CN113377663B - Automatic driving simulation test method based on scene element removing characteristics - Google Patents
Automatic driving simulation test method based on scene element removing characteristics Download PDFInfo
- Publication number
- CN113377663B CN113377663B CN202110703885.1A CN202110703885A CN113377663B CN 113377663 B CN113377663 B CN 113377663B CN 202110703885 A CN202110703885 A CN 202110703885A CN 113377663 B CN113377663 B CN 113377663B
- Authority
- CN
- China
- Prior art keywords
- test platform
- test
- loop
- scene
- hardware
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
Abstract
The invention relates to an automatic driving simulation test method based on scene element removing characteristics. The method comprises the following steps: firstly, establishing an automatic driving automobile sensing system hardware-in-loop test platform, an execution system hardware-in-loop test platform and a vehicle-in-loop simulation test platform; secondly, determining scene elements to be tested according to the characteristics of the platform to be tested; thirdly, constructing a test scene according to the scene elements to be tested; analyzing the test result, and selectively removing the tested scene elements; inputting the residual scene elements into the vehicle-in-loop simulation test platform, and carrying out final test on the vehicle-in-loop simulation test platform; and sixthly, finally analyzing the performance of the tested automatic driving algorithm according to the test result. The invention can well combine pure software simulation, hardware-in-loop test of a sensing system, hardware-in-loop test of an execution system and vehicle-in-loop simulation test to form an automatic driving overall simulation test, and can effectively accelerate the overall test by a method for removing element characteristics.
Description
Technical Field
The invention belongs to the technical field of test evaluation of an automatic driving automobile, and particularly relates to an automatic driving simulation test method based on scene-removing element characteristics.
Background
With the development of the related technology of the automatic driving automobile, the realization of mass production of the automatic driving automobile has already possessed technical feasibility, and a lot of enterprises have promoted mass production vehicles possessing an automatic driving function. However, how to ensure the safety of the automatic driving automobile on the road is not uniformly known, and no mature regulation is provided for the safety verification process of the automatic driving automobile. In the test process aiming at automatic driving, simulation test occupies a very important position, and a plurality of organizations research related contents, but research is only carried out on a single simulation test platform, and a whole tool chain for simulation test is not run through.
Disclosure of Invention
The invention provides an automatic driving simulation test method based on scene element removing characteristics, which can well combine pure software simulation, hardware-in-loop test of a sensing system, hardware-in-loop test of an execution system and vehicle-in-loop simulation test to form an automatic driving overall simulation test flow and can effectively accelerate the overall test through the method of removing the element characteristics.
The technical scheme of the invention is described as follows by combining the attached drawings:
an automatic driving simulation test method based on scene element removing characteristics comprises the following steps:
firstly, establishing an automatic driving automobile sensing system hardware-in-loop test platform, an execution system hardware-in-loop test platform and a vehicle-in-loop simulation test platform;
determining scene elements to be tested according to the characteristics of the hardware-in-loop test platform and the hardware-in-loop test platform of the execution system of the automatic driving automobile sensing system to be tested;
thirdly, determining scene elements to be tested according to the second step, and constructing a test scene of the hardware-in-loop test platform of the perception system and the hardware-in-loop test platform of the execution system in a combined test mode;
analyzing the test results of the hardware-in-loop test platform and the hardware-in-loop test platform of the automatic driving automobile sensing system, and selectively removing the tested scene elements;
inputting a failure scene of hardware of the perception system in the vehicle-in-loop test platform and a test scene generated by the residual scene elements through a combined test method into the vehicle-in-loop simulation test platform, establishing a vehicle-in-loop simulation test platform test scene set, and performing final test on the vehicle-in-loop simulation test platform;
and step six, finally analyzing the performance of the tested automatic driving algorithm according to the test results of the hardware-in-loop test platform of the automatic driving automobile sensing system, the hardware-in-loop test platform of the execution system and the vehicle-in-loop simulation test platform.
The specific method of the first step is as follows:
establishing a hardware-in-loop test platform of the sensing system and a hardware-in-loop test platform of the execution system by analyzing the working characteristics and the used hardware equipment of the automatic driving automobile, wherein the hardware equipment comprises sensing system characteristics and execution system characteristics; the hardware-in-loop test platform of the sensing system and the hardware-in-loop test platform of the execution system comprise a millimeter wave radar sensing system-in-loop test platform, a camera sensing system-in-loop test platform, a V2X sensing system-in-loop test platform, an ultrasonic wave sensing system-in-loop test platform, a hardware-in-loop test platform of a brake system and a hardware-in-loop test platform of a steering system; the vehicle is integrally embedded into a test environment, and a vehicle-in-loop simulation test platform is built.
The specific method of the second step is as follows:
setting n-dimension of scene elements required to be tested for functions of the system to be tested, wherein x is the representation of each scene dimension, and the scene elements required to be tested are all the same
X＝[x1,x2,x3,…xn]
When all the seven test beds need to test all the types of elements, assuming that each scene element has M optional parameters, the number of the scenes which need to be tested is M;
M＝7×mn
with the increase of the dimension of the parameter to be tested, the phenomenon of dimension explosion is easy to occur to the number of the tested scenes; therefore, a test flow for removing elements is required to be established;
selecting the type of the scene element to be tested according to the characteristics of a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave sensing system in-loop test platform and a brake system hardware in-loop test platform, wherein the scene element to be tested of the millimeter wave radar sensing system in the loop test platform is [ x ] xk1,…xn-k1](ii) a The scene element of the camera perception system tested on the ring test platform is [ x ]k2,…xn-k2](ii) a The scene element tested by the V2X perception system on the ring test platform is [ x ]k3,…xn-k3](ii) a Sense of arrival of ultrasonic wavesThe scene element of the system tested on the ring test platform is xk4,…xn-k4](ii) a The scene element of the brake system hardware in the ring test platform is [ x ]k5,…xn-k5](ii) a The scene element of the steering system hardware in the ring test platform is [ x ]k6,…xn-k6](ii) a x is a test element and subscripts refer to different test elements.
The concrete method of the third step is as follows:
after determining scene elements to be tested of each testing platform, combining the testing scenes by an orthogonal table method in the combined test to generate, wherein the orthogonal table method combines values of all the elements to cover all the possible elements;
at this time, except the vehicle-in-loop simulation test platform, the number M' of the scenes needing to be tested of the other six platforms is
M’＝mn-k1+mn-k2+mn-k3+mn-k4+mn-k5+mn-k6
Wherein m is the number of the assumed optional parameters of each scene element, and for the convenience of calculation, the number of the assumed optional parameters of all the scene elements is the same; n-k1, n-k2, n-k3, n-k4, n-k5 and n-k6 refer to the number of test scene elements of six test platforms, and all of the test scene elements are less than or equal to n.
The concrete method of the fourth step is as follows:
inputting test scenes of the hardware-in-the-loop test platform of the sensing system and the hardware-in-the-loop test platform of the execution system, which need to be tested, into corresponding test platforms, and analyzing test results after all the test results are obtained through testing; firstly, determining dangerous scenes obtained by testing a millimeter wave radar sensing system in a ring test platform, a camera sensing system in a ring test platform, a V2X sensing system in a ring test platform and an ultrasonic wave sensing system in a ring test platform respectively, wherein the dangerous scenes are obtained by correcting errors to a certain extent through possible data fusion among sensors in the actual running process of an automatic driving automobile, so that the partially failed scenes need to be put into a failure scene set for subsequent in-ring vehicle tests; for the danger generated in the process of executing the system test, subsequent vehicle in-loop test verification is not needed, and the part of scenes can be directly judged as final danger; in addition, for scene elements in a non-vehicle running state class, if the scene elements are fully tested by a millimeter wave radar sensing system in a ring test platform, a camera sensing system in a ring test platform, a V2X sensing system in a ring test platform, an ultrasonic wave sensing system in a ring test platform, brake system hardware in a ring test platform and steering system hardware in a ring test platform, the scene elements in the vehicle in the ring test are removed; generating a failure scene set by the rest scene elements and the hardware-in-the-loop test platform of the sensing system to jointly form a test scene set of the vehicle-in-the-loop simulation test platform;
the number of the scenes needing to be tested in the vehicle in-loop test is M*
M*＝o+M**
In the formula, o is the number of scenes of a failure scene set generated by the sensing system on the ring platform; m**The number of test scenes generated for the remaining scene element parameters.
The concrete method of the step five is as follows:
generating the rest test scene elements in a combined mode through an orthogonal table method in the combined test, inputting the generated test scenes into a vehicle in-loop simulation test platform, and generating the number M of the test scenes**
M**＝mn-k7。
In the formula, n-k7 is the type of the scene element which needs to be tested when the vehicle is in the ring; and m is the number of optional parameters of each assumed scene element.
The concrete method of the sixth step is as follows:
according to the test results of the hardware-in-loop test platform of the sensing system, the hardware-in-loop test platform of the execution system and the vehicle-in-loop simulation test platform, determining the result weights of different test platforms by using factor analysis, and comprehensively obtaining the final score of the tested automatic driving algorithm, wherein the specific steps are as follows:
S＝a1*S1+a2*S2+a3*S3+a4*S4+a5*S5+a6*S6+a7*S7
in the formula, a1、a2、a3、a4、a5、a6、a7The method comprises the following steps of S weighting the scores of a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave class perception system in-loop test platform, a brake system hardware in-loop test platform, a steering system hardware in-loop test platform and a vehicle in-loop simulation test platform which are obtained by using a factor analysis method1、S2、S3、S4、S5、S6、S7The method is characterized by scoring test results of a tested automatic driving system in a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave radar sensing system in-loop test platform, a brake system hardware in-loop test platform, a steering system hardware in-loop test platform and a vehicle in-loop simulation test platform.
The invention has the beneficial effects that:
the invention can well combine the prior pure software simulation, hardware-in-loop test of the sensing system, hardware-in-loop test of the executing system and vehicle-in-loop simulation test to form an automatic driving overall simulation test process, and can effectively accelerate the overall test by a method for removing element characteristics.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a schematic diagram of a test sequence framework of various test platforms according to the present invention;
FIG. 2 is a schematic flow chart of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 and 2, an automatic driving simulation test method based on scene-removing element features includes the following steps:
firstly, establishing an automatic driving automobile sensing system hardware-in-loop test platform, an execution system hardware-in-loop test platform and a vehicle-in-loop simulation test platform;
establishing a hardware-in-loop test platform of the sensing system and a hardware-in-loop test platform of the execution system by analyzing the working characteristics and the used hardware equipment of the automatic driving automobile, wherein the hardware equipment comprises sensing system characteristics and execution system characteristics; the hardware-in-loop test platform of the sensing system and the hardware-in-loop test platform of the execution system comprise a millimeter wave radar sensing system-in-loop test platform, a camera sensing system-in-loop test platform, a V2X sensing system-in-loop test platform, an ultrasonic wave sensing system-in-loop test platform, a hardware-in-loop test platform of a brake system and a hardware-in-loop test platform of a steering system; the vehicle is integrally embedded into a test environment, and a vehicle-in-loop simulation test platform is built.
Determining scene elements to be tested according to the characteristics of the hardware-in-loop test platform and the hardware-in-loop test platform of the execution system of the automatic driving automobile sensing system to be tested;
setting n-dimension of scene elements required to be tested for functions of the system to be tested, wherein x is the representation of each scene dimension, and the scene elements required to be tested are all the same
X＝[x1,x2,x3,…xn]
When all the seven test beds need to test all the types of elements, assuming that each scene element has M optional parameters, the number of the scenes which need to be tested is M;
M＝7×mn
with the increase of the dimension of the parameter to be tested, the phenomenon of dimension explosion is easy to occur to the number of the tested scenes; therefore, a test flow for removing elements is required to be established;
selecting the type of the scene element to be tested according to the characteristics of a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave sensing system in-loop test platform and a brake system hardware in-loop test platform, wherein the scene element to be tested of the millimeter wave radar sensing system in the loop test platform is [ x ] xk1,…xn-k1](ii) a The scene element of the camera perception system tested on the ring test platform is [ x ]k2,…xn-k2](ii) a The scene element tested by the V2X perception system on the ring test platform is [ x ]k3,…xn-k3](ii) a The scene element of the ultrasonic wave analog-to-digital sensing system in the ring test platform is [ x ]k4,…xn-k4](ii) a The scene element of the brake system hardware in the ring test platform is [ x ]k5,…xn-k5](ii) a The scene element of the steering system hardware in the ring test platform is [ x ]k6,…xn-k6](ii) a x is a test element and subscripts refer to different test elements.
Thirdly, determining scene elements to be tested according to the second step, and constructing a test scene of the hardware-in-loop test platform of the perception system and the hardware-in-loop test platform of the execution system in a combined test mode;
after determining scene elements to be tested of each testing platform, combining the testing scenes by an orthogonal table method in the combined test to generate, wherein the orthogonal table method combines values of all the elements to cover all the possible elements;
at this time, except the vehicle-in-loop simulation test platform, the number M' of the scenes needing to be tested of the other six platforms is
M’＝mn-k1+mn-k2+mn-k3+mn-k4+mn-k5+mn-k6
Wherein m is the number of the assumed optional parameters of each scene element, and for the convenience of calculation, the number of the assumed optional parameters of all the scene elements is the same; n-k1, n-k2, n-k3, n-k4, n-k5 and n-k6 refer to the number of test scene elements of six test platforms, and all of the test scene elements are less than or equal to n.
Analyzing the test results of the hardware-in-loop test platform and the hardware-in-loop test platform of the automatic driving automobile sensing system, and selectively removing the tested scene elements;
inputting test scenes of the hardware-in-the-loop test platform of the sensing system and the hardware-in-the-loop test platform of the execution system, which need to be tested, into corresponding test platforms, and analyzing test results after all the test results are obtained through testing; firstly, determining dangerous scenes obtained by testing a millimeter wave radar sensing system in a ring test platform, a camera sensing system in a ring test platform, a V2X sensing system in a ring test platform and an ultrasonic wave sensing system in a ring test platform respectively, wherein the dangerous scenes are obtained by correcting errors to a certain extent through possible data fusion among sensors in the actual running process of an automatic driving automobile, so that the partially failed scenes need to be put into a failure scene set for subsequent in-ring vehicle tests; for the danger generated in the process of executing the system test, subsequent vehicle in-loop test verification is not needed, and the part of scenes can be directly judged as final danger; in addition, for scene elements in a non-vehicle running state class, if the scene elements are fully tested by a millimeter wave radar sensing system in a ring test platform, a camera sensing system in a ring test platform, a V2X sensing system in a ring test platform, an ultrasonic wave sensing system in a ring test platform, brake system hardware in a ring test platform and steering system hardware in a ring test platform, the scene elements in the vehicle in the ring test are removed; generating a failure scene set by the rest scene elements and the hardware-in-the-loop test platform of the sensing system to jointly form a test scene set of the vehicle-in-the-loop simulation test platform;
the number of the scenes needing to be tested in the vehicle in-loop test is M*
M*＝o+M**
In the formula, o is the number of scenes of a failure scene set generated by the sensing system on the ring platform; m**The number of test scenes generated for the remaining scene element parameters.
Inputting a failure scene of hardware of the perception system in the vehicle-in-loop test platform and a test scene generated by the residual scene elements through a combined test method into the vehicle-in-loop simulation test platform, establishing a vehicle-in-loop simulation test platform test scene set, and performing final test on the vehicle-in-loop simulation test platform;
generating the rest test scene elements in a combined mode through an orthogonal table method in the combined test, inputting the generated test scenes into a vehicle in-loop simulation test platform, and generating the number M of the test scenes**
M**＝mn-k7。
In the formula, n-k7 is the type of the scene element which needs to be tested when the vehicle is in the ring; and m is the number of optional parameters of each assumed scene element.
And step six, finally analyzing the performance of the tested automatic driving algorithm according to the test results of the hardware-in-loop test platform of the automatic driving automobile sensing system, the hardware-in-loop test platform of the execution system and the vehicle-in-loop simulation test platform.
According to the test results of the hardware-in-loop test platform of the sensing system, the hardware-in-loop test platform of the execution system and the vehicle-in-loop simulation test platform, determining the result weights of different test platforms by using factor analysis, and comprehensively obtaining the final score of the tested automatic driving algorithm, wherein the specific steps are as follows:
S＝a1*S1+a2*S2+a3*S3+a4*S4+a5*S5+a6*S6+a7*S7
in the formula, a1、a2、a3、a4、a5、a6、a7The method is characterized in that a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a millimeter wave radar sensor and a sensor are obtained by using a factor analysis method,Scoring weights of a V2X perception system in-loop test platform, an ultrasonic wave class arrival perception system in-loop test platform, a brake system hardware in-loop test platform, a steering system hardware in-loop test platform and a vehicle in-loop simulation test platform, S1、S2、S3、S4、S5、S6、S7The method is characterized by scoring test results of a tested automatic driving system in a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave radar sensing system in-loop test platform, a brake system hardware in-loop test platform, a steering system hardware in-loop test platform and a vehicle in-loop simulation test platform.
Examples
And selecting a curve following function of the expressway for testing, and selecting a tested logic scene as an emergency deceleration of a vehicle in front of the curve. According to the layering characteristics of six layers of scene elements, lane information, road infrastructure, temporary change of roads and road infrastructure, dynamic traffic participants, weather and digital information, and considering the sensor used by the automatic driving function and the effect required to be achieved, the selected test scene elements comprise: lane information, namely lane line color, lane line type, lane line fouling condition, lane width, lane curvature, road surface adhesion coefficient and local water stain position; road infrastructure classes, i.e. traffic signs and centre strip types; temporary changes of roads and road infrastructures, i.e. construction facilities; the dynamic traffic participant type, namely the RCS value, the position and the motion state of the target vehicle of the type of the target vehicle; the weather types are illumination intensity, illumination angle, lateral wind intensity and lateral wind angle; the digital information category includes vehicle-mounted V2X information, V2X signal interference frequency and V2X signal interference strength.
For 21 types of scene elements, optional parameters of partial discrete data are shown in table 1.
And performing self-defined dispersion on other continuous scenes according to the test requirements. At this time, it is assumed that the scene elements of the continuous class are all discretely acquired with 10 optional variables, and the number of scenes to be tested at this time is
M＝7*2*5*4*3*4*4*1015＝1.344*1019
It can be seen that the number of scenes to be tested is extremely large, and actual test requirements are difficult to meet.
The method provided by the invention is used for processing the elements of the test scene, and the elements can be obtained by analyzing the characteristics of the test platform, and the elements to be tested of several hardware-in-the-loop test platforms are as follows: testing elements, namely the RCS value of a target vehicle, the position of the target vehicle and the motion state of the target vehicle, of a millimeter wave radar sensing system on an annular testing platform; the camera perception system tests elements on the ring test platform, namely lane line color, lane line type, lane line fouling condition, lane width, lane curvature, traffic signs, middle belt type, construction facilities, target vehicle type, target vehicle position, target vehicle motion state, illumination intensity and illumination angle; the V2X perception system tests elements on the ring test platform, namely vehicle-mounted V2X information, V2X signal interference frequency, V2X signal interference strength, target vehicle position and target vehicle motion state; testing elements, namely the position of a target vehicle and the motion state of the target vehicle, of the ultrasonic radar sensing system on an annular testing platform; testing elements, namely lane width, lane curvature, road surface adhesion coefficient, local water stain position, target vehicle motion state, lateral wind strength and lateral wind angle, of the braking system hardware on the ring test platform; the hardware of the steering system tests the elements, namely lane width, lane curvature, road adhesion coefficient, local water stain position, target vehicle motion state, lateral wind strength and lateral wind angle, on the ring test platform.
Constructing test scenes of a hardware-in-loop test platform by using an orthogonal table method in the combined test, wherein the test scenes are respectively as follows:
the test quantity of the millimeter wave radar sensing system on the ring test platform is as follows:
Mmillimeter wave radar＝103
The number of the camera perception system tests on a ring test platform is as follows:
Mcamera with a camera module＝2*5*4*3*4*4*107＝1.92*1010
The number of tests of the V2X perception system on the ring test platform is as follows:
MV2X＝105
the test quantity of the ultrasonic radar sensing system on the ring test platform is as follows:
Multrasonic wave analog radar＝102
The test quantity of the hardware of the brake system on a ring test platform is as follows:
Mbrake system＝108
The test quantity of the hardware of the steering system on the ring test platform is as follows:
Msteering system＝108
Summarizing the dangerous scene sets of the hardware of the perception system on the ring test platform, wherein the approximate scene number is
MSensing system＝106
Because sensing system signals such as RCS values of target vehicles, lane line colors, lane line types, lane line fouling conditions, traffic signs, intermediate belt types, construction facilities, target vehicle types, illumination intensity, illumination angles, vehicle-mounted V2X information, V2X signal interference frequencies and V2X signal interference intensities are fully tested and failure parameters of corresponding sensors are known, the final vehicle-in-loop test can eliminate the part of scene elements and only pay attention to dangerous scenes. The remaining scene elements comprise lane width, lane curvature, road surface adhesion coefficient, local water stain position, target vehicle motion state, lateral wind strength and lateral wind angle, the scene elements related to the vehicle motion state need to be subjected to combined test to generate a test scene, and the test scene and a dangerous scene generated by a sensing system test are input into a vehicle in-loop simulation test platform together, and the number of scenes generated by the remaining scene elements is
MRemainder of＝108
Thus, the number of test scenarios as a whole
M is whole as MMillimeter wave radar+MCamera with a camera module+MV2X+MUltrasonic wave analog radar+MBrake system+MSteering system+MSensing system+MRemainder of＝103+1.92*1010+105+102+108+108+106+108≈1.95*1010
Much less than 1.344 x 10 scenes required for direct testing19。
In conclusion, the invention can effectively accelerate the whole test process.
Although the preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, the scope of the present invention is not limited to the specific details of the above embodiments, and any person skilled in the art can substitute or change the technical solution of the present invention and its inventive concept within the technical scope of the present invention, and these simple modifications belong to the scope of the present invention.
It should be noted that the various technical features described in the above embodiments can be combined in any suitable manner without contradiction, and the invention is not described in any way for the possible combinations in order to avoid unnecessary repetition.
In addition, any combination of the various embodiments of the present invention is also possible, and the same should be considered as the disclosure of the present invention as long as it does not depart from the spirit of the present invention.
Claims (6)
1. An automatic driving simulation test method based on scene element removing characteristics is characterized by comprising the following steps:
firstly, establishing an automatic driving automobile sensing system hardware-in-loop test platform, an execution system hardware-in-loop test platform and a vehicle-in-loop simulation test platform;
determining scene elements to be tested according to the characteristics of the hardware-in-loop test platform and the hardware-in-loop test platform of the execution system of the automatic driving automobile sensing system to be tested;
thirdly, determining scene elements to be tested according to the second step, and constructing a test scene of the hardware-in-loop test platform of the perception system and the hardware-in-loop test platform of the execution system in a combined test mode;
analyzing the test results of the hardware-in-loop test platform and the hardware-in-loop test platform of the automatic driving automobile sensing system, and selectively removing the tested scene elements;
inputting a failure scene of hardware of the perception system in the vehicle-in-loop test platform and a test scene generated by the residual scene elements through a combined test method into the vehicle-in-loop simulation test platform, establishing a vehicle-in-loop simulation test platform test scene set, and performing final test on the vehicle-in-loop simulation test platform;
sixthly, performing final analysis on the performance of the tested automatic driving algorithm according to the test results of the hardware-in-loop test platform of the automatic driving automobile sensing system, the hardware-in-loop test platform of the execution system and the vehicle-in-loop simulation test platform;
the specific method of the second step is as follows:
setting n-dimension of scene elements required to be tested for functions of the system to be tested, wherein x is the representation of each scene dimension, and the scene elements required to be tested are all the same
X＝[x1,x2,x3,…xn]
When all the seven test beds need to test all the types of elements, assuming that each scene element has M optional parameters, the number of the scenes which need to be tested is M;
M＝7×mn
with the increase of the dimension of the parameter to be tested, the phenomenon of dimension explosion is easy to occur to the number of the tested scenes; therefore, a test flow for removing elements is required to be established;
according to the characteristics of a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave sensing system in-loop test platform and a brake system hardware in-loop test platform,selecting the types of the scene elements to be tested, wherein the scene element tested by the millimeter wave radar sensing system on the ring test platform is [ x ]k1,…xn-k1](ii) a The scene element of the camera perception system tested on the ring test platform is [ x ]k2,…xn-k2](ii) a The scene element tested by the V2X perception system on the ring test platform is [ x ]k3,…xn-k3](ii) a The scene element of the ultrasonic wave analog-to-digital sensing system in the ring test platform is [ x ]k4,…xn-k4](ii) a The scene element of the brake system hardware in the ring test platform is [ x ]k5,…xn-k5](ii) a The scene element of the steering system hardware in the ring test platform is [ x ]k6,…xn-k6](ii) a x is a test element and subscripts refer to different test elements.
2. The automatic driving simulation test method based on scene-removing element characteristics according to claim 1, wherein the specific method of the first step is as follows:
establishing a hardware-in-loop test platform of the sensing system and a hardware-in-loop test platform of the execution system by analyzing the working characteristics and the used hardware equipment of the automatic driving automobile, wherein the hardware equipment comprises sensing system characteristics and execution system characteristics; the hardware-in-loop test platform of the sensing system and the hardware-in-loop test platform of the execution system comprise a millimeter wave radar sensing system-in-loop test platform, a camera sensing system-in-loop test platform, a V2X sensing system-in-loop test platform, an ultrasonic wave sensing system-in-loop test platform, a hardware-in-loop test platform of a brake system and a hardware-in-loop test platform of a steering system; the vehicle is integrally embedded into a test environment, and a vehicle-in-loop simulation test platform is built.
3. The automatic driving simulation test method based on scene-removing element characteristics according to claim 1, wherein the specific method of the third step is as follows:
after determining scene elements to be tested of each testing platform, combining the testing scenes by an orthogonal table method in the combined test to generate, wherein the orthogonal table method combines values of all the elements to cover all the possible elements;
at this time, except the vehicle-in-loop simulation test platform, the number M' of the scenes needing to be tested of the other six platforms is
M’＝mn-k1+mn-k2+mn-k3+mn-k4+mn-k5+mn-k6
Wherein m is the number of the assumed optional parameters of each scene element, and for the convenience of calculation, the number of the assumed optional parameters of all the scene elements is the same; n-k1, n-k2, n-k3, n-k4, n-k5 and n-k6 refer to the number of test scene elements of six test platforms, and all of the test scene elements are less than or equal to n.
4. The automatic driving simulation test method based on scene-removing element characteristics according to claim 1, wherein the specific method of the fourth step is as follows:
inputting test scenes of the hardware-in-the-loop test platform of the sensing system and the hardware-in-the-loop test platform of the execution system, which need to be tested, into corresponding test platforms, and analyzing test results after all the test results are obtained through testing; firstly, determining dangerous scenes obtained by testing a millimeter wave radar sensing system in a ring test platform, a camera sensing system in a ring test platform, a V2X sensing system in a ring test platform and an ultrasonic wave sensing system in a ring test platform respectively, wherein the dangerous scenes are obtained by correcting errors to a certain extent through possible data fusion among sensors in the actual running process of an automatic driving automobile, so that the partially failed scenes need to be put into a failure scene set for subsequent in-ring vehicle tests; for the danger generated in the process of executing the system test, subsequent vehicle in-loop test verification is not needed, and the danger scene generated in the process of executing the system test can be directly judged as the final danger; in addition, for scene elements in a non-vehicle running state class, if the scene elements are fully tested by a millimeter wave radar sensing system in a ring test platform, a camera sensing system in a ring test platform, a V2X sensing system in a ring test platform, an ultrasonic wave sensing system in a ring test platform, brake system hardware in a ring test platform and steering system hardware in a ring test platform, the scene elements in the vehicle in the ring test are removed; generating a failure scene set by the rest scene elements and the hardware-in-the-loop test platform of the sensing system to jointly form a test scene set of the vehicle-in-the-loop simulation test platform;
the number of the scenes needing to be tested in the vehicle in-loop test is M*
M*＝o+M**
In the formula, o is the number of scenes of a failure scene set generated by the sensing system on the ring platform; m**The number of test scenes generated for the remaining scene element parameters.
5. The automatic driving simulation test method based on scene-removing element characteristics according to claim 1, wherein the concrete method of the fifth step is as follows:
generating the rest test scene elements in a combined mode through an orthogonal table method in the combined test, inputting the generated test scenes into a vehicle in-loop simulation test platform, and generating the number M of the test scenes**
M**＝mn-k7；
In the formula, n-k7 is the type of the scene element which needs to be tested when the vehicle is in the ring; and m is the number of optional parameters of each assumed scene element.
6. The automatic driving simulation test method based on scene-removing element characteristics according to claim 1, wherein the specific method of the sixth step is as follows:
according to the test results of the hardware-in-loop test platform of the sensing system, the hardware-in-loop test platform of the execution system and the vehicle-in-loop simulation test platform, determining the result weights of different test platforms by using factor analysis, and comprehensively obtaining the final score of the tested automatic driving algorithm, wherein the specific steps are as follows:
S＝a1*S1+a2*S2+a3*S3+a4*S4+a5*S5+a6*S6+a7*S7
in the formula, a1、a2、a3、a4、a5、a6、a7The method comprises the following steps of S weighting the scores of a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave class perception system in-loop test platform, a brake system hardware in-loop test platform, a steering system hardware in-loop test platform and a vehicle in-loop simulation test platform which are obtained by using a factor analysis method1、S2、S3、S4、S5、S6、S7The method is characterized by scoring test results of a tested automatic driving system in a millimeter wave radar sensing system in-loop test platform, a camera sensing system in-loop test platform, a V2X sensing system in-loop test platform, an ultrasonic wave radar sensing system in-loop test platform, a brake system hardware in-loop test platform, a steering system hardware in-loop test platform and a vehicle in-loop simulation test platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110703885.1A CN113377663B (en) | 2021-06-24 | 2021-06-24 | Automatic driving simulation test method based on scene element removing characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110703885.1A CN113377663B (en) | 2021-06-24 | 2021-06-24 | Automatic driving simulation test method based on scene element removing characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113377663A CN113377663A (en) | 2021-09-10 |
| CN113377663B true CN113377663B (en) | 2021-12-10 |
Family
ID=77578911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110703885.1A Active CN113377663B (en) | 2021-06-24 | 2021-06-24 | Automatic driving simulation test method based on scene element removing characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113377663B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117056746A (en) * | 2023-10-11 | 2023-11-14 | 长春汽车工业高等专科学校 | Big data-based automobile test platform and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106873397A (en) * | 2017-01-23 | 2017-06-20 | 同济大学 | Intelligent network joins automobile " hardware in loop " accelerated loading emulation test system |
| CN111841012A (en) * | 2020-06-23 | 2020-10-30 | 北京航空航天大学 | Automatic driving simulation system and test resource library construction method thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8600556B2 (en) * | 2009-06-22 | 2013-12-03 | Johnson Controls Technology Company | Smart building manager |
-
2021
- 2021-06-24 CN CN202110703885.1A patent/CN113377663B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106873397A (en) * | 2017-01-23 | 2017-06-20 | 同济大学 | Intelligent network joins automobile " hardware in loop " accelerated loading emulation test system |
| CN111841012A (en) * | 2020-06-23 | 2020-10-30 | 北京航空航天大学 | Automatic driving simulation system and test resource library construction method thereof |
Non-Patent Citations (2)
| Title |
|---|
| X在环在汽车整车开发中的应用;牛文旭等;《机电一体化》;20150131;第49-52页 * |
| 基于场景的自动驾驶汽车虚拟测试研究进展;朱冰等;《中国公路学报》;20190630;第32卷(第6期);第1-19页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113377663A (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210197851A1 (en) | Method for building virtual scenario library for autonomous vehicle | |
| CN111009153B (en) | Training method, device and equipment of trajectory prediction model | |
| CN111795832B (en) | Intelligent driving vehicle testing method, device and equipment | |
| CN109884916A (en) | A kind of automatic Pilot Simulation Evaluation method and device | |
| WO2020079698A1 (en) | Adas systems functionality testing | |
| CN112116031B (en) | Target fusion method, system, vehicle and storage medium based on road side equipment | |
| Nentwig et al. | Hardware-in-the-loop testing of computer vision based driver assistance systems | |
| CN112613169A (en) | Expected function safety analysis method for misoperation of automatic driving vehicle | |
| CN113377663B (en) | Automatic driving simulation test method based on scene element removing characteristics | |
| CN113515105A (en) | Platform, method and storage medium for vehicle expected function safety simulation test | |
| CN112685289A (en) | Scene generation method, and scene-based model in-loop test method and system | |
| CN114354219A (en) | Test method and device for automatic driving vehicle | |
| CN113779705A (en) | Intelligent grade assessment method and system for automatic driving automobile | |
| US20200384989A1 (en) | Method for the improved detection of objects by a driver assistance system | |
| King et al. | A taxonomy and survey on validation approaches for automated driving systems | |
| CN112987711B (en) | Optimization method of automatic driving regulation algorithm and simulation testing device | |
| KR20230159308A (en) | Method, system and computer program product for calibrating and validating an advanced driver assistance system (adas) and/or an automated driving system (ads) | |
| CN111767630A (en) | Virtual simulation test method and device applied to intelligent driving | |
| CN112631151B (en) | Simulation test method and device | |
| US20210182707A1 (en) | Method for Testing and Testing Device | |
| CN111932829B (en) | Fatigue driving prevention facility utility testing method and system | |
| CN112396235B (en) | Traffic accident occurrence time prediction modeling method based on eyeball motion tracking | |
| CN115309074A (en) | Automatic driving simulation test method and device, simulation equipment and storage medium | |
| CN115292816A (en) | Automatic driving test method, device, equipment and storage medium | |
| CN110954341B (en) | Test method and system for intelligent networking automobile test scene |
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 |