include weather conditions, lighting conditions, traffic participant behaviors, driving environment conditions, and driving data of the vehicle.
4. The method of claim 1, wherein in step one three, the static variables include a vehicle motion state, a road state, an environmental state, and other traffic participant motion states in a scene event.
5. The method of claim 1, wherein in step one three, the dynamic variables include a vehicle dynamic parameter, a dynamic parameter of the vehicle relative to other traffic participants, and a self dynamic parameter of the other traffic participants.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010750334.6A CN111985092B (en) | 2020-07-30 | 2020-07-30 | Intelligent automobile simulation test matrix generation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010750334.6A CN111985092B (en) | 2020-07-30 | 2020-07-30 | Intelligent automobile simulation test matrix generation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111985092A CN111985092A (en) | 2020-11-24 |
CN111985092B true CN111985092B (en) | 2024-05-31 |
Family
ID=73444767
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010750334.6A Active CN111985092B (en) | 2020-07-30 | 2020-07-30 | Intelligent automobile simulation test matrix generation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111985092B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113157578B (en) * | 2021-01-11 | 2022-05-13 | 北京赛目科技有限公司 | Automatic driving simulation test method and device based on scene |
CN113066282B (en) * | 2021-02-26 | 2022-05-27 | 北京航空航天大学合肥创新研究院(北京航空航天大学合肥研究生院) | Method and system for modeling vehicle following coupling relation in mixed-traveling environment |
CN113823096B (en) * | 2021-11-25 | 2022-02-08 | 禾多科技(北京)有限公司 | Random traffic flow obstacle object arrangement method for simulation test |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104298540A (en) * | 2014-10-23 | 2015-01-21 | 浙江大学 | Underlaying model parameter correction method of microscopic traffic simulation software |
CN105608267A (en) * | 2015-12-21 | 2016-05-25 | 许昌学院 | Multivariable global optimization algorithm |
CN105740592A (en) * | 2016-05-05 | 2016-07-06 | 中国人民解放军国防科学技术大学 | Latin hypercube experiment design method based on sequential sampling |
CN107038306A (en) * | 2017-04-14 | 2017-08-11 | 哈尔滨工业大学 | A kind of optimal Latin hypercube experimental design method based on self-adapted genetic algorithm |
CN107436971A (en) * | 2017-07-07 | 2017-12-05 | 东南大学 | Suitable for the improvement Latin Hypercube Sampling method of non-positive definite form correlation control |
CN110263381A (en) * | 2019-05-27 | 2019-09-20 | 南京航空航天大学 | A kind of automatic driving vehicle test emulation scene generating method |
-
2020
- 2020-07-30 CN CN202010750334.6A patent/CN111985092B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104298540A (en) * | 2014-10-23 | 2015-01-21 | 浙江大学 | Underlaying model parameter correction method of microscopic traffic simulation software |
CN105608267A (en) * | 2015-12-21 | 2016-05-25 | 许昌学院 | Multivariable global optimization algorithm |
CN105740592A (en) * | 2016-05-05 | 2016-07-06 | 中国人民解放军国防科学技术大学 | Latin hypercube experiment design method based on sequential sampling |
CN107038306A (en) * | 2017-04-14 | 2017-08-11 | 哈尔滨工业大学 | A kind of optimal Latin hypercube experimental design method based on self-adapted genetic algorithm |
CN107436971A (en) * | 2017-07-07 | 2017-12-05 | 东南大学 | Suitable for the improvement Latin Hypercube Sampling method of non-positive definite form correlation control |
CN110263381A (en) * | 2019-05-27 | 2019-09-20 | 南京航空航天大学 | A kind of automatic driving vehicle test emulation scene generating method |
Also Published As
Publication number | Publication date |
---|---|
CN111985092A (en) | 2020-11-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111985092B (en) | Intelligent automobile simulation test matrix generation method | |
US20210197851A1 (en) | Method for building virtual scenario library for autonomous vehicle | |
DE102018121595B4 (en) | UNSUPERVISED TRAINING OF AGENTS FOR AUTONOMOUS DRIVING APPLICATIONS | |
Jia et al. | Long short‐term memory and convolutional neural network for abnormal driving behaviour recognition | |
CN112036001A (en) | Automatic driving test scene construction method, device and equipment and readable storage medium | |
CN112215337A (en) | Vehicle trajectory prediction method based on environment attention neural network model | |
CN111079800B (en) | Acceleration method and acceleration system for intelligent driving virtual test | |
Huang et al. | A versatile approach to evaluating and testing automated vehicles based on kernel methods | |
CN113010967A (en) | Intelligent automobile in-loop simulation test method based on mixed traffic flow model | |
CN113505440B (en) | Real-time prediction method for aerodynamic performance parameters of automobile based on three-dimensional deep learning | |
CN113094808A (en) | Simulation data and artificial intelligence based automobile collision damage grade real-time prediction method | |
CN112163285B (en) | Modeling method of road surface type prediction model for simulating driving system | |
CN115238958A (en) | Dangerous event chain extraction method and system based on complex traffic scene | |
Cheng et al. | Behavioral classification of drivers for driving efficiency related ADAS using artificial neural network | |
CN110763483A (en) | Automatic generation method and device of security level test scene library | |
CN113222239A (en) | Short-time traffic flow prediction method based on CNN-LSTM-At neural network | |
DE102020128978A1 (en) | TRAINING DEEP NEURAL NETWORKS WITH SYNTHETIC IMAGES | |
Watanabe et al. | Scenario mining for development of predictive safety functions | |
Du et al. | Finding diverse failure scenarios in autonomous systems using adaptive stress testing | |
Yang et al. | Survey on autonomous vehicle simulation platforms | |
CN110949380A (en) | Location prediction for dynamic objects | |
CN107506568A (en) | A kind of traffic accident data analyze the method with prediction in real time | |
CN110213741B (en) | Method for detecting authenticity of vehicle sending information in real time based on width learning | |
CN113706911B (en) | Automatic driving method based on digital traffic scene | |
CN115376129A (en) | Mining area pavement semantic segmentation method, device, chip, terminal and computer equipment |
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 |