CN115597889B - Method and system for testing vehicle based on software definition and software definition testing system - Google Patents

Method and system for testing vehicle based on software definition and software definition testing system Download PDF

Info

Publication number
CN115597889B
CN115597889B CN202211462704.1A CN202211462704A CN115597889B CN 115597889 B CN115597889 B CN 115597889B CN 202211462704 A CN202211462704 A CN 202211462704A CN 115597889 B CN115597889 B CN 115597889B
Authority
CN
China
Prior art keywords
test
vehicle
vehicles
tested
information
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
Application number
CN202211462704.1A
Other languages
Chinese (zh)
Other versions
CN115597889A (en
Inventor
李浩宇
南金瑞
曹万科
王文伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Automotive Research Institute of Beijing University of Technology
Original Assignee
Shenzhen Automotive Research Institute of Beijing University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Automotive Research Institute of Beijing University of Technology filed Critical Shenzhen Automotive Research Institute of Beijing University of Technology
Priority to CN202211462704.1A priority Critical patent/CN115597889B/en
Publication of CN115597889A publication Critical patent/CN115597889A/en
Application granted granted Critical
Publication of CN115597889B publication Critical patent/CN115597889B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M17/00—Testing of vehicles
    • G01M17/007—Wheeled or endless-tracked vehicles

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to the technical field of automatic driving, in particular to a method and a system for testing a vehicle based on software definition and a software definition testing system. The method comprises the steps of obtaining real-time road information, demand information of a tested vehicle and initial test scene data corresponding to the test demand information through a cluster structure, then carrying out data processing on the initial test scene data according to the test demand information and the real-time road information of the tested vehicle to obtain comprehensive test scene data corresponding to a test project, and finally adjusting the running parameters of other tested vehicles in a test fleet according to the comprehensive test scene data to further realize the test of the tested vehicle. By adopting the method and the system for testing the vehicle, the real driving scene can be simulated to test the vehicle to be tested, and the test result is obtained, so that the automatic driving function test is effectively guaranteed.

Description

Method and system for testing vehicle based on software definition and software definition testing system
Technical Field
The invention relates to the technical field of automatic driving vehicle testing, in particular to a method and a system for testing a vehicle based on software definition and a software definition testing system.
Background
As vehicle technology develops, more and more vehicles are equipped with advanced automatic driving functions. As a main form of future traffic participation, the reliability and safety of the functions of autonomous vehicles are important areas that should be paid attention to. However, at present, the test for the automatic driving vehicle mostly stays on a road under a simple working condition design, and a complex test cannot be performed according to a real-time road condition, so that the deviation between a result when the test for the automatic driving vehicle is actually simulated driving and a test time is large, the road condition cannot be truly reflected, the vehicle information cannot be fed back in real time, and the automatic driving function test cannot be guaranteed.
Disclosure of Invention
The invention provides a method and a system for testing vehicles based on software definition and a software definition testing system, which solve the problem that the automatic driving function test cannot be effectively guaranteed because the automatic driving vehicles cannot be accurately tested according to the real-time road condition in the prior art.
According to a first aspect, there is provided in one embodiment a method of testing a vehicle based on software definition, comprising:
acquiring real-time road information, test requirement information of a tested vehicle and initial test scene data corresponding to the test requirement information; the road information comprises movement information of a movable object on a road surface on which the detected vehicle runs; the test requirement information comprises test items for testing the automatic driving function of the tested vehicle, wherein the automatic driving function comprises overtaking, lane changing, curve or pedestrian risk avoiding;
based on the test demand information of the tested vehicle and the real-time road information, performing data processing on the initial test scene data to obtain comprehensive test scene data corresponding to the test project;
and adjusting the running parameters of other test vehicles in the test fleet according to the comprehensive test scene data, and communicating with other test vehicles in the test fleet through a content-based routing algorithm.
In an embodiment of possible implementation, the obtaining initial test scenario data corresponding to the test requirement information includes: receiving initial test scene data which are sent by a cloud platform and correspond to the test requirement information; the initial test scene data is obtained by processing test requirement information submitted by a tested vehicle after the test requirement information is received by the cloud platform; the method further comprises the following steps:
and controlling the running parameters of other test vehicles in the test fleet according to the initial test scene data.
In one possible implementation example, after adjusting the driving parameters of other test vehicles in the test fleet according to the comprehensive test scenario data, the method further includes:
judging whether the current comprehensive test scene data meets the condition that the tested vehicle realizes the automatic driving function corresponding to the test item;
if not, adjusting the current comprehensive test scene data to enable the comprehensive test scene data to meet the conditions that the tested vehicle achieves the automatic driving function corresponding to the test items.
In one possible implementation, the adjusting the current comprehensive test scenario data includes:
determining a new cluster head node vehicle;
the new cluster head node vehicle adjusts the current comprehensive test scene data according to the performance evaluation index of each test vehicle in the test fleet; the performance evaluation index at least comprises the distance between each test vehicle and the tested vehicle, the time delay between each test vehicle and the load demand between each test vehicle.
In one possible implementation, the determining a new cluster head node vehicle includes:
obtaining a strength level of each test vehicle;
determining a new cluster head node vehicle according to the robustness level; the robustness level is used for representing the performance of each test vehicle in the test fleet, and the performance comprises the communication performance between the cluster head node vehicle and the tested vehicle and between the cluster head node vehicle and the rest test vehicles.
According to a second aspect, there is provided in one embodiment a system for testing a vehicle based on software definition, comprising:
the system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring real-time road information, test requirement information of a tested vehicle and initial test scene data corresponding to the test requirement information; the road information comprises movement information of a movable object on a road surface on which the vehicle to be detected runs; the test requirement information comprises test items for testing the automatic driving function of the tested vehicle, wherein the automatic driving function comprises overtaking, lane changing, curve or pedestrian risk avoiding;
the data processing module is used for carrying out data processing on the initial test scene data based on the test requirement information of the tested vehicle and the real-time road information to obtain comprehensive test scene data corresponding to the test items;
and the adjusting module is used for adjusting the running parameters of other test vehicles in the test fleet according to the comprehensive test scene data and communicating with the other test vehicles in the test fleet through a content-based routing algorithm.
In one possible implementation embodiment, the acquiring, by the acquiring module, initial test scenario data corresponding to the test requirement information includes: receiving initial test scene data corresponding to the test demand information sent by the cloud platform; and the initial test scene data is obtained by processing the test requirement information after the cloud platform receives the test requirement information submitted by the vehicle to be tested.
In one possible implementation, the system further comprises:
and the control module is used for controlling the running parameters of other test vehicles in the test fleet according to the initial test scene data.
In an embodiment of a possible implementation, the data processing module is further configured to:
judging whether the current comprehensive test scene data meets the condition that the tested vehicle realizes the automatic driving function corresponding to the test item;
if not, adjusting the current comprehensive test scene data to enable the comprehensive test scene data to meet the conditions that the tested vehicle achieves the automatic driving function corresponding to the test items.
According to a third aspect, there is provided in one embodiment a software defined testing system comprising:
the cloud platform is used for providing test standard data corresponding to the test requirement information and initial test scene data corresponding to the test requirement information for the test fleet;
a test fleet comprising a plurality of test vehicles for:
acquiring real-time road information and test requirement information of a tested vehicle; the road information comprises movement information of a movable object on a road surface on which the vehicle to be detected runs; the test requirement information comprises test items for testing the automatic driving function of the tested vehicle, wherein the automatic driving function comprises overtaking, lane changing, curve or pedestrian risk avoiding;
based on the test demand information of the tested vehicle and the real-time road information, performing data processing on the initial test scene data to obtain comprehensive test scene data corresponding to the test project;
adjusting the running parameters of other test vehicles in the test fleet according to the comprehensive test scene data, and communicating with other test vehicles in the test fleet through a content-based routing algorithm;
testing the tested vehicle and forming a test result;
sending the test result to the cloud platform;
and the cloud platform is also used for evaluating the test result according to the test standard data to form test reference information, and the test reference information is stored along with the user information of the tested vehicle user.
According to the method and the system for testing the vehicle based on the software definition, the real-time road information, the demand information of the tested vehicle and the initial test scene data corresponding to the test demand information are obtained, then the initial test scene data are subjected to data processing according to the test demand information and the real-time road information of the tested vehicle, comprehensive test scene data corresponding to the test items are obtained, and finally the running parameters of the rest of the tested vehicles in the test fleet are adjusted according to the comprehensive test scene data, so that the tested vehicle is tested. By adopting the test method, the real driving scene can be simulated to test the tested vehicle, and the test result is obtained, so that the automatic driving function test is effectively guaranteed.
Drawings
FIG. 1 is a test scenario diagram of a test vehicle based on software definition according to the present embodiment;
FIG. 2 is a flowchart of a testing method for testing a vehicle based on software definition according to this embodiment;
fig. 3 is a flowchart for determining whether the current comprehensive test scenario data satisfies the automatic driving function according to this embodiment;
fig. 4 is a flowchart for adjusting current comprehensive test scenario data according to this embodiment;
fig. 5 is a flowchart of determining a new cluster head node vehicle according to the embodiment;
fig. 6 is a block diagram illustrating a system for testing a vehicle based on software definition according to the present embodiment.
An icon: 100. testing a motorcade; 101. a front vehicle of a side lane; 102. front vehicle; 103. a first environmental vehicle; 104. a second ambient vehicle; 105. carrying out rear vehicle; 106. a rear vehicle on a side lane; 200. a drive test unit device; 300. a cloud platform; 400. a vehicle under test; 500. an acquisition module; 600. a data processing module; 700. an adjustment module; 800. and a control module.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. Wherein like elements in different embodiments are numbered with like associated elements. In the following description, numerous details are set forth in order to provide a better understanding of the present application. However, one skilled in the art will readily recognize that some of the features may be omitted or replaced with other elements, materials, methods in different instances. In some instances, certain operations related to the present application have not been shown or described in detail in order to avoid obscuring the core of the present application from excessive description, and it is not necessary for those skilled in the art to describe these operations in detail, so that they may be fully understood from the description in the specification and the general knowledge in the art.
Furthermore, the features, operations, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. Also, the various steps or actions in the description of the methods may be transposed or transposed in order, as will be apparent to a person skilled in the art. Thus, the various sequences in the specification and drawings are for the purpose of describing certain embodiments only and are not intended to imply a required sequence unless otherwise indicated where such sequence must be followed.
The numbering of the components as such, e.g., "first", "second", etc., is used herein only to distinguish the objects as described, and does not have any sequential or technical meaning. The term "connected" and "coupled" when used in this application, unless otherwise indicated, includes both direct and indirect connections (couplings).
Compared with the existing vehicle testing mode, the software definition-based vehicle testing method provided by the invention provides a testing fleet which simulates the road scene under a more complex driving condition, can test the tested vehicle under the road scene under the more complex driving condition, and can effectively solve the problems that the road condition cannot be truly reflected and the vehicle information cannot be fed back in real time in the existing self-driving testing model, so that the automatic driving function testing cannot be guaranteed.
The method, system and system for testing vehicles based on software definition according to the present invention will be described in detail with reference to the accompanying drawings.
The tested vehicle of the embodiment is an automatic driving vehicle, in a system for testing the automatic driving vehicle, a test vehicle system of a test fleet can flexibly define test scenes according to different test requirements, and test scene data are acquired from a data center stored by a cloud platform or the test vehicle system, so that the aim of carrying out all-around test on different functions of the tested vehicle is fulfilled, and the test vehicles can be flexibly added or reduced according to the test requirements.
For example, as shown in fig. 1, when detecting one of the automatic driving functions of the vehicle under test 400, the vehicle under test system may autonomously define test scenario data according to a test requirement, that is, define a clustering strategy of the vehicle under test 100, specifically, the drive test unit device 200 transmits information, such as a relative distance, a relative direction, and a number between a movable object (a vehicle, a pedestrian, etc.) on a driving road of the vehicle under test 400 and the vehicle under test 400, to the cloud platform 300, and the cloud platform 300 selects the test scenario data (that is, the clustering strategy of the vehicle under test 100) according to the information and the test requirement information of the vehicle under test 400.
The test scenario data (i.e. the clustering strategy of the test fleet 100) is that the test fleet 100 is composed of a preset number of test vehicles, and the test vehicles may be distributed in front and rear of the same lane of the vehicle 400, left and right lanes of the vehicle 400, and the like. The preset number can be set according to the test requirement, and the preset number is 6 in the embodiment, that is, the test fleet 100 can be composed of six test vehicles, namely, a front vehicle 101 located in a side lane of the tested vehicle 400, a rear vehicle 106 located in a side lane of the tested vehicle 400, a front vehicle 102 located in front of the tested vehicle 400, a rear vehicle 105 located behind the tested vehicle 400, a first environmental vehicle 103 located in another side lane of the tested vehicle 400, and a second environmental vehicle 104 located in another side lane of the tested vehicle 400. Then, the cloud platform 300 of the test system selects one of the vehicles as a cluster head node vehicle, and the running parameters of the rest of the test vehicles in the test fleet 100 are controlled by the cluster head node vehicle, so that the test of the automatic driving function of the tested vehicle 400 is realized. The functional modules configured by the individual nodes of each test vehicle are identical to the physical structures, the positions of the vehicles can be randomly arranged, the test vehicle nodes can be reasonably added according to a test scene, the design can also avoid the condition that the whole test fleet 100 cannot be tested due to the fault of one test vehicle, the test platform does not need to be replaced, and the redundancy of the structural system is improved. In addition, interruption is not needed in the testing process, the testing fleet 100 can set a testing scheme according to the cloud platform 300, time is effectively saved, and testing efficiency is improved.
Before the following steps are carried out, cluster head node vehicles need to be selected, and particularly, the selection of the cluster head node vehicles is determined according to the robustness level of each test vehicle. The strength level can be understood as the strength of the communication performance between the test vehicles, and specifically is as follows: the data information of all the test vehicles in the test fleet 100 is acquired through the cloud platform 300 in the test system, wherein the data information includes time delay, packet loss rate and the like, then the cloud platform 300 judges the communication performance between each test vehicle and other test vehicles through the data information, under normal conditions, the time delay is generally lower than 20ms, the packet loss rate when the distance between two vehicles is less than 100m is less than 0.1%, and when the packet loss rate when the distance between two vehicles exceeds 100m is greater than 0.5%, the normal communication between data is affected. It should be noted that, in this embodiment, for the communication performance of the vehicle, the determination conditions of the time delay and the packet loss rate of the vehicle under different test scenes are different, the determination conditions of the time delay and the packet loss rate are applicable to urban roads, buildings and trees are arranged on two sides of the road, other vehicles and traffic participants exist in the road, and for roads with fewer obstacles on two sides of the road and wide space, the smaller the packet loss rate is, the better the communication performance is within a certain distance. The test vehicle with the strongest communication performance is selected as the cluster head node vehicle through the cloud platform 300, and then the running parameters of the rest of the test vehicles are controlled through the cluster head node vehicle according to the test requirements.
Specifically, the test method for controlling the test fleet 100 to execute the test function of the tested vehicle 400 by the cluster head node vehicle is as follows:
referring to fig. 2, the present embodiment provides a method for testing a vehicle based on software definition, which includes the following steps:
step 1: acquiring real-time road information, test requirement information of a tested vehicle 400 and initial test scene data corresponding to the test requirement information; the road information includes movement information of a movable object on a road surface on which the vehicle 400 under test travels; the test demand information includes test items for testing the automatic driving function of the vehicle 400 under test, the automatic driving function including passing, lane changing, curve, pedestrian avoidance, hill braking, or lane keeping. These automatic driving functions may be fully automatic or may be driving assistance. In this embodiment, the test requirement information may include one or more test items, such as one or more of a test item for testing overtaking, a test item for testing lane change, a test item for testing curve, a test item for testing pedestrian risk avoidance, a test item for testing braking on/off a slope, and a test item for testing lane keeping. The initial test scenario data in this embodiment is a test scenario stored in the test system and used for testing the automatic driving function of the vehicle 400 to be tested, and the test fleet 100 tests the vehicle 400 to be tested based on the test scenario.
In the test fleet 100, control of the remaining test vehicles is achieved by the cluster head node vehicles. Specifically, the cluster head node vehicles acquire road information in real time through the drive test unit devices 200, wherein the drive test unit devices 200 are measuring devices located on the side of a test road, power is supplied through a road surface power supply device, and information acquisition and interaction can be performed on the road with elements such as running vehicles, pedestrians, moving objects and the like, information is transmitted to the cluster head node vehicles, data filtering and feature extraction can be performed on specific attention parameters, specific feature information is transmitted to the cloud platform 300, and the specific feature information can include real-time position information, distance information and the like of each test vehicle.
In addition, the cluster head node vehicle may further perform information acquisition through its own communication module, for example, acquire a real-time relative position between the cluster head node vehicle and the rest of the test vehicles and acquire a real-time relative position between the cluster head node vehicle and the tested vehicle 400, and the cluster head node vehicle may further directly communicate with the tested vehicle 400 through the communication module to acquire the test requirement information of the tested vehicle 400. Specifically, the test requirement information is information for testing various automatic driving functions of the vehicle 400 to be tested, and includes overtaking, lane changing, curve, pedestrian risk avoidance (braking), passing (braking performance on an uphill or downhill), control stability, ride comfort, and the like.
In addition, the cluster head node vehicles can also directly communicate with the cloud platform 300 through the communication modules of the cluster head node vehicles. The user can operate the tested vehicle 400 to generate the test requirement information, the tested vehicle 400 sends the test requirement information to the cloud platform 300, or the user can operate the mobile terminal to generate the test requirement information, and the mobile terminal sends the test requirement information to the cloud platform 300. A corresponding computing center exists in the cloud platform 300, the computing center may combine specific feature information transmitted by the drive test unit device 200 to form initial test scenario data corresponding to the test demand information according to the test demand information, and then the cluster head node vehicle may directly acquire the initial test scenario data from the cloud platform 300. It should be noted that, in the present scheme, the calculation and storage of the initial test scenario data are not limited to the cloud platform 300, and may be other data processing systems, for example, a data processing system directly disposed on the test vehicle, as long as the functions and functions of the data processing system can be implemented. It should be noted that, other test vehicles in the test fleet 100 also have the function of cluster head node vehicles, and each test vehicle is equipped with a module capable of communicating with the vehicle 400 to be tested, so that functional signal transmission can be performed with the vehicle 400 to be tested in all directions at any time, failure in communication caused by interference and shielding of signals of individual vehicles or module failure is avoided, redundancy of signals is enhanced, and reliability of achieving a test function target is enhanced.
For example, any one of the test vehicles may communicate with the other test vehicles through its own communication module, and communicate with the cloud platform 300. When the test function changes, a new cluster head node vehicle needs to be selected from the test fleet 100, and at this time, the selected test vehicle as the new cluster head node vehicle can also receive initial test scenario data corresponding to the test demand information sent by the cloud platform 300, and can correspondingly control and adjust the test fleet 100 according to the initial test scenario data.
For the communication function of the cluster head node vehicles, a test vehicle can be selected independently to serve as a gateway vehicle, information interaction between the gateway vehicle and surrounding cluster fleet is achieved, namely communication between the cluster head node vehicles and the tested vehicles 400 and communication between the test vehicle and the cluster head node vehicles are achieved through the gateway vehicle, therefore, high efficiency, time saving and information unification of relevant information can be guaranteed in the transmission process, and the gateway vehicle selects one with the best communication performance in the test fleet 100 to serve as the gateway vehicle.
And 2, step: based on the test demand information and the real-time road information of the vehicle 400 to be tested, the initial test scenario data is subjected to data processing, and comprehensive test scenario data corresponding to the test items is obtained. And optimizing the initial test scene data by using the real-time road information, thereby obtaining the comprehensive test scene data corresponding to the test project.
The cluster head node vehicle acquires initial test scene data stored by the cloud platform 300, and then performs data optimization calculation on the initial test scene data based on the acquired test requirement information and real-time road information of the tested vehicle 400, and specifically, a GAAC-based content service network routing optimization algorithm applicable to a vehicle network and extending according to a routing protocol in a self-organizing network is realized by adding a control layer on routing forwarding, and is used for controlling routing forwarding of a forwarding information base of a content center network, so that optimization processing is performed on the initial test scene model in a big data background, and comprehensive test scene data corresponding to a test item is obtained, and the comprehensive test scene data is specifically used for making a test cluster strategy model corresponding to the test item of the tested vehicle 400, the test cluster strategy model realizes flexible setting of a vehicle test scene based on the demand, ensures updating of a test function, and realizes optimization processing on the test cluster model made by the big data background, so as to realize real simulation of the tested vehicle fleet function of the tested vehicle 400, and optimization of the tested vehicle fleet parameters in the test process of the tested vehicle fleet.
And step 3: the driving parameters of other test vehicles in the test fleet 100 are adjusted according to the integrated test scenario data and communicated with the other test vehicles in the test fleet 100 through a content-based routing algorithm.
When the test fleet 100 of the present embodiment performs a test on the vehicle 400 under test, the functional modules and physical modules configured for each test vehicle in the test fleet 100 have the same structure.
In actual testing, the test fleet 100 is changed in real time, for example, to test the overtaking function, when the vehicle 400 under test finds that the front vehicle 102 in the test fleet 100 is traveling too slowly, which needs to perform the overtaking operation, for a specific road scene, such as a straight road, the following situations may occur when testing the test item of the overtaking function: when the front vehicle 102 in the test fleet 100 suddenly slows down, the tested vehicle 400 cannot timely turn to a lane change and the front vehicle 102 has rear-end collision and is stuck; when the tested vehicle 400 is successfully driven by changing lanes, the rear vehicle 106 of the side lane in the tested vehicle team 100 collides with the rear vehicle; when the tested vehicle 400 is successfully driven for lane change, but the tested vehicle collides with a front vehicle 101 of a side lane in the test fleet 100; when the tested vehicle 400 successfully changes the lane to run, but the vehicle 101 in front of the side lane in the test fleet 100 also runs too slowly at the moment, so that the tested vehicle 400 cannot realize overtaking operation and is stuck; when the tested vehicle 400 is successfully driven for lane change, but the front vehicle 102 in the tested vehicle team 100 knocks into the rear when the road is changed back; when the tested vehicle 400 is successfully changed to run and is smoothly changed back to the middle lane, the successful overtaking is realized.
In the above case, the position distance of the test fleet 100 needs to be adjusted in time, but the relative position between the test vehicles does not change in order to ensure the stability of the test. Specifically, the cluster head node vehicle performs comprehensive judgment calculation on the test function of the test scene according to the obtained comprehensive test scene data, and transmits a control signal command to the other test vehicles in the test fleet 100 through information interaction between the cluster head node vehicle and the other test vehicles, so that the other test vehicles can be accurately located at corresponding positions in the current test scene, the test condition can be sent to the tested vehicle 400 through the interaction information between the cluster head node vehicle and the tested vehicle 400, the tested vehicle 400 executes corresponding functions, and the test on the corresponding automatic driving function of the tested vehicle 400 is stably and reliably realized.
The algorithm for communication between the cluster head node vehicle and the rest of the test vehicles in the test fleet 100 is implemented by using a content-based routing algorithm, specifically, the algorithm is a content-oriented and service-network-oriented routing optimization algorithm GAAC, and the algorithm implementation steps are as follows:
(1) Initializing a network, and deleting paths which do not meet constraint conditions;
(2) Initializing a genetic algorithm population and genetic parameters, and mainly used for initializing cross probability and mutation probability in the genetic algorithm, wherein the cross indicates that a next generation individual exchanges one of two previous generation individuals on a certain same gene to obtain a new individual, and the cross probability is used for indicating that the new individual has higher fitness than the two previous generation individuals; the mutation is similar to gene mutation in the biological propagation process and is the key for diversifying the organisms, the chromosome is mutated at a certain gene position with a certain probability, the mutation can be carried out in situ or can be mutated into other positions, so that the local performance of the algorithm can be improved, the mutation probability represents the mutation probability of an individual, the mutation probability has a certain influence on the whole population, if the mutation probability is too small, the information lost too early can not be recovered, the generation of new individuals is not facilitated, and if the mutation probability is too large, the random search can be realized;
(3) Performing a plurality of iterations according to a genetic algorithm, and generating network initial pheromone distribution;
(4) Initializing control parameters of the ant colony algorithm, wherein the control parameters are mainly used for initializing a weighting coefficient loaded in the ant colony algorithm and an evaporation coefficient of an pheromone, the evaporation coefficient represents the evaporation rate of the pheromone, and the pheromone evaporation can avoid that the algorithm converges to a local optimal solution too fast, thereby being beneficial to the expansion of a search area;
(5) And carrying out a plurality of iterations according to the ant colony algorithm to obtain an optimal solution.
The ant colony algorithm in the embodiment is a known algorithm in the prior art, and the algorithm is adopted to provide a hybrid genetic ant colony algorithm to improve CCN route forwarding so as to improve the route forwarding performance, reduce redundant retrieval and reduce the transmission of network repeated content.
In one possible implementation, the obtaining initial test scenario data corresponding to the test requirement information includes: receiving initial test scene data which is sent by the cloud platform 300 and corresponds to the test requirement information; the initial test scene data is obtained by processing the test requirement information after the cloud platform 300 receives the test requirement information submitted by the vehicle 400 to be tested; the method of testing an autonomous vehicle further comprises: the driving parameters of other test vehicles in the test fleet 100 are controlled according to the initial test scenario data.
In this embodiment, specifically, the cloud platform 300 receives test requirement information sent by a user, a corresponding processing module and a corresponding memory module exist in the cloud platform 300, the processing module is configured to form initial test scenario data for the requirement from the test requirement information submitted by the user, and the initial test scenario data formed after processing by the processing module is stored in the memory module. After acquiring the initial test scenario data in the memory module, the cluster head node vehicle can control the driving parameters of the rest of the test vehicles in the test fleet 100 according to the initial test scenario data, and then test the tested vehicle 400. The memory module comprises a user file submodule, the cluster-head node vehicle can send test data formed after the test is completed to the cloud end platform 300, and the cloud end platform 300 stores the test data into the user file submodule, so that a user can conveniently call the test data at any offline place, and a software defined test function framework for the user service of the tested vehicle 400 is realized. In addition, test data formed after the test is finished can be stored in the user file submodule as new initial test scene data, so that the initial test scene data can be updated and optimized continuously, and the accuracy of test performance is guaranteed.
In one possible implementation, as shown in fig. 3, after adjusting the driving parameters of the other test vehicles in the test fleet 100 according to the comprehensive test scenario data, the method further includes:
and 4, step 4: and judging whether the current comprehensive test scene data meets the condition that the tested vehicle 400 realizes the automatic driving function corresponding to the test item.
And 5: if not, adjusting the current comprehensive test scene data to enable the comprehensive test scene data to meet the condition that the tested vehicle 400 achieves the automatic driving function corresponding to the test item.
In the actual testing process, after the driving parameters of other testing vehicles in the testing fleet 100 are adjusted according to the comprehensive testing scenario data, it is further required to determine whether the adjusted testing fleet 100 can meet the testing conditions of the tested vehicle 400. The conditions for determining that the vehicle 400 under test realizes the automatic driving function corresponding to the test item specifically include: the tested vehicle 400 obtains the distance between the tested vehicle and the surrounding vehicles; judging whether the distance is within a safe distance for testing an automatic driving function; if the distance is within the safe distance, the judgment condition is met, otherwise, the judgment condition is not met.
Specifically, for example, the lane change function of the vehicle 400 to be tested needs to be tested, before lane change, it needs to be determined in advance whether the vehicle 400 to be tested meets the lane change condition, the vehicle 400 to be tested collects information of surrounding vehicles through its own radar or sensor, and then calculates through its own central processing unit, reasonably estimates the distance in the test scene, and determines whether the vehicle can change lanes on the premise of ensuring the safe distance, the safe distance is determined to be multidirectional, for example, the lateral direction, the longitudinal direction and the like should meet the lane change condition; if the lane change condition is met through calculation, information meeting the lane change condition is sent to the cluster head node vehicles, current comprehensive test scene data is triggered, and the cluster head node vehicles control the test fleet 100 to test the automatic driving function of the tested vehicle 400; if the tested vehicle 400 does not meet the lane changing condition through calculation of the central processing unit of the tested vehicle 400, the information which does not meet the lane changing condition is transmitted to the cluster head node vehicle, the cluster head node vehicle adjusts the current comprehensive test scene data, and other test vehicles in the test fleet 100 are adjusted according to the distance between each test vehicle and the tested vehicle 400, the time delay between each test vehicle and the load requirement between each test vehicle to enable the test vehicles to meet the test condition, so that the high efficiency of information transmission between the test vehicles is guaranteed. If the result calculated by the cpu of the vehicle under test 400 is contrary to the actual test scenario data, the corresponding test scenario data is modified until the test accuracy can meet the test requirement.
In one possible implementation, as shown in fig. 4, adjusting the current comprehensive test scenario data includes:
step 51: a new cluster head node vehicle is determined.
Step 52: the new cluster head node vehicle adjusts the current comprehensive test scene data according to the performance evaluation index of each test vehicle in the test fleet 100; the performance evaluation index includes at least a distance between each test vehicle and the vehicle 400 under test, a time delay between each test vehicle, and a load demand between each test vehicle.
Since the testing process is a dynamic process, real-time updates to the test vehicles in the test fleet 100 are required. In the actual testing process, with the change of the testing scene, the cluster head node vehicles need to be updated first, after the updating is completed, the new cluster head node vehicles can adjust the current testing scene data, and specifically, the relevant positions of the other testing vehicles in the current testing scene are adjusted according to the performance evaluation indexes of the other testing vehicles in the testing fleet 100. The performance evaluation index of each test vehicle specifically includes a distance between each test vehicle and the vehicle 400 to be tested, a time delay between each test vehicle, and a load demand between each test vehicle.
It should be noted that, during the lane changing motion process, the relative position between the test fleet 100 and the tested Vehicle 400 changes faster, so that the information interaction between the two is more frequent, and in order to ensure the timeliness of the information transmission and the rapid confirmation of the relative position during the motion process, the Communication between the vehicles can adopt one or more of different wireless transmission modes such as Cellular Vehicle networking (C-V2X), short Distance Short Range Communication (DSRC), wireless sensing networks (ZigBee, bluetooth, etc.), wireless local area networks (WLAN, wiFi), etc., and a wireless transmission management system matched with the same; preferably, the wireless transmission mode can be executed according to communication protocols under standards such as a dedicated short-range communication protocol DSRC and an LTE-V/C-V2X, redundancy of communication signals can be achieved, the communication modules are the same as the international standard, the vehicle 400 to be tested does not need to specially designate a special signal module according to the test requirement, the test cost can be effectively saved, and the test applicability is enhanced. In addition, in the process of information interaction between the test fleet 100 and the vehicle 400, a functional test needs to be performed on the premise of ensuring safety, so when the criteria of the two are in conflict, the determination of the vehicle 400 should be used as the criterion, and the corresponding criterion in the test fleet 100 is corrected through information transmission.
In this embodiment, the updating of the cluster head node vehicles is based on the same principle as the selection of the cluster head node vehicles, and referring to fig. 5, the method specifically includes the following steps:
step 511: a robustness rating is obtained for each test vehicle.
Step 512: determining a new cluster head node vehicle according to the robustness level; the robustness level is used to characterize the performance of each test vehicle in the test fleet 100, including the communication performance between the cluster head node vehicle and the vehicle under test 400, and between the cluster head node vehicle and the remaining test vehicles.
In this embodiment, the selection of the new cluster head node vehicle is determined according to the robustness level of each test vehicle, that is, by judging the communication performance of each test vehicle, and selecting the vehicle with the strongest communication performance as the new cluster head node vehicle. Specifically, the cloud platform 300 of the test system obtains data information of all test vehicles in the test fleet 100, where the data information includes time delay, packet loss rate, jitter, and the like, and then the cloud platform 300 determines the communication performance between each vehicle and other vehicles through the data information, selects the test vehicle with the strongest communication performance as a new cluster head node vehicle, and then controls the running parameters of the other test vehicles according to the test requirements through the new cluster head node vehicle. It should be noted that, in this embodiment, the calculation of the time delay, the packet loss rate, the jitter, and the like of each vehicle can be realized by using the existing technology, and this embodiment does not make much requirements on this technology.
According to a second aspect, as shown in fig. 6, the system for testing a vehicle based on software definition provided in the embodiment includes:
an obtaining module 500, configured to obtain real-time road information, test requirement information of the vehicle 400 to be tested, and initial test scenario data corresponding to the test requirement information; the road information includes movement information of a movable object on a road surface on which the vehicle 400 under test travels; the test demand information includes test items for testing the automatic driving function of the vehicle 400 under test, the automatic driving function including passing, lane changing, curve, or pedestrian danger avoidance.
The data processing module 600 is configured to perform data processing on the initial test scenario data based on the test demand information and the real-time road information of the vehicle 400 to be tested, so as to obtain comprehensive test scenario data corresponding to the test item.
And the adjusting module 700 is configured to adjust the driving parameters of other test vehicles in the test fleet 100 according to the comprehensive test scenario data, and communicate with the other test vehicles in the test fleet 100 through a content-based routing algorithm.
In the test fleet 100, the cluster head node vehicles control the remaining test vehicles. Specifically, the acquisition module 500 of the cluster head node vehicle acquires the road information in real time through the drive test unit device 200, wherein the drive test unit device 200 supplies power through a road surface power supply device, and performs information acquisition and interaction with elements such as driving vehicles, pedestrians, moving objects and the like on the road, and transmits the information to the cluster head node vehicle, and after data filtering and feature extraction can be performed on specific attention parameters, specific feature information is transmitted to the cloud platform 300, wherein the specific feature information can include real-time position information, distance information and the like of each test vehicle.
The cluster head node vehicle acquisition module 500 may be a radar, a sensor, or the like, which may directly communicate with the vehicle 400 to acquire the test requirement information of the vehicle 400. Specifically, the test requirement information is information for testing various automatic driving functions of the vehicle 400 to be tested, and includes overtaking, lane changing, curve, pedestrian risk avoidance (braking), passing a bridge (braking performance on an uphill slope or a downhill slope), handling stability, smoothness, and the like.
In addition, the cluster head node vehicle acquisition module 500 may also communicate directly with the cloud platform 300. The user sends the test requirement information of the vehicle 400 to be tested to the cloud platform 300, or the user can operate the mobile terminal to generate the test requirement information, and the mobile terminal sends the test requirement information to the cloud platform 300. A corresponding computing center exists in the cloud platform 300, the computing center combines the specific feature information transmitted by the drive test unit device 200, initial test scenario data corresponding to the test requirement information is formed according to the test requirement information, and then the acquisition module 500 can directly acquire the initial test scenario data from the cloud platform 300. It should be noted that, in the present scheme, the calculation and storage of the initial test scenario data are not limited to the cloud platform 300, and may be other data processing systems as long as the functions and functions of the system can be realized. It should be noted that, other test vehicles in the test fleet 100 also have the function of cluster head node vehicles, and each test vehicle is equipped with a module capable of communicating with the vehicle 400 to be tested, so that functional signal transmission can be performed with the vehicle 400 to be tested in all directions at any time, and communication failure caused by interference and shielding of signals of individual vehicles or module failure is avoided, thereby enhancing signal redundancy and enhancing reliability of achieving a test function target.
For example, any one of the test vehicles may communicate with the other test vehicles and with the cloud platform 300 through its own communication module. When the test function changes, a new cluster head node vehicle needs to be selected from the test fleet 100, and at this time, the selected test vehicle as the new cluster head node vehicle can also receive initial test scenario data corresponding to the test demand information sent by the cloud platform 300, and can correspondingly control and adjust the test fleet 100 according to the initial test scenario data.
The data processing module 600 of the cluster head node vehicle performs data optimization calculation on the initial test scenario data by retrieving the initial test scenario data stored by the cloud platform 300, and then performs a GAAC-based content service network routing optimization algorithm applicable to a vehicle network, which is extended according to a routing protocol in a self-organizing network, based on the obtained test requirement information and real-time road information of the vehicle under test 400, the algorithm is to add a control layer on routing forwarding for controlling routing forwarding of a forwarding information base FIB of a content center network, so as to achieve optimization processing on the initial test scenario model in a big data background, and obtain comprehensive test scenario data corresponding to a test item, the comprehensive test scenario data is specifically used for making a test cluster policy model corresponding to the test item of the vehicle under test 400, the test cluster policy model achieves flexible setting of a vehicle test scenario based on research and development requirements, updating of a test function is ensured, and after passing the big data, the optimized processing on the test cluster policy model made by the platform is performed, so as to achieve simulation of the test function of the vehicle under test 400 in the test fleet, and optimization process of the vehicle under test fleet is performed in the test process of the vehicle under test 400.
When the test fleet 100 of the present embodiment performs a test on the vehicle under test 400, each test vehicle in the test fleet 100 has the same configuration of functional modules and physical modules.
In actual testing, the test fleet 100 is changed in real time, for example, to test the overtaking function, when the vehicle 400 under test finds that the front vehicle 102 in the test fleet 100 is traveling too slowly, which requires overtaking operation, for a specific road scene, such as a straight road, the following situations may occur when testing the overtaking function test item: when the front vehicle 102 in the test fleet 100 suddenly slows down, the tested vehicle 400 cannot timely steer to change lanes and the front vehicle 102 collides with the tail and is stuck; when the tested vehicle 400 successfully changes the lane to drive, but the rear vehicle 106 of the side lane in the tested motorcade 100 knocks into the rear; when the tested vehicle 400 successfully changes the lane to drive, but the tested vehicle collides with a front vehicle 101 of a side lane in the test fleet 100; when the tested vehicle 400 successfully changes the lane to drive, but the front vehicle 101 of the side lane in the test fleet 100 also drives too slowly at this time, so that the tested vehicle 400 cannot realize overtaking operation and is stuck; when the tested vehicle 400 is successfully driven for lane change, but the front vehicle 102 in the tested vehicle team 100 knocks into the rear when the road is changed back; when the tested vehicle 400 is successfully changed to run and is smoothly changed back to the middle lane, the successful overtaking is realized.
In the above case, the position distance of the test fleet 100 needs to be adjusted in time, but the relative position between the test vehicles does not change in order to ensure the stability of the test. Specifically, the cluster head node vehicle performs comprehensive judgment calculation on the test function of the test scene according to the obtained comprehensive test scene data, and transmits a control signal command to the other test vehicles in the test fleet 100 through information interaction between the cluster head node vehicle and the other test vehicles, so that the other test vehicles can be accurately positioned at corresponding positions in the current test scene, the test condition can be sent to the tested vehicle 400 through the interaction information between the cluster head node vehicle and the tested vehicle 400, the tested vehicle 400 executes corresponding functions, and the test on the corresponding automatic driving function of the tested vehicle 400 is stably and reliably realized.
The algorithm for communication between the cluster head node vehicle and the rest of the test vehicles in the test fleet 100 is implemented by using a content-based routing algorithm, specifically, the algorithm is a content-oriented and service-network-oriented routing optimization algorithm GAAC, and the algorithm implementation steps are as follows:
(1) Initializing the network, and deleting paths which do not meet constraint conditions;
(2) Initializing a genetic algorithm population and genetic parameters, and mainly used for initializing cross probability and variation probability in the genetic algorithm, wherein the cross indicates that a next generation individual exchanges one of two previous generation individuals on a certain same gene to obtain a new individual, and the cross probability is used for indicating that the new individual has higher fitness than the two previous generation individuals; the mutation is similar to gene mutation in the process of biological propagation and is the key for diversifying the organisms, a chromosome is mutated at a certain gene position with a certain probability, the mutation can be carried out in situ or can be mutated into other positions, so that the local performance of the algorithm can be improved, the mutation probability represents the mutation probability of an individual, the mutation probability has a certain influence on the whole population, if the mutation probability is too small, the gene position loses information too early and cannot be recovered, the generation of new individuals is not facilitated, and if the mutation probability is too large, random search can be realized;
(3) Performing a plurality of iterations according to a genetic algorithm, and generating network initial pheromone distribution;
(4) Initializing control parameters of the ant colony algorithm, wherein the control parameters are mainly used for initializing a weighting coefficient loaded in the ant colony algorithm and an evaporation coefficient of an pheromone, the evaporation coefficient represents the evaporation rate of the pheromone, and the pheromone evaporation can avoid the algorithm from converging into a local optimal solution too fast and is beneficial to the expansion of a search area;
(5) And carrying out a plurality of iterations according to the ant colony algorithm to obtain an optimal solution.
The ant colony algorithm in the embodiment is a known algorithm in the prior art, and the algorithm is adopted to provide a hybrid genetic ant colony algorithm to improve CCN route forwarding so as to improve the route forwarding performance, reduce redundant retrieval and reduce the transmission of network repeated content.
The selection of cluster head node vehicles is determined according to the robustness level of each test vehicle. The test vehicle system acquires data information of all test vehicles in the test fleet 100, judges the communication performance between each vehicle and other vehicles according to the data information, selects the test vehicle with the strongest communication performance as a cluster head node vehicle, and controls the running parameters of the other test vehicles according to test requirements through the cluster head node vehicle.
In one possible implementation embodiment, the obtaining module 500 obtains the initial test scenario data corresponding to the test requirement information, including: receiving initial test scene data corresponding to the test demand information sent by the cloud platform 300; the initial test scenario data is obtained by performing data processing on the test requirement information after the cloud platform 300 receives the test requirement information submitted by the vehicle 400 to be tested.
In this embodiment, specifically, the user submits the test requirement information to the cloud platform 300, the cloud platform 300 has a corresponding processing module and a corresponding memory module, the processing module is configured to form the test requirement information submitted by the user into initial test scenario data for the requirement, and the initial test scenario data formed after the processing module processes the test requirement information is stored in the memory module. After acquiring the initial test scene data in the memory module, the cluster head node vehicle can control the running parameters of the rest of the test vehicles in the test fleet 100 according to the initial test scene data, and then test the tested vehicle 400. The memory module further comprises a user profile submodule, test data formed after the test is completed can be sent to the cloud platform 300 by the cluster head node vehicle, and the cloud platform 300 stores the test data into the user profile submodule, so that a user can conveniently call the test data at any offline place, and a software defined test function framework for the user service of the tested vehicle 400 is realized. In addition, test data formed after the test is finished can be stored in the user file submodule as new initial test scene data, so that the initial test scene data can be updated and optimized continuously, and the accuracy of test performance is guaranteed.
In one possible implementation example, referring to fig. 6, the test vehicle system further includes a control module 800, and the control module 800 is configured to control the driving parameters of other test vehicles in the test fleet 100 according to the initial test scenario data.
In this embodiment, the processing module in the cloud platform 300 may form the test requirement information submitted by the user into initial test scenario data for the requirement, and the control module 800 of the cluster head node vehicle can control the driving parameters of the rest of the test vehicles in the test fleet 100 according to the initial test scenario data after receiving the initial test scenario data, so as to test the tested vehicle 400.
In one possible implementation, the data processing module 600 is further configured to:
and judging whether the current comprehensive test scene data meets the condition that the tested vehicle 400 realizes the automatic driving function corresponding to the test item.
If not, adjusting the current comprehensive test scene data to enable the comprehensive test scene data to meet the condition that the tested vehicle 400 realizes the automatic driving function corresponding to the test item.
In this embodiment, in the actual testing process, after the driving parameters of other testing vehicles in the testing fleet 100 are adjusted according to the comprehensive testing scenario data, the data processing module 600 further needs to determine whether the adjusted testing fleet 100 can meet the testing conditions of the tested vehicle 400. Specifically, for example, the lane change function of the vehicle 400 to be tested needs to be tested, before lane change, it needs to be determined in advance whether the vehicle 400 to be tested meets the lane change condition, the vehicle 400 to be tested collects information of surrounding vehicles through its own radar or sensor, and then calculates through its own central processing unit, reasonably estimates the distance in the test scene, and determines whether the vehicle can change lanes on the premise of ensuring the safe distance, the safe distance is determined to be multidirectional, for example, the lateral direction, the longitudinal direction and the like should meet the lane change condition; if the lane change condition is met through calculation, information meeting the lane change condition is sent to the cluster head node vehicles, current comprehensive test scene data is triggered, and the cluster head node vehicles control the test fleet 100 to test the automatic driving function of the tested vehicle 400; if the detected vehicle 400 does not meet the lane changing condition calculated by the central processing unit of the detected vehicle 400, the information which does not meet the lane changing condition is transmitted to the cluster head node vehicle, the cluster head node vehicle adjusts the current comprehensive test scene data, and the rest of the test vehicles in the test fleet 100 are adjusted according to the distance between each test vehicle and the detected vehicle 400, the time delay between each test vehicle and the load requirement between each test vehicle to enable the test vehicles to meet the test condition, so that the efficiency of information transmission between the test vehicles is also ensured.
In the data processing module 600, adjusting the current comprehensive test scenario data includes: determining a new cluster head node vehicle; then, the new cluster head node vehicle adjusts the current comprehensive test scene data according to the performance evaluation index of each test vehicle in the test fleet 100; the performance evaluation index includes at least a distance between each test vehicle and the vehicle 400 under test, a time delay between each test vehicle, and a load demand between each test vehicle.
Since the testing process is a dynamic process, real-time updates to the test vehicles in the test fleet 100 are required. In the actual testing process, with the change of the testing scene, the cluster head node vehicles need to be updated first, after the updating is completed, the new cluster head node vehicles can adjust the current testing scene data, and specifically, the relevant positions of the other testing vehicles in the current testing scene are adjusted according to the performance evaluation indexes of the other testing vehicles in the testing fleet 100. The performance evaluation index of each test vehicle specifically includes a distance between each test vehicle and the vehicle 400 to be tested, a time delay between each test vehicle, and a load demand between each test vehicle.
It should be noted that, during the lane change movement, the relative position between the test fleet 100 and the Vehicle 400 to be tested changes faster, so that the information interaction between the two is more frequent, and in order to ensure the timeliness of the information transmission and the rapid confirmation of the relative position during the movement, one or more of different wireless transmission modes such as Cellular Vehicle networking (Cellular Vehicle-to-electrical, C-V2X), short Distance Short Range Communication (DSRC), wireless sensing networks (ZigBee, bluetooth, etc.), wireless local area networks (WLAN, wiFi), etc., and a wireless transmission management system matched with the wireless transmission modes can be adopted; preferably, the wireless transmission mode can be executed according to communication protocols under standards such as a dedicated short-range communication protocol DSRC and an LTE-V/C-V2X, redundancy of communication signals can be achieved, the communication modules are the same as the international standard, the vehicle 400 to be tested does not need to specially designate a special signal module according to the test requirement, the test cost can be effectively saved, and the test applicability is enhanced. In addition, in the process of information interaction between the test fleet 100 and the vehicle 400, a functional test needs to be performed on the premise of ensuring safety, so when the criteria of the two are in conflict, the determination of the vehicle 400 should be used as the criterion, and the corresponding criterion in the test fleet 100 is corrected through information transmission.
In the embodiment, the updating of the cluster head node vehicles is the same as the selection principle of the cluster head node vehicles, and the strength grade of each test vehicle is obtained firstly; then determining a new cluster head node vehicle according to the robustness level; the robustness level is used to characterize the performance of each test vehicle in the test fleet 100, including the communication performance between cluster head node vehicles and the vehicle under test 400, and between cluster head node vehicles and the remaining test vehicles.
Specifically, the determination is made according to the robustness level of each test vehicle, the cloud platform 300 of the test system acquires data information of all test vehicles in the test fleet 100, where the data information includes time delay, packet loss rate, and the like, then the cloud platform 300 determines the communication performance between each vehicle and other vehicles through the data information, selects the test vehicle with the strongest communication performance as a new cluster head node vehicle, and then controls the running parameters of the other test vehicles according to the test requirements through the new cluster head node vehicle.
According to a third aspect, a software definition testing system provided in this embodiment is shown in fig. 1, and includes:
and the cloud platform 300 is configured to provide the test fleet 100 with test standard data corresponding to the test requirement information and initial test scenario data corresponding to the test requirement information.
A test fleet 100 comprising a plurality of test vehicles for: acquiring real-time road information and test requirement information of the tested vehicle 400; the road information includes movement information of a movable object on a road surface on which the vehicle 400 under test travels; the test demand information includes test items for testing an automatic driving function of the vehicle 400 under test, the automatic driving function including passing, lane changing, curve, or pedestrian danger avoidance; based on the test demand information and the real-time road information of the vehicle 400 to be tested, performing data processing on the initial test scene data to obtain comprehensive test scene data corresponding to the test items; adjusting the driving parameters of other test vehicles in the test fleet 100 according to the comprehensive test scene data, and communicating with other test vehicles in the test fleet 100 through a content-based routing algorithm; testing the tested vehicle 400 and forming a test result; and sending the test result to the cloud platform 300.
The cloud platform 300 is further configured to evaluate the test result according to the test standard data to form test reference information, and the test reference information is stored along with the user information of the user of the vehicle 400 to be tested. Specifically, the memory module of the cloud platform 300 further stores test standard data used for providing the test fleet 100 with information corresponding to the test requirement, and after the test result is sent to the cloud platform 300 through the cluster head node vehicle of the test fleet 100, the cloud platform 300 evaluates the test result and the test standard data corresponding to the test requirement information to form test reference information and stores the test reference information along with the user information of the tested vehicle 400 user.
In the test system, please refer to the method for testing a vehicle based on software definition in the above embodiment, which is not described herein in detail, for a specific implementation manner of performing an automatic driving function test on a vehicle 400 to be tested through the cloud platform 300 and the test fleet 100.
In addition, it should be noted that, the test vehicle fleet 100 in this embodiment may adopt various types of test vehicles, which may perform function tests in various forms, such as a common four-wheel, three-wheel, multi-wheel, or multi-wheel composite vehicle fleet, to have good compatibility, and also to improve the adaptability of this test method.
The present invention has been described in terms of specific examples, which are provided to aid understanding of the invention and are not intended to be limiting. For a person skilled in the art to which the invention pertains, several simple deductions, modifications or substitutions may be made according to the idea of the invention.

Claims (7)

1. A method of testing a vehicle based on software definition, comprising:
acquiring real-time road information, test requirement information of a tested vehicle and initial test scene data corresponding to the test requirement information; the road information includes movement information of a movable object on a traveling road surface of the vehicle under test; the test requirement information comprises test items for testing the automatic driving function of the tested vehicle, wherein the automatic driving function comprises overtaking, lane changing, curve or pedestrian risk avoiding;
based on the test demand information of the tested vehicle and the real-time road information, performing data processing on the initial test scene data to obtain comprehensive test scene data corresponding to the test project;
adjusting the running parameters of other test vehicles in the test fleet according to the comprehensive test scene data, and judging whether the current comprehensive test scene data meets the condition that the tested vehicle realizes the automatic driving function corresponding to the test project; if not, determining a new cluster head node vehicle; the new cluster head node vehicle adjusts the current comprehensive test scene data according to the performance evaluation indexes of all test vehicles in the test fleet so as to meet the condition that the tested vehicle realizes the automatic driving function corresponding to the test project; and communicating with other test vehicles in the test fleet via a content-based routing algorithm;
the performance evaluation index at least comprises the distance between each test vehicle and the tested vehicle, the time delay between each test vehicle and the load demand between each test vehicle.
2. The software-definition-based method for testing a vehicle according to claim 1, wherein the obtaining initial test scenario data corresponding to the test requirement information comprises: receiving initial test scene data which are sent by a cloud platform and correspond to the test requirement information; the initial test scene data is obtained by processing test requirement information submitted by a tested vehicle after the test requirement information is received by the cloud platform; the method further comprises the following steps:
and controlling the running parameters of other test vehicles in the test fleet according to the initial test scene data.
3. The method of testing vehicles based on software definition according to claim 1, wherein said determining new cluster head node vehicles comprises:
obtaining a strength level of each test vehicle;
determining a new cluster head node vehicle based on the robustness level; the robustness level is used for representing the performance of each test vehicle in the test fleet, and the performance comprises the communication performance between the cluster head node vehicle and the tested vehicle and between the cluster head node vehicle and the rest test vehicles.
4. A system for testing a vehicle based on software definition, comprising:
the system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring real-time road information, test requirement information of a tested vehicle and initial test scene data corresponding to the test requirement information; the road information includes movement information of a movable object on a traveling road surface of the vehicle under test; the test requirement information comprises test items for testing the automatic driving function of the tested vehicle, wherein the automatic driving function comprises overtaking, lane changing, curve or pedestrian risk avoiding;
the data processing module is used for carrying out data processing on the initial test scene data based on the test requirement information of the tested vehicle and the real-time road information to obtain comprehensive test scene data corresponding to the test items; the system is also used for judging whether the current comprehensive test scene data meets the condition that the tested vehicle realizes the automatic driving function corresponding to the test item; if not, determining a new cluster head node vehicle; the new cluster head node vehicle adjusts the current comprehensive test scene data according to the performance evaluation indexes of all test vehicles in the test fleet so as to enable the data to meet the conditions that the tested vehicles realize the automatic driving function corresponding to the test items, wherein the performance evaluation indexes at least comprise the distances between all the test vehicles and the tested vehicles, the time delays among all the test vehicles and the load requirements among all the test vehicles;
and the adjusting module is used for adjusting the running parameters of other test vehicles in the test fleet according to the comprehensive test scene data and communicating with the other test vehicles in the test fleet through a content-based routing algorithm.
5. The software-definition-based vehicle testing system of claim 4, wherein the obtaining module obtaining initial test scenario data corresponding to the test requirement information comprises: receiving initial test scene data corresponding to the test demand information sent by the cloud platform; and the initial test scene data is obtained by processing the test requirement information after the cloud platform receives the test requirement information submitted by the vehicle to be tested.
6. The software definition-based vehicle testing system of claim 4, further comprising:
and the control module is used for controlling the running parameters of other test vehicles in the test fleet according to the initial test scene data.
7. A software defined test system, comprising:
the cloud platform is used for providing test standard data corresponding to the test requirement information and initial test scene data corresponding to the test requirement information for the test fleet;
a test fleet comprising a plurality of test vehicles for:
acquiring real-time road information and test demand information of a tested vehicle; the road information includes movement information of a movable object on a traveling road surface of the vehicle under test; the test requirement information comprises test items for testing the automatic driving function of the tested vehicle, wherein the automatic driving function comprises overtaking, lane changing, curve or pedestrian risk avoiding;
based on the test demand information of the tested vehicle and the real-time road information, performing data processing on the initial test scene data to obtain comprehensive test scene data corresponding to the test project;
adjusting the running parameters of other test vehicles in the test fleet according to the comprehensive test scene data, and judging whether the current comprehensive test scene data meets the condition that the tested vehicle realizes the automatic driving function corresponding to the test project; if not, determining a new cluster head node vehicle; the new cluster head node vehicle adjusts the current comprehensive test scene data according to the performance evaluation index of each test vehicle in the test fleet so as to meet the condition that the tested vehicle realizes the automatic driving function corresponding to the test project; and communicating with other test vehicles in the test fleet via a content-based routing algorithm; the performance evaluation indexes at least comprise the distance between each test vehicle and the tested vehicle, the time delay between each test vehicle and the load demand between each test vehicle;
testing the tested vehicle and forming a test result;
sending the test result to the cloud platform;
the cloud platform is further used for evaluating the test result according to the test standard data to form test reference information, and the test reference information is stored along with the user information of the tested vehicle user.
CN202211462704.1A 2022-11-22 2022-11-22 Method and system for testing vehicle based on software definition and software definition testing system Active CN115597889B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211462704.1A CN115597889B (en) 2022-11-22 2022-11-22 Method and system for testing vehicle based on software definition and software definition testing system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211462704.1A CN115597889B (en) 2022-11-22 2022-11-22 Method and system for testing vehicle based on software definition and software definition testing system

Publications (2)

Publication Number Publication Date
CN115597889A CN115597889A (en) 2023-01-13
CN115597889B true CN115597889B (en) 2023-03-14

Family

ID=84853651

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211462704.1A Active CN115597889B (en) 2022-11-22 2022-11-22 Method and system for testing vehicle based on software definition and software definition testing system

Country Status (1)

Country Link
CN (1) CN115597889B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108829087A (en) * 2018-07-19 2018-11-16 山东省科学院自动化研究所 A kind of intelligent test system and test method of autonomous driving vehicle
CN111226268A (en) * 2017-05-02 2020-06-02 密歇根大学董事会 Simulated vehicular traffic for autonomous vehicles
CN111768612A (en) * 2020-07-10 2020-10-13 江苏智能交通及智能驾驶研究院 C-V2X-based vehicle formation driving strategy control method
CN115346367A (en) * 2022-07-26 2022-11-15 上海交通大学 Test system and test method of vehicle-road cooperative system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111226268A (en) * 2017-05-02 2020-06-02 密歇根大学董事会 Simulated vehicular traffic for autonomous vehicles
CN108829087A (en) * 2018-07-19 2018-11-16 山东省科学院自动化研究所 A kind of intelligent test system and test method of autonomous driving vehicle
CN111768612A (en) * 2020-07-10 2020-10-13 江苏智能交通及智能驾驶研究院 C-V2X-based vehicle formation driving strategy control method
CN115346367A (en) * 2022-07-26 2022-11-15 上海交通大学 Test system and test method of vehicle-road cooperative system

Also Published As

Publication number Publication date
CN115597889A (en) 2023-01-13

Similar Documents

Publication Publication Date Title
Gao et al. V2VR: reliable hybrid-network-oriented V2V data transmission and routing considering RSUs and connectivity probability
US11027751B2 (en) Reinforcement and model learning for vehicle operation
Chen et al. ASGR: An artificial spider-web-based geographic routing in heterogeneous vehicular networks
US11110941B2 (en) Centralized shared autonomous vehicle operational management
CN110435667A (en) System and method for controlling autonomous vehicle
CN103890824B (en) Method for operating a driver assistance system and method for processing vehicle environment data
CN107426694B (en) Fuzzy clustering method for vehicle-mounted self-organizing network
CN111845754B (en) Decision prediction method of automatic driving vehicle based on edge calculation and crowd-sourcing algorithm
US11584248B2 (en) Method of parking an autonomous driving vehicle for autonomous charging
CN105208616A (en) Road topology based adaptive multi-copy routing method in vehicular ad hoc network
CN103118440B (en) Construction method of dynamic mesh of vehicle-mounted ad hoc network
CN108650656B (en) Distributed city Internet of vehicles routing method based on intersection
CN106441324A (en) Navigation method, navigation apparatus, road condition information collection method and road condition information collection system
CN111225336A (en) Base station selection and switching method and system based on intelligent lamp pole
US20220196418A1 (en) Navigation Map Learning for Intelligent Hybrid-Electric Vehicle Planning
KR102256644B1 (en) Artificial intelligence traffic signal host server using BIM object model and control system comprising it and method of controlling traffic signal
CN115597889B (en) Method and system for testing vehicle based on software definition and software definition testing system
CN114237237A (en) Path optimization method for autonomous steering control of unmanned vehicle
US11899454B2 (en) Objective-based reasoning in autonomous vehicle decision-making
CN115963840A (en) Intelligent vehicle formation control method based on deep reinforcement learning under communication topology switching
CN110297486A (en) System and method for directing the user to vehicle
Khan et al. A game theory approach for smart traffic management
Lu et al. An anti-collision algorithm for self-organizing vehicular ad-hoc network using deep learning
CN118097989A (en) Multi-agent traffic area signal control method based on digital twin
Wadea et al. Enhancing VANET connectivity through utilizing autonomous vehicles

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