CN111841012A - Automatic driving simulation system and test resource library construction method thereof - Google Patents
Automatic driving simulation system and test resource library construction method thereof Download PDFInfo
- Publication number
- CN111841012A CN111841012A CN202010581488.7A CN202010581488A CN111841012A CN 111841012 A CN111841012 A CN 111841012A CN 202010581488 A CN202010581488 A CN 202010581488A CN 111841012 A CN111841012 A CN 111841012A
- Authority
- CN
- China
- Prior art keywords
- index
- test
- scene
- traffic
- evaluation
- 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
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 141
- 238000004088 simulation Methods 0.000 title claims abstract description 86
- 238000010276 construction Methods 0.000 title claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims abstract description 54
- 230000003068 static effect Effects 0.000 claims abstract description 31
- 238000000034 method Methods 0.000 claims description 40
- 238000004422 calculation algorithm Methods 0.000 claims description 28
- 238000012216 screening Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000013210 evaluation model Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000002452 interceptive effect Effects 0.000 claims description 5
- 230000008447 perception Effects 0.000 claims description 5
- 206010039203 Road traffic accident Diseases 0.000 claims description 4
- 230000010355 oscillation Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 239000002131 composite material Substances 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 11
- 230000006399 behavior Effects 0.000 description 10
- 230000003993 interaction Effects 0.000 description 10
- 238000011160 research Methods 0.000 description 9
- 238000011161 development Methods 0.000 description 8
- 238000013461 design Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000005094 computer simulation Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/50—Controlling the output signals based on the game progress
- A63F13/52—Controlling the output signals based on the game progress involving aspects of the displayed game scene
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/60—Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/80—Special adaptations for executing a specific game genre or game mode
- A63F13/803—Driving vehicles or craft, e.g. cars, airplanes, ships, robots or tanks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
An automatic driving simulation system and a test resource library construction method thereof. The automatic driving simulation system provided by the invention constructs a test environment based on a game engine such as an illusion 4 engine, creatively provides a dynamic and static combined simulation test resource library construction method, realizes high real dynamic scene simulation, and finally constructs an automatic driving system performance comprehensive evaluation system by combining the experience of automatic driving evaluation at home and abroad and the characteristics of automatic driving test at L1-L3 level. The invention provides a dynamic and static combined simulation test resource library construction method.
Description
Technical Field
The invention relates to the field of traffic, in particular to an analog simulation system for automatically driving a vehicle.
Background
As a systematic project, the automatic driving technology has various related modules and complex business logic, and needs frequent real-vehicle on-road testing for ensuring the overall safety. However, this is just a big challenge in the development of the current automatic driving system — the real vehicle is able to measure the "road resistance and length". According to data in Chinese autopilot simulation technology research report (2019): a fleet of 1000 autonomous test vehicles is deployed that takes approximately 50 years to complete a sufficient mileage test. Enterprises are difficult to continuously invest huge resources to support wide real vehicle road tests, the wide road tests are still difficult to cover complex and various traffic scenes, and meanwhile, the frequent real vehicle road tests bring huge potential safety risks to traffic safety. In 2016, for example, a pedestrian suddenly appearing in the center of the road has been hit by a unmanned Uber vehicle. Therefore, the automatic driving virtual simulation system is emphasized by various research organizations, is used for an early verification algorithm, solves possible technical problems before real vehicle drive test, and accelerates the research and development of the automatic driving system.
The automatic driving simulation test is to establish a mathematical model of a real static traffic scene and a real dynamic traffic scene by a computer simulation technology, so that an automatic driving automobile and an algorithm are subjected to driving test in a virtual traffic scene. Compared with the real vehicle road test, the simulation test is easier, the test cost can be effectively reduced, and the algorithm development iteration period is shortened; meanwhile, a customized traffic test scene with strong pertinence can be constructed through simulation software, and algorithm specificity upgrading iteration and enhancement are realized; meanwhile, for some scenes with high recurring costs, such as an expressway in rainy and snowy days and the like, a real object test scene does not need to be built or a special test occasion is waited for, such as rainy and snowy days, the simulation greatly reduces the difficulty in realizing the test scene and enlarges the test range. At present, the high-efficiency test flow of unmanned driving is that firstly, an algorithm is upgraded on computer software through a simulation test technology to ensure that the unmanned driving can successfully run in a virtual environment, then a closed field test is carried out, and finally an open road test is carried out.
At present, the automatic driving simulation technology is being introduced into development processes by more and more unmanned vehicle research and development enterprises and scientific research institutions, for example, simulation tests are used as necessary items before public road tests are carried out by the Waymo unmanned vehicle; the simulation platform is regarded as an Apollo main manifestation way by hundredths, and the simulation cloud service is provided for Apollo partners to create income; many automated driving initiatives such as roadstar ai, pony ai, etc. are developing simulators autonomously. Chinese automated driving simulation technology research report (2019) indicates that the total size of the international market for simulation software and test in the next 5 years is about billion dollars, and that the commercial simulation software meeting the requirements of automated driving algorithm development and vehicle system integration will certainly become one of the bottom-level basic tools in the development chain. China also puts the development of the automatic driving simulation technology to a strategic height. In the beginning of 2020, the "intelligent automobile innovation development strategy" issued by 11 departments such as the department of industry and informatization and joint stamping clearly requires to research and develop technologies and verification tools such as virtual simulation, software and hardware combined simulation, real automobile road test and the like and a multi-level test evaluation system.
The premise of the high-reliability automatic driving simulation test is a reasonable and diversified test scene, and the performance potential of the automatic driving system can be fully mined and algorithm loopholes can be exposed only by the time of the simulation test. The existing simulation test scene design at home and abroad basically focuses on reproducing important natural driving scenes, and although the method can test the safety of the automatic driving system to a certain extent, other dangerous boundary situations are ignored. More importantly, the design of the current test scene cannot perfectly solve the dynamic element simulation in the scene, such as the simulation of motor vehicles, pedestrians and non-motor vehicles, which relate to the simulation of other traffic participants. The existing dynamic simulation scene simulation systems such as the simulation world Carcraft of Waymo, the open source simulation test environment CARLA and the like all have the problems that simulation traffic participants are lack of interaction with a tested vehicle in simulation tests, a generalization method of real scenes is lack of deep research and the like. In the simulation test process, vehicles in the simulation environment run according to actual measurement or simulation tracks, lack of interaction with a tested vehicle or unreasonable unreal interaction is generated. The simulation dynamic scene generalization has the problem of reality loss, for example, a dense traffic flow case is generalized, after a vehicle track is changed, a plurality of peripheral tracks are actually affected and spread, the driving disturbance of a single vehicle sometimes causes instability of the whole traffic flow, and the phenomenon can be difficult to reproduce by a simple generalization sample expansion method.
Therefore, the invention innovatively provides a dynamic and static combined simulation test resource library construction method, realizes highly real dynamic scene simulation, and firstly constructs a typical static test scene library with each level and function of test value by using a combined inference method and a scene screening rule on the basis of a complex scene group formed by the arrangement and combination of the relative positions and the motion relations of a tested vehicle and all surrounding interference vehicles. On the basis of a typical static test scene library, an online drainage system based on multiple platforms such as a webpage and an android is developed, the driving right of the interference vehicle around the tested vehicle is taken over through the intervention of an online game player, and real dynamic scene simulation is realized through the participation of real users. Further, real-person driving data under a targeted test scene is extracted, an intelligent traffic model is trained on the basis of artificial intelligent algorithms such as deep learning, and real-person players in the online drainage system are supplemented/replaced. The invention systematically designs a highly real automatic driving simulation system by taking a dynamic and static combined simulation test resource library as a core, wherein a test environment is constructed based on a game engine such as an illusion 4 engine, a platform during the test can record all fine expressions after a tested vehicle starts from a starting point, and the performance of an algorithm is evaluated from the aspects of vehicle intelligence, driving safety, driving comfort and traffic coordination. The invention well solves the problem of behavior distortion of traffic participants in the current dynamic scene simulation process, greatly improves the test reliability of a simulation system and assists the development of the automatic driving technology in China.
Disclosure of Invention
The invention aims to solve the technical problem of providing an automatic driving simulation system, which aims to solve the problem of behavior distortion of traffic participants in the dynamic scene simulation process of the conventional simulation system and improve the test reliability of the simulation system.
The technical scheme adopted by the invention for solving the technical problems is as follows: an automatic driving simulation system is designed, a test environment is built based on a game engine such as an illusion 4 engine, a dynamic and static combined simulation test resource library construction method is provided, high real dynamic scene simulation is achieved, and finally a comprehensive performance evaluation system of the automatic driving system is built by combining experience of automatic driving evaluation at home and abroad and automatic driving test characteristics of L1-L3. The invention provides a dynamic and static combined simulation test resource library construction method, which is characterized in that a typical static test scene library with each level and function of test value is constructed on the basis of a complex scene group formed by arrangement and combination of relative positions and motion relations of a tested vehicle and all surrounding interference vehicles based on a combined inference method and a scene screening rule. On the basis of a typical static test scene library, an online drainage system based on multiple platforms such as a webpage and an android is developed, the driving right of the interference vehicle around the tested vehicle is taken over through the intervention of an online game player, and real dynamic scene simulation is realized through the participation of real users. Further, real-person driving data under a targeted test scene is extracted, an intelligent traffic model is trained on the basis of artificial intelligent algorithms such as deep learning, and real-person players in the online drainage system are supplemented/replaced. After the automatic driving algorithm is accessed to the simulation platform and starts to be tested, the platform can record all fine performances of the tested vehicle after the tested vehicle starts from the starting point, and the performance of the algorithm is evaluated in the aspects of vehicle intelligence, driving safety, driving comfort and traffic coordination. The invention systematically designs a highly real automatic driving simulation system by taking a dynamic and static combined simulation test resource library construction method as a core, well solves the problem of behavior distortion of traffic participants in the current dynamic scene simulation process, and greatly improves the test reliability of the simulation system.
The basic principle of the simulation technology is to simulate and generate obstacles, pedestrians, vehicles, weather and the like in the real physical world in a virtual scene, use the obstacles, the pedestrians, the vehicles, the weather and the like as sensor sensing objects to test a sensing algorithm, simulate the moving tracks of traffic participants such as the vehicles, the pedestrians and the like and the arrangement of related scenes to test a decision planning algorithm, and simulate the dynamic characteristics of the vehicles to test a control algorithm. The automatic driving simulation technology comprises the steps of constructing a test environment, constructing a test scene library, simulating a dynamic scene and establishing an automatic driving system evaluation system.
The constructed test environment generally includes roads (lane lines, road surface materials, etc.), traffic elements (traffic lights and traffic signboards), traffic participants (motor vehicles, non-motor vehicles and pedestrians), road surrounding elements (including street lamps and buildings), and the like. The invention constructs a test environment based on the game engine, can simulate road traffic scenes and virtual city scenes with high precision, and restores the illumination conditions such as refraction and reflection and the change of weather conditions such as wind, frost, rain and snow by means of the game engine, so that the scenes and the simulated test conditions of the sensors are close to reality, and the authenticity of the test effect of the perception algorithm is further ensured. In addition, a physical engine built in the game engine can be directly used for simulating the effects of vehicle slipping, collision and the like, and support is provided for testing of a control algorithm.
The test scene is the comprehensive reflection of the environment and the driving behavior in a certain time and space range, and describes the external states of roads, traffic facilities, meteorological conditions, traffic participants and the like, and the information of the driving task and the state of the tested vehicle and the like. An organic collection of multiple scenes is a "scene library". The automatic driving test scene library is a database formed by a series of automatic driving test scenes meeting certain test requirements. The invention constructs a typical static test scene library with each level and function of test value based on a complex scene group formed by permutation and combination of relative positions and motion relations of a tested vehicle and all surrounding interference vehicles and based on a combinatorial reasoning method and a scene screening rule. The combined inference method comprises the following specific steps:
determining a complex scene group: aiming at a specified road traffic environment, analyzing the combination of possible relative positions and movement directions of a tested vehicle and N surrounding traffic participants, and determining the relative position range of the tested vehicle and the N traffic participants on the premise of covering the combination with the most complex driving task or the combination possibly causing traffic accidents;
screening a test scene: designing a scene screening rule, and screening out scenes with test values from the complex scene group to form a scene group such as a normal driving working condition, a pre-collision working condition and the like;
Determining parameters: necessary constraint conditions are added to the screened test scene groups, and all the combined scene groups are covered.
In view of the fact that the test scenes in the complex scene group obtained by greedy combination are not all effective, and the cost of the test is unnecessarily increased for scenes which are not realizable or have repeated utility, the invention also needs to eliminate useless scenes in the complex scene group according to the following scene screening rules:
separating from the actual scene and deleting;
deleting scenes with small influence on the trial run motion in the similar scenes;
for the scenes with the same influence on the tested motion, only 1 scene is reserved.
The dynamic scene mainly refers to parts such as management and control with dynamic characteristics, traffic flow and the like in simulation, is a key component of a simulation test scene, and mainly comprises the following steps: traffic management control simulation, motor vehicle/non-motor vehicle/pedestrian motion planning and feedback simulation and the like. Most of the existing dynamic simulation scene simulation systems have the problems that simulation traffic participants lack interaction with a tested vehicle in a simulation test process, a generalization method for a real scene lacks deep research, and the like. The online drainage system is designed, online game clients based on multiple platforms such as web pages and android are researched and developed, tests are carried out for the society, scientific research institutions and government departments for compensation, and real traffic interaction is achieved through participation of real users. Furthermore, real-person driving data are collected through a background, generalized behavior models are extracted based on artificial intelligence algorithms such as a deep neural network and the like to form a dynamic traffic model library, the model library calls intelligent traffic body models in corresponding scenes during testing to take over corresponding interference vehicles and pedestrians, and respective behavior modes are modified through interaction specificity of real traffic participants to achieve high anthropomorphic effect. The dynamic traffic model library and the static test scene library jointly form the core of the invention, namely, the dynamic and static combined simulation test resource library provides support for high-fidelity simulation in a typical test scene.
During testing, the automatic driving simulation platform can record all fine performances of the tested vehicle after the tested vehicle starts from the starting point, and accordingly, the algorithm performance of the tested vehicle can be evaluated. The comprehensive evaluation system for the performance of the automatic driving system, which is constructed by the invention, comprises the four aspects of vehicle intelligence, driving safety, driving comfort and traffic harmony, and obtains a comprehensive performance evaluation value based on a multi-level fuzzy comprehensive evaluation method. The performance index system constructed based on the empirical method comprises the following steps:
vehicle intelligence: evaluating the measured automatic driving algorithm from the aspects of perception, path planning, control and the like, wherein the perception aspect comprises the identification requirements on objects such as signal lamp phases, pedestrians and the like; the path planning aspect comprises requirements on stable track, obstacle avoidance and the like; the control aspect is the requirements for the execution of the planning result, such as maximum deviation, oscillation times, oscillation period and the like;
driving safety: whether the tested vehicle can cause danger to other traffic participants when running on a road is similar to whether the vehicle is stopped in a stop line at a red light, whether pedestrians are collided, whether traffic accidents occur and the like;
driving comfort is as follows: the ride experience and interaction experience of the driver during the trial run on the road. During the test, the driving experience evaluates whether the vehicle is driven stably and whether the vehicle turns smoothly according to the states of an accelerator, a brake and a steering recorded by the platform in the driving process; the interactive experience is that the operation quality and other interactive experiences of the tested automatic driving system are determined in a questionnaire mode through the on-loop of the driver;
Traffic coordination: the tested vehicle is tested and driven on the road, and the tested vehicle performs relative traffic movement with other traffic participants. During the test period, the traffic coordination is evaluated according to the indexes such as the running time, the traffic flow, the effective utilization rate of road resources and the like.
After a performance index system is established, the overall performance of the automatic driving algorithm is determined based on a multi-level fuzzy comprehensive evaluation method:
(1) index type reconciliation process
Quantitatively expressing the qualitative index as a maximum index;
quantitative indicators (maximum, minimum, intermediate and interval) are converted into maximum indicators;
(2) dimensionless treatment of indexes
The indexes can be weighted and integrated after being subjected to dimensionless treatment, and the normalization treatment method is adopted:
xijrepresenting the measured value of the jth index at the ith test.
(3) Determination of the weights of the indicators
Determining the weight of each index by adopting a sequence relation analysis method based on a 'function driving' principle:
in the index set { x1，x2，...，xj，...，xmSequentially selecting the most important indexes according to a user-defined evaluation criterion;
determination of the evaluation index x according to Table 1j-1And xjThe ratio r of the degrees of importance betweenj(wj-1/wj)；
Calculating the weight coefficient w corresponding to each index after the sortingj：
wj-1＝rjwj，j＝m，m-1，...，3，2 (3)
TABLE 1 rjValue assigning table
(4) Multilevel fuzzy comprehensive evaluation
Aiming at a single factor (a certain layer f evaluation index) x on the basis of the determination of an index systemj(j is 1, 2, m, m is the index number) to obtain the evaluation grade yk(k 1, 2.. multidot.h, h is the number of grades) and a degree of membership l between (k 1, 2.. multidot.h)jkThen, an evaluation matrix L is constructed:
determining a weight set W ═ W (W) based on a sequence relation method1，w2，...，wj，...，wm) And synthesizing with the evaluation matrix L to obtain a fuzzy comprehensive evaluation model C:
starting calculation from the lowest layer, forming an evaluation matrix of the upper-level evaluation index by the obtained result, and iterating to obtain a total evaluation model Ct；
Determining the fraction set mu (mu) corresponding to the evaluation grade1，μ2，…，μh)TPost-calculation composite score G:
G＝100Ctμ＝(c1，c2，...，ch)·(μ1，μ2，...，μh)T×100 (6)
due to the adoption of the technical scheme, compared with the prior art, the invention has the following advantages and positive effects: the support of the game engine enables the simulation test environment to be more vivid and the vehicle dynamics simulation to be simpler and more convenient. And establishing a dynamic and static combined simulation test resource library, establishing a static test scene library with each level and function of test value based on a combined inference method and a scene screening rule, and realizing real traffic interaction through the participation of real players. Furthermore, real-person driving data is collected through a background, intelligent traffic body model training is driven, a dynamic traffic model library is constructed, and support is provided for AI player driving behaviors. And finally, combining the experience of home and abroad automatic driving evaluation and the characteristics of the L1-L3 automatic driving test to construct an automatic driving system performance comprehensive evaluation system, and providing comprehensive and reasonable evaluation for algorithm test. Therefore, in general, the method has great significance for improving the overall performance of the vehicle automatic driving simulation.
Drawings
FIG. 1 is a dynamic and static combined architecture diagram of an autopilot simulation system;
FIG. 2 is a schematic diagram of a test repository construction with dynamic and static integration;
FIG. 3 is an autopilot system performance index architecture;
FIG. 4 is an autopilot system performance index architecture.
Detailed Description
The invention will be further illustrated with reference to the following specific examples.
It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.
In the embodiment of the dynamic and static combined automatic driving simulation system and the test resource library construction method thereof, the framework of the dynamic and static combined automatic driving simulation system and the construction of the test resource library thereof are respectively shown in fig. 1 and fig. 2. The system framework in fig. 1 is divided into four layers, namely a bottom layer support, a core module, a simulation platform and an evaluation system. A game engine in a bottom layer support, such as a unreal 4 engine, is used for constructing a simulation test environment, machine learning libraries such as TensorFlow and Torch7 are used for supporting training of a traffic intelligent agent model in a dynamic traffic model library, databases such as MySQL support player driving data acquisition, test scene storage and model training, ROS and the like are used as communication mechanisms of all modules in a simulation system, and finally an online drainage system is developed based on platforms such as android. Based on the bottom layer support, a core module, namely a dynamic and static combined simulation resource library is constructed, wherein the resource library comprises a static test scene library and a dynamic traffic model library. On the core module, an analog simulation server and a multi-platform user side based on platforms such as android and web pages are developed. And finally, establishing a set of evaluation system aiming at various test data of the tested vehicle recorded by the automatic driving platform.
In the embodiment, a construction scheme of the dynamic-static combination simulation resource library provided by the invention is specifically shown in fig. 2. Firstly, a typical static test scene library with test values of all levels and functions is constructed on the basis of a complex scene group formed by permutation and combination of relative positions and motion relations of a tested vehicle and all surrounding interference vehicles and on the basis of a combinatorial reasoning method and a scene screening rule. As shown in fig. 3, the present embodiment describes a static test scenario library construction method based on a combined inference method and a scenario screening rule by taking a straight behavior example of a vehicle to be tested in a two-lane straight-ahead section scenario. In the scenario shown in fig. 3, the presence of the interfering vehicles V1 to V3 has little influence on the running of the test vehicle T and is not considered. Therefore, only three movement directions (straight movement, left turning and braking) of the disturbance vehicles V4 and V5 need to be considered, and at the same time, the situation that V4 or V5 does not exist, and 12 test scenes formed by combining the disturbance vehicles with the tested vehicle in a straight movement mode are shown in the following table.
TABLE 2 straight-driving test scene of tested vehicle under double-lane straight-driving section scene
The non-influence interference vehicles are already eliminated when the vehicles are arranged and combined, so that the second and third conditions in the scene screening rule are considered, and the 12 test scenes in the table 2 are found to be not in accordance with the screening rule, so that all the test scenes are reserved. And then, determining test scene parameters by referring to an L1 level automatic driving existing test rule such as EURO-NCAP, realizing a corresponding test scene by using a game engine, and storing the corresponding test scene into a static test scene library.
And then, a multi-platform user side based on platforms such as android and webpage is opened to carry out tests for the society, scientific research institutions and government departments for compensation, and real traffic interaction is realized through participation of real users. Furthermore, real-person driving data is collected through a background, generalized behavior models are extracted based on artificial intelligence algorithms such as a deep neural network and the like to form an intelligent traffic model library under a corresponding scene, the model library calls the intelligent traffic model under the corresponding scene during testing to take over corresponding interference vehicles and pedestrians, and respective behavior modes are modified through interaction specificity with real traffic participants to achieve high anthropomorphic effect.
During testing, the automatic driving simulation platform can record all fine performances of the tested vehicle after the tested vehicle starts from the starting point, and accordingly, the performance of the algorithm is evaluated in the aspects of vehicle intelligence, driving safety, driving comfort and traffic coordination. This embodiment establishes an automatic driveability evaluation index system as shown in fig. 4. Then, the overall performance of the automatic driving algorithm is determined based on a multi-level fuzzy comprehensive evaluation method, and the detailed steps are as follows:
(1) index type reconciliation process
Quantitatively expressing the qualitative index as a maximum index;
Quantitative indicators (maximum, minimum, intermediate and interval) are converted into maximum indicators;
(2) dimensionless treatment of indexes
Carrying out dimensionless treatment on the indexes by adopting a normalization treatment method;
(3) determination of the weights of the indicators
Determining the weight of each index by adopting a sequence relation analysis method based on a 'function driving' principle:
in the index set { x1，x2，...，xj，...，xmSequentially selecting the most important indexes according to a user-defined evaluation criterion;
determination of the evaluation index x according to Table 1j-1And xjThe ratio r of the degrees of importance betweenj(wj-1/wj)；
Calculating the weight coefficient w corresponding to each index after the sortingj；
(4) Multilevel fuzzy comprehensive evaluation
Aiming at a single factor (a certain layer f evaluation index) x on the basis of the determination of an index systemj(j is 1, 2, …, m, m is index number) to obtain the evaluation grade yk(k is 1, 2, …, h, h is the number of grades) andjkthen, an evaluation matrix L is formed;
determining a weight set W ═ W (W) based on a sequence relation method1，w2，…，wj，…，wm) And synthesizing the fuzzy comprehensive evaluation model C with the evaluation matrix L. Starting calculation from the lowest layer, forming an evaluation matrix of the upper-level evaluation index by the obtained result, and iterating to obtain a total evaluation model Ct；
Determining the fraction set mu (mu) corresponding to the evaluation grade 1，μ2，...，μh)TAnd calculating a comprehensive score G.
It should be noted that the present invention can be embodied in other specific forms, and various changes and modifications can be made by those skilled in the art without departing from the spirit and scope of the invention.
Claims (9)
1. An automatic driving simulation system and a test resource library construction method thereof are characterized in that a static test scene library is constructed based on combined reasoning, real players are introduced into a test system through an online drainage system and take over interference vehicles and pedestrian traffic participating objects around a tested vehicle to realize real dynamic scene simulation, real driving data under a targeted test scene are extracted, an intelligent traffic model is trained based on a deep learning artificial intelligence algorithm to construct a dynamic traffic model library, the real players in the online drainage system are supplemented/replaced, and finally an automatic driving system performance comprehensive evaluation system is constructed by combining experience of automatic driving evaluation at home and abroad and automatic driving test characteristics of a level L1-L3, wherein the simulation test resource library comprises a static test scene library and a dynamic traffic model library, and the automatic driving system performance comprehensive evaluation system comprises vehicle intelligence, vehicle performance, And obtaining a comprehensive performance evaluation value based on a multi-level fuzzy comprehensive evaluation method.
2. The method of claim 1, wherein: the specific process of constructing the static test scene library based on the combined inference is to construct a typical static test scene library with each level and function of test value by using a combined inference method and a scene screening rule on the basis of a complex scene group formed by arranging and combining the relative positions and the motion relations of a tested vehicle and all surrounding interference vehicles.
3. The method of claim 2, wherein: the combined reasoning method comprises the following specific steps:
(1) determining a complex scene group: aiming at a specified road traffic environment, analyzing the combination of possible relative positions and movement directions of a tested vehicle and N surrounding traffic participants, and determining the relative position range of the tested vehicle and the N traffic participants on the premise of covering the combination with the most complex driving task or the combination possibly causing traffic accidents;
(2) screening a test scene: designing a scene screening rule, and screening out scenes with test values from the complex scene group to form a scene group such as a normal driving working condition, a pre-collision working condition and the like;
(3) determining parameters: necessary constraint conditions are added to the screened test scene groups, and all the combined scene groups are covered.
4. The method of claim 2, wherein: the scene screening rule is specifically as follows:
(1) separating from the actual scene and deleting;
(2) deleting scenes with small influence on the trial run motion in the similar scenes;
(3) for the scenes with the same influence on the tested motion, only 1 scene is reserved.
5. The method of claim 1, wherein: the online drainage system is a multi-platform client developed based on platforms such as a smart phone, a webpage, a driving simulator and the like, a player logs in an automatic driving simulation platform server through a network and takes over traffic participation objects including interference vehicles and pedestrians around a tested vehicle, the dynamic traffic model library is essentially a traffic body model set 'scene-intelligent body model library' corresponding to all test scenes in the static test scene library, and elements in the set are traffic participant models driven based on a learning algorithm in corresponding scenes.
6. The method of claim 5, wherein: the vehicle intelligence evaluates the measured automatic driving algorithm from a perception aspect, a path planning aspect and a control aspect, wherein the perception aspect comprises identification requirements for objects including signal lamp phases and pedestrians; the path planning aspect comprises the requirements on stable track and obstacle avoidance; the control aspects are the requirements to be performed on the planning results, such as maximum deviation, number of oscillations and period of oscillations.
7. The method of claim 5, wherein: the driving safety refers to whether the tested vehicle can cause danger to other traffic participants when running on a road, whether the tested vehicle stops in a stop line when being similar to a red light, whether the tested vehicle collides with pedestrians, and whether a traffic accident occurs, the driving comfort refers to the driving experience and interactive experience of a driver during the running of the tested vehicle on the road, and during the testing period, the driving experience evaluates whether the vehicle is driven stably and whether the vehicle turns smoothly according to the states of an accelerator, a brake and a steering recorded by a platform during the running process; and the interactive experience is determined by the operation quality of the tested automatic driving system in a questionnaire mode through the ring of the driver.
8. The method of claim 5, wherein: the traffic coordination refers to the traffic movement performance of the tested vehicle relative to other traffic participants when the tested vehicle runs on a road, and during the test period, the traffic coordination is evaluated through indexes including running time, traffic flow and road resource effective utilization rate.
9. The method of claim 5, wherein: the specific steps of the multilevel fuzzy comprehensive evaluation method after determining the comprehensive evaluation system of the automatic driving system performance are as follows:
(1) Index type reconciliation process
Quantitatively expressing the qualitative index as a maximum index;
converting quantitative index such as maximum type, minimum type, intermediate type and interval type into maximum type index;
(2) dimensionless treatment of indexes
The indexes can be weighted and integrated after being subjected to dimensionless treatment, and the normalization treatment method is adopted:
xijrepresents the measured value of the jth index at the ith test;
(3) determination of the weights of the indicators
Determining the weight of each index by adopting a sequence relation analysis method based on a 'function driving' principle:
in the index set { x1，x2，...，xj，...，xmSequentially selecting the most important indexes according to a user-defined evaluation criterion;
determination of the evaluation index x according to Table 1j-1And xjThe ratio r of the degrees of importance betweenj(wj-1/wj)；
Calculating the weight coefficient w corresponding to each index after the sortingj：
wj-1＝rjwj，j＝m，m-1，...，3，2 (3)
TABLE 1 rjValue assigning table
(4) Multilevel fuzzy comprehensive evaluation
Aiming at a single factor (a certain layer f evaluation index) x on the basis of the determination of an index systemj(j is 1, 2, m, m is the index number) to obtain the evaluation grade y k(k 1, 2.. multidot.h, h is the number of grades) and a degree of membership l between (k 1, 2.. multidot.h)jkAnd then form an evaluation matrix L
Determining a weight set W ═ W (W) based on a sequence relation method1，w2，...，wj，...，wm) And synthesizing the fuzzy comprehensive evaluation model C with the evaluation matrix L
Starting calculation from the lowest layer, forming an evaluation matrix of the upper-level evaluation index by the obtained result, and iterating to obtain a total evaluation model Ct
Determining the fraction set mu (mu) corresponding to the evaluation grade1，μ2，...，μh)TPost-calculation composite score G
G＝100Ctμ＝(c1，c2，...，ch)·(μ1，μ2，...，μh)T×100 (6)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010581488.7A CN111841012B (en) | 2020-06-23 | 2020-06-23 | Automatic driving simulation system and test resource library construction method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010581488.7A CN111841012B (en) | 2020-06-23 | 2020-06-23 | Automatic driving simulation system and test resource library construction method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111841012A true CN111841012A (en) | 2020-10-30 |
| CN111841012B CN111841012B (en) | 2024-05-17 |
Family
ID=72988411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010581488.7A Active CN111841012B (en) | 2020-06-23 | 2020-06-23 | Automatic driving simulation system and test resource library construction method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111841012B (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112286206A (en) * | 2020-11-17 | 2021-01-29 | 苏州智加科技有限公司 | Automatic driving simulation method, system, equipment, readable storage medium and platform |
| CN112464461A (en) * | 2020-11-20 | 2021-03-09 | 北京赛目科技有限公司 | Method and device for constructing automatic driving test scene |
| CN112541258A (en) * | 2020-12-08 | 2021-03-23 | 特路（北京）科技有限公司 | Test scene library of automatic driving automobile test field |
| CN112686979A (en) * | 2021-03-22 | 2021-04-20 | 中智行科技有限公司 | Simulated pedestrian animation generation method and device and electronic equipment |
| CN112732671A (en) * | 2020-12-31 | 2021-04-30 | 华东师范大学 | Automatic driving safety scene element modeling method driven by space-time trajectory data |
| CN112859810A (en) * | 2021-01-13 | 2021-05-28 | 自行科技(武汉)有限公司 | ADAS algorithm verification method and device based on Carla platform |
| CN113032261A (en) * | 2021-03-23 | 2021-06-25 | 北京三快在线科技有限公司 | Simulation test method and device |
| CN113052118A (en) * | 2021-04-07 | 2021-06-29 | 上海浩方信息技术有限公司 | Method, system, device, processor and storage medium for realizing scene change video analysis and detection based on high-speed dome camera |
| CN113064839A (en) * | 2021-06-03 | 2021-07-02 | 中智行科技有限公司 | System evaluation method and device |
| CN113377663A (en) * | 2021-06-24 | 2021-09-10 | 吉林大学 | Automatic driving simulation test method based on scene element removing characteristics |
| CN113536548A (en) * | 2021-06-29 | 2021-10-22 | 的卢技术有限公司 | Carla-based prefabricated track simulation scene construction method |
| CN113609016A (en) * | 2021-08-05 | 2021-11-05 | 北京赛目科技有限公司 | Method, device, equipment and medium for constructing vehicle automatic driving test scene |
| CN113610166A (en) * | 2021-08-10 | 2021-11-05 | 吉林大学 | Method for establishing test scene library for intelligent vehicle |
| CN113867315A (en) * | 2021-09-24 | 2021-12-31 | 同济大学 | Virtual-real combined high-fidelity traffic flow intelligent vehicle test platform and test method |
| CN114065549A (en) * | 2021-11-30 | 2022-02-18 | 广州文远知行科技有限公司 | Automatic driving level evaluation method, device, equipment and storage medium |
| CN114355864A (en) * | 2021-12-28 | 2022-04-15 | 重庆长安汽车股份有限公司 | Model-in-loop testing method, system and computer-readable storage medium for intelligent driving model development |
| WO2022160900A1 (en) * | 2021-01-29 | 2022-08-04 | 华为技术有限公司 | Test environment construction method and device |
| CN115688484A (en) * | 2022-11-30 | 2023-02-03 | 西部科学城智能网联汽车创新中心(重庆)有限公司 | WebGL-based V2X simulation method and system |
| CN116302364A (en) * | 2023-05-17 | 2023-06-23 | 清华大学 | Automatic driving reliability test method, device, equipment, medium and program product |
| CN116680932A (en) * | 2023-07-27 | 2023-09-01 | 西安深信科创信息技术有限公司 | Evaluation method and device for automatic driving simulation test scene |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140144921A (en) * | 2013-06-12 | 2014-12-22 | 국민대학교산학협력단 | Simulation system for autonomous vehicle using virtual reality |
| GB201709348D0 (en) * | 2017-06-13 | 2017-07-26 | Kompetenzzentrum-Das Virtuelle Fahrzeug | Method and process for co-simulation with virtual testing of real environments with pedestrian interaction |
| CN107727411A (en) * | 2017-10-30 | 2018-02-23 | 青岛慧拓智能机器有限公司 | A kind of automatic driving vehicle test and appraisal scene generation system and method |
| CN108595901A (en) * | 2018-07-09 | 2018-09-28 | 黄梓钥 | A kind of autonomous driving vehicle normalized security simulating, verifying model data base system |
| CN109657355A (en) * | 2018-12-20 | 2019-04-19 | 安徽江淮汽车集团股份有限公司 | A kind of emulation mode and system of road vehicle virtual scene |
| CN109765060A (en) * | 2018-12-29 | 2019-05-17 | 同济大学 | A kind of automatic driving vehicle traffic coordinating virtual test system and method |
| CN109902430A (en) * | 2019-03-13 | 2019-06-18 | 上海车右智能科技有限公司 | Traffic scene generation method, device, system, computer equipment and storage medium |
| CN110007675A (en) * | 2019-04-12 | 2019-07-12 | 北京航空航天大学 | A kind of Vehicular automatic driving decision system based on driving situation map and the training set preparation method based on unmanned plane |
| CN110243610A (en) * | 2019-05-13 | 2019-09-17 | 北京航空航天大学 | A kind of the multidimensional interference automatic Pilot test macro and method of movement and fixed Combination |
| 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 |
| CN110275859A (en) * | 2018-03-15 | 2019-09-24 | 明日娱乐欧洲公司 | Artificial intelligence engine on game engine and chip |
| CN110647053A (en) * | 2019-09-19 | 2020-01-03 | 北京智行者科技有限公司 | Automatic driving simulation method and system |
| CN110779730A (en) * | 2019-08-29 | 2020-02-11 | 浙江零跑科技有限公司 | L3-level automatic driving system testing method based on virtual driving scene vehicle on-ring |
| WO2020079066A1 (en) * | 2018-10-16 | 2020-04-23 | Five AI Limited | Autonomous vehicle planning and prediction |
| CN111123920A (en) * | 2019-12-10 | 2020-05-08 | 武汉光庭信息技术股份有限公司 | Method and device for generating automatic driving simulation test scene |
-
2020
- 2020-06-23 CN CN202010581488.7A patent/CN111841012B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140144921A (en) * | 2013-06-12 | 2014-12-22 | 국민대학교산학협력단 | Simulation system for autonomous vehicle using virtual reality |
| GB201709348D0 (en) * | 2017-06-13 | 2017-07-26 | Kompetenzzentrum-Das Virtuelle Fahrzeug | Method and process for co-simulation with virtual testing of real environments with pedestrian interaction |
| CN107727411A (en) * | 2017-10-30 | 2018-02-23 | 青岛慧拓智能机器有限公司 | A kind of automatic driving vehicle test and appraisal scene generation system and method |
| CN110275859A (en) * | 2018-03-15 | 2019-09-24 | 明日娱乐欧洲公司 | Artificial intelligence engine on game engine and chip |
| CN108595901A (en) * | 2018-07-09 | 2018-09-28 | 黄梓钥 | A kind of autonomous driving vehicle normalized security simulating, verifying model data base system |
| WO2020079066A1 (en) * | 2018-10-16 | 2020-04-23 | Five AI Limited | Autonomous vehicle planning and prediction |
| CN109657355A (en) * | 2018-12-20 | 2019-04-19 | 安徽江淮汽车集团股份有限公司 | A kind of emulation mode and system of road vehicle virtual scene |
| CN109765060A (en) * | 2018-12-29 | 2019-05-17 | 同济大学 | A kind of automatic driving vehicle traffic coordinating virtual test system and method |
| CN109902430A (en) * | 2019-03-13 | 2019-06-18 | 上海车右智能科技有限公司 | Traffic scene generation method, device, system, computer equipment and storage medium |
| CN110007675A (en) * | 2019-04-12 | 2019-07-12 | 北京航空航天大学 | A kind of Vehicular automatic driving decision system based on driving situation map and the training set preparation method based on unmanned plane |
| CN110243610A (en) * | 2019-05-13 | 2019-09-17 | 北京航空航天大学 | A kind of the multidimensional interference automatic Pilot test macro and method of movement and fixed Combination |
| 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 |
| CN110779730A (en) * | 2019-08-29 | 2020-02-11 | 浙江零跑科技有限公司 | L3-level automatic driving system testing method based on virtual driving scene vehicle on-ring |
| CN110647053A (en) * | 2019-09-19 | 2020-01-03 | 北京智行者科技有限公司 | Automatic driving simulation method and system |
| CN111123920A (en) * | 2019-12-10 | 2020-05-08 | 武汉光庭信息技术股份有限公司 | Method and device for generating automatic driving simulation test scene |
Non-Patent Citations (2)
| Title |
|---|
| KYOUNGWOOK MIN.ET AL: "SAE Level 3 Autonomous Driving Technology of the ETRI", 2019 INTERNATIONAL CONFERENCE ON INFORMATION AND COMMUNICATION TECHNOLOGY CONVERGENCE (ICTC), pages 464 - 466 * |
| 吴琼: "构建基于真实场景下自动驾驶车辆模拟仿真测试", 智能网联汽车, pages 70 - 71 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112286206A (en) * | 2020-11-17 | 2021-01-29 | 苏州智加科技有限公司 | Automatic driving simulation method, system, equipment, readable storage medium and platform |
| CN112286206B (en) * | 2020-11-17 | 2024-01-23 | 苏州智加科技有限公司 | Automatic driving simulation method, system, equipment, readable storage medium and platform |
| CN112464461A (en) * | 2020-11-20 | 2021-03-09 | 北京赛目科技有限公司 | Method and device for constructing automatic driving test scene |
| CN112464461B (en) * | 2020-11-20 | 2021-09-28 | 北京赛目科技有限公司 | Method and device for constructing automatic driving test scene |
| CN112541258A (en) * | 2020-12-08 | 2021-03-23 | 特路（北京）科技有限公司 | Test scene library of automatic driving automobile test field |
| CN112732671A (en) * | 2020-12-31 | 2021-04-30 | 华东师范大学 | Automatic driving safety scene element modeling method driven by space-time trajectory data |
| CN112859810A (en) * | 2021-01-13 | 2021-05-28 | 自行科技(武汉)有限公司 | ADAS algorithm verification method and device based on Carla platform |
| WO2022160900A1 (en) * | 2021-01-29 | 2022-08-04 | 华为技术有限公司 | Test environment construction method and device |
| CN112686979A (en) * | 2021-03-22 | 2021-04-20 | 中智行科技有限公司 | Simulated pedestrian animation generation method and device and electronic equipment |
| CN113032261A (en) * | 2021-03-23 | 2021-06-25 | 北京三快在线科技有限公司 | Simulation test method and device |
| CN113052118A (en) * | 2021-04-07 | 2021-06-29 | 上海浩方信息技术有限公司 | Method, system, device, processor and storage medium for realizing scene change video analysis and detection based on high-speed dome camera |
| CN113064839A (en) * | 2021-06-03 | 2021-07-02 | 中智行科技有限公司 | System evaluation method and device |
| CN113377663B (en) * | 2021-06-24 | 2021-12-10 | 吉林大学 | Automatic driving simulation test method based on scene element removing characteristics |
| CN113377663A (en) * | 2021-06-24 | 2021-09-10 | 吉林大学 | Automatic driving simulation test method based on scene element removing characteristics |
| CN113536548A (en) * | 2021-06-29 | 2021-10-22 | 的卢技术有限公司 | Carla-based prefabricated track simulation scene construction method |
| CN113609016B (en) * | 2021-08-05 | 2024-03-15 | 北京赛目科技股份有限公司 | Method, device, equipment and medium for constructing automatic driving test scene of vehicle |
| CN113609016A (en) * | 2021-08-05 | 2021-11-05 | 北京赛目科技有限公司 | Method, device, equipment and medium for constructing vehicle automatic driving test scene |
| CN113610166B (en) * | 2021-08-10 | 2023-12-26 | 吉林大学 | Method for establishing test scene library for intelligent vehicle |
| CN113610166A (en) * | 2021-08-10 | 2021-11-05 | 吉林大学 | Method for establishing test scene library for intelligent vehicle |
| CN113867315A (en) * | 2021-09-24 | 2021-12-31 | 同济大学 | Virtual-real combined high-fidelity traffic flow intelligent vehicle test platform and test method |
| CN114065549A (en) * | 2021-11-30 | 2022-02-18 | 广州文远知行科技有限公司 | Automatic driving level evaluation method, device, equipment and storage medium |
| CN114065549B (en) * | 2021-11-30 | 2023-04-07 | 广州文远知行科技有限公司 | Automatic driving level evaluation method, device, equipment and storage medium |
| CN114355864A (en) * | 2021-12-28 | 2022-04-15 | 重庆长安汽车股份有限公司 | Model-in-loop testing method, system and computer-readable storage medium for intelligent driving model development |
| CN114355864B (en) * | 2021-12-28 | 2024-05-03 | 重庆长安汽车股份有限公司 | Model-in-loop test method, system and computer readable storage medium for intelligent driving model development |
| CN115688484A (en) * | 2022-11-30 | 2023-02-03 | 西部科学城智能网联汽车创新中心(重庆)有限公司 | WebGL-based V2X simulation method and system |
| CN115688484B (en) * | 2022-11-30 | 2023-07-25 | 西部科学城智能网联汽车创新中心(重庆)有限公司 | V2X simulation method and system based on WebGL |
| CN116302364A (en) * | 2023-05-17 | 2023-06-23 | 清华大学 | Automatic driving reliability test method, device, equipment, medium and program product |
| CN116302364B (en) * | 2023-05-17 | 2023-08-15 | 清华大学 | Automatic driving reliability test method, device, equipment, medium and program product |
| CN116680932B (en) * | 2023-07-27 | 2023-11-21 | 安徽深信科创信息技术有限公司 | Evaluation method and device for automatic driving simulation test scene |
| CN116680932A (en) * | 2023-07-27 | 2023-09-01 | 西安深信科创信息技术有限公司 | Evaluation method and device for automatic driving simulation test scene |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111841012B (en) | 2024-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111841012B (en) | Automatic driving simulation system and test resource library construction method thereof | |
| Krishan et al. | Air quality modelling using long short-term memory (LSTM) over NCT-Delhi, India | |
| Li et al. | Survey on artificial intelligence for vehicles | |
| Li et al. | Humanlike driving: Empirical decision-making system for autonomous vehicles | |
| Chen et al. | Learning deep representation from big and heterogeneous data for traffic accident inference | |
| Algers et al. | Review of micro-simulation models | |
| Zheng et al. | Car-following behavior with instantaneous driver–vehicle reaction delay: A neural-network-based methodology | |
| CN112465395A (en) | Multi-dimensional comprehensive evaluation method and device for automatically-driven automobile | |
| Luo et al. | Guided game-based learning using fuzzy cognitive maps | |
| Xu et al. | Spatial ensemble prediction of hourly PM2. 5 concentrations around Beijing railway station in China | |
| Kumar et al. | Developing context sensitive planning criteria for transit oriented development (TOD): A fuzzy-group decision approach | |
| Meftah et al. | A virtual simulation environment using deep learning for autonomous vehicles obstacle avoidance | |
| CN117313931B (en) | Subway platform passenger evacuation time prediction method based on fire scene | |
| CN117521389A (en) | Vehicle perception test method based on vehicle-road collaborative perception simulation platform | |
| CN116946183A (en) | Commercial vehicle driving behavior prediction method considering driving capability and vehicle equipment | |
| Wang et al. | A method to automatic measuring riding comfort of autonomous vehicles: Based on passenger subjective rating and vehicle parameters | |
| Elhenawy et al. | Modeling the perception reaction time and deceleration level for different surface conditions using machine learning techniques | |
| Fouladinejad et al. | Development of a surrogate-based vehicle dynamic model to reduce computational delays in a driving simulator | |
| Malleson et al. | Agent-based modelling for Urban Analytics: State of the art and challenges | |
| Chu et al. | Three-dimensional simulation for training autonomous vehicles in smart city environments | |
| Shi et al. | Serious Games for Transportation Research: An Overview | |
| Liu et al. | Using classification and regression tree and dimension reduction in analyzing motor vehicle traffic accidents | |
| Mussone et al. | A Methodology for Modelling Driver Behaviour in Signalized Urban Intersections Using Artificial Neural Networks | |
| Ni et al. | A method for evaluating the complexity of autonomous driving road scenes | |
| Silva | Traffic lights optimization of an intersection through the application of metaheuristics |
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 |