CN117763818A - Vehicle-mounted laser radar simulation model evaluation method and device and computer equipment - Google Patents
Vehicle-mounted laser radar simulation model evaluation method and device and computer equipment Download PDFInfo
- Publication number
- CN117763818A CN117763818A CN202311694725.0A CN202311694725A CN117763818A CN 117763818 A CN117763818 A CN 117763818A CN 202311694725 A CN202311694725 A CN 202311694725A CN 117763818 A CN117763818 A CN 117763818A
- Authority
- CN
- China
- Prior art keywords
- laser radar
- simulation model
- modeling
- radar simulation
- determining
- 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.)
- Pending
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 454
- 238000011156 evaluation Methods 0.000 title claims abstract description 146
- 238000012360 testing method Methods 0.000 claims abstract description 150
- 238000000034 method Methods 0.000 claims abstract description 33
- 230000015654 memory Effects 0.000 claims description 27
- 238000001514 detection method Methods 0.000 claims description 17
- 238000013461 design Methods 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000012795 verification Methods 0.000 abstract description 10
- 238000012827 research and development Methods 0.000 abstract description 4
- 238000005259 measurement Methods 0.000 description 25
- 238000010586 diagram Methods 0.000 description 9
- 238000011161 development Methods 0.000 description 5
- 238000013459 approach Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Optical Radar Systems And Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of intelligent automobile simulation test, and discloses a vehicle-mounted laser radar simulation model evaluation method, a device and computer equipment, wherein the method comprises the following steps: determining a modeling grade of a laser radar simulation model; acquiring an application scene of a laser radar simulation model; carrying out importance assessment on the laser radar simulation model to obtain an importance assessment result of the laser radar simulation model; determining working conditions required by evaluating the laser radar simulation model based on the importance evaluation result; collecting first test data of a laser radar simulation model under working conditions and second test data of a real laser radar under working conditions; and performing simulation degree evaluation on the laser radar simulation model based on the first test data and the second test data to obtain a simulation degree evaluation result of the laser radar simulation model. The intelligent driving system simulation method considers different requirements of the intelligent automobile on the simulation degree of the laser radar simulation model in the simulation stage, and meets the current situation of the research and development verification requirements of the intelligent driving system.
Description
Technical Field
The invention relates to the technical field of intelligent automobile simulation tests, in particular to a vehicle-mounted laser radar simulation model evaluation method, a vehicle-mounted laser radar simulation model evaluation device and computer equipment.
Background
The laser radar obtains information such as relative distance, speed, angle and the like of a target by sending a laser beam to the target, and is an important component in an intelligent automobile sensing system for detecting the automobile running environment and supporting a decision control system to finish dynamic and static driving tasks. Meanwhile, the laser radar sensing technology is still in a development and exploration stage, and in order to reduce the test and verification cost, the intelligent driving system needs to be fully verified before the driver.
Currently, related groups in the industry have issued laser radar hardware evaluation standards for the industry focus laser radar hardware evaluation method. However, in the aspect of evaluating the laser radar simulation model, the existing evaluation method of the laser radar simulation model does not consider different requirements of the intelligent automobile on the simulation degree of the laser radar simulation model in the simulation stage, and the actual application scene of the laser radar sensor is not covered in the evaluation index dimension.
Disclosure of Invention
In view of the above, the invention provides a vehicle-mounted laser radar simulation model evaluation method, a device and computer equipment, so as to solve the problems that the existing laser radar simulation model evaluation method does not consider different requirements of intelligent automobiles on the simulation degree of the laser radar simulation model in a simulation stage, does not cover the actual application scene of a laser radar sensor in the dimension of an evaluation index, and is difficult to meet the current situation of the research, development and verification requirements of an intelligent driving system.
In a first aspect, the present invention provides a vehicle-mounted lidar simulation model evaluation method, the method comprising:
determining a modeling grade of the laser radar simulation model based on modeling information of the laser radar simulation model;
acquiring an application scene of the laser radar simulation model based on the intelligent automobile design operation range;
based on the application scene and modeling level of the laser radar simulation model, carrying out importance assessment on the laser radar simulation model to obtain an importance assessment result of the laser radar simulation model;
determining working conditions required by evaluating the laser radar simulation model based on an importance evaluation result of the laser radar simulation model;
collecting first test data of the laser radar simulation model under the working condition and second test data of the real laser radar under the working condition;
and performing simulation evaluation on the laser radar simulation model based on the first test data and the second test data to obtain a simulation evaluation result of the laser radar simulation model.
According to the embodiment of the application method, the importance evaluation is carried out on the laser radar simulation model based on the application scene and the modeling level of the laser radar simulation model, the evaluation dimension covers the actual application scene of the laser radar sensor, different test working conditions are selected according to different requirements of the intelligent automobile on the simulation degree of the laser radar simulation model in the simulation stage, the simulation degree evaluation of the laser radar simulation model is carried out, and the current situation of the research, development and verification requirements of the intelligent driving system is met.
In an alternative embodiment, the determining the modeling level of the lidar simulation model based on the modeling information of the lidar simulation model includes:
and determining the modeling grade of the laser radar simulation model based on the modeling mode and the model output of the laser radar simulation model.
In an alternative embodiment, the determining the modeling level of the lidar simulation model based on the modeling manner and the model output of the lidar simulation model includes:
determining the modeling level of the laser radar simulation model as high-level modeling under the condition that the modeling mode of the laser radar simulation model is Maxwell equation set and the model output is primary data;
determining that the modeling level of the laser radar simulation model is medium modeling under the condition that the modeling mode of the laser radar simulation model is a black box model and the model output is a detection list;
and determining the modeling level of the laser radar simulation model as low-level modeling under the condition that the modeling mode of the laser radar simulation model is a true value list and the model output is a target list.
In an optional implementation manner, the performing importance assessment on the lidar simulation model based on the application scenario and the modeling level of the lidar simulation model to obtain an importance assessment result of the lidar simulation model includes:
Determining the influence degree of the application scene of the laser radar simulation model on a simulation test result according to the application scene of the laser radar simulation model;
inputting the modeling grade of the laser radar simulation model and the influence degree of the application scene of the laser radar simulation model on the simulation test result into an importance evaluation matrix to obtain the importance evaluation result of the laser radar simulation model.
According to the embodiment of the application, importance evaluation is carried out on the laser radar simulation model, the evaluation dimension covers the actual application scene of the laser radar sensor, and the current situation of research and development verification requirements of the intelligent driving system is met by considering different requirements of the intelligent automobile on the simulation degree of the laser radar simulation model in the simulation stage.
In an alternative embodiment, the determining, based on the importance evaluation result of the lidar simulation model, the working conditions required for evaluating the lidar simulation model includes:
under the condition that the importance assessment result of the laser radar simulation model is high in simulation degree, determining working conditions required by assessing the laser radar simulation model to be laser radar performance dimension working conditions, point cloud quality dimension working conditions and application scene performance dimension working conditions;
Under the condition that the importance assessment result of the laser radar simulation model is a requirement of intermediate degree of reality, determining a working condition required by assessing the laser radar simulation model as a laser radar performance dimension working condition;
and under the condition that the importance evaluation result of the laser radar simulation model is that the simulation degree is low, determining that the laser radar simulation model does not need to be subjected to simulation degree evaluation.
According to the embodiment of the application, different test working conditions are selected according to different requirements of the laser radar model on the simulation degree, the simulation degree of the laser radar simulation model is evaluated, and the current situation of research and development verification requirements of an intelligent driving system is met.
In an optional implementation manner, the performing the simulation evaluation on the lidar simulation model based on the first test data and the second test data to obtain a simulation evaluation result of the lidar simulation model includes:
determining a deviation value between the first test data and the second test data;
and under the condition that the deviation value is in a preset range, determining that the simulation degree evaluation result of the laser radar simulation model is good in simulation degree.
In an alternative embodiment, the method further comprises:
And under the condition that the deviation value is not in a preset range, determining that the simulation evaluation result of the laser radar simulation model is not up to the expected value.
In a second aspect, the present invention provides an on-vehicle lidar simulation model evaluation device, the device comprising:
the modeling grade determining module is used for determining the modeling grade of the laser radar simulation model based on the modeling information of the laser radar simulation model;
the application scene acquisition module is used for acquiring an application scene of the laser radar simulation model based on the intelligent automobile design operation range;
the importance evaluation result acquisition module is used for carrying out importance evaluation on the laser radar simulation model based on the application scene and the modeling level of the laser radar simulation model to obtain an importance evaluation result of the laser radar simulation model;
the working condition determining module is used for determining working conditions required by evaluating the laser radar simulation model based on the importance evaluation result of the laser radar simulation model;
the test data acquisition module is used for acquiring first test data of the laser radar simulation model under the working condition and second test data of the real laser radar under the working condition;
And the simulation degree evaluation result acquisition module is used for carrying out simulation degree evaluation on the laser radar simulation model based on the first test data and the second test data to obtain a simulation degree evaluation result of the laser radar simulation model.
In a third aspect, the present invention provides a computer device comprising: the system comprises a memory and a processor, wherein the memory and the processor are in communication connection, the memory stores computer instructions, and the processor executes the computer instructions so as to execute the vehicle-mounted laser radar simulation model evaluation method according to the first aspect or any implementation mode corresponding to the first aspect.
In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method for evaluating a vehicle-mounted lidar simulation model of the first aspect or any of the embodiments corresponding thereto.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for evaluating a simulation model of an in-vehicle lidar according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method for evaluating a simulation model of an in-vehicle lidar according to an embodiment of the present invention;
FIG. 3 is a flow chart of yet another vehicle-mounted lidar simulation model evaluation method according to an embodiment of the present invention;
FIG. 4 is a schematic illustration of an fidelity requirement in accordance with an embodiment of the present invention;
FIG. 5 is a schematic diagram of a detection accuracy evaluation condition in a lidar performance dimension according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a blind zone evaluation operating mode in a lidar performance dimension according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of an angular accuracy evaluation regime in the lidar performance dimension according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of an angular resolution evaluation regime in the lidar performance dimension according to an embodiment of the present invention;
FIG. 9 is a schematic view of view angle assessment at a lidar performance dimension according to an embodiment of the present invention;
FIG. 10 is a schematic diagram of a vehicle distance accuracy evaluation working condition and an European clustering result evaluation working condition under a point cloud quality dimension according to an embodiment of the invention;
FIG. 11 is a schematic diagram of a point cloud result evaluation condition under an application scenario performance dimension according to an embodiment of the present invention;
FIG. 12 is a schematic illustration of a laser radar data acquisition Sudoku selection according to an embodiment of the present invention;
FIG. 13 is a block diagram of a vehicle-mounted lidar simulation model evaluation apparatus according to an embodiment of the present invention;
fig. 14 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The laser radar obtains information such as relative distance, speed, angle and the like of a target by sending a laser beam to the target, and is an important component in an intelligent automobile sensing system for detecting the automobile running environment and supporting a decision control system to finish dynamic and static driving tasks. Meanwhile, the laser radar sensing technology is still in a development and exploration stage, and in order to reduce the test and verification cost, the intelligent driving system needs to be fully verified before the driver.
The industry focus laser radar hardware evaluation method has related groups issued laser radar hardware evaluation standards in the industry. However, in the aspect of evaluating the laser radar simulation model, in the related art, the method for evaluating the laser radar simulation model does not consider different requirements of the intelligent automobile on the simulation degree of the laser radar model in the simulation stage, and the actual application scene of the laser radar sensor is not covered in the evaluation index dimension.
In order to solve the problems that the actual application scene of the laser radar sensor is not covered in the evaluation index dimension and the current situation of the research and development verification requirements of the intelligent driving system is difficult to meet in the evaluation method of the laser radar simulation model, which does not consider the different requirements of the intelligent automobile on the simulation degree of the laser radar simulation model in the simulation stage, the embodiment of the invention provides the vehicle-mounted laser radar simulation model evaluation method, the vehicle-mounted laser radar simulation model evaluation device, the electronic equipment and the storage medium.
According to an embodiment of the present invention, there is provided an embodiment of a method for evaluating a simulation model of an on-vehicle lidar, it being noted that the steps shown in the flowchart of the drawings may be performed in a computer system such as a set of computer-executable instructions, and that although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in an order different from that herein.
In this embodiment, a vehicle-mounted lidar simulation model evaluation method is provided, which may be used in the above mobile terminal, such as a central processing unit, a server, etc., fig. 1 is a flowchart of a vehicle-mounted lidar simulation model evaluation method according to an embodiment of the present invention, and as shown in fig. 1, the flowchart includes the following steps:
step S101, determining the modeling grade of the laser radar simulation model based on the modeling information of the laser radar simulation model.
In order to determine the requirement of the intelligent automobile on the simulation model of the laser radar, the embodiment determines the requirement of the intelligent automobile on the simulation model of the laser radar by acquiring the modeling grade of the simulation model of the laser radar and the application scene of the simulation model of the laser radar.
First, a modeling level of a lidar simulation model is determined based on modeling information of the lidar simulation model to be evaluated.
Step S102, based on the intelligent automobile design operation range, acquiring an application scene of a laser radar simulation model.
The intelligent automobile design operation range is a scene that the intelligent internet-connected automobile can operate, which is specified in the existing intelligent internet-connected automobile operation specification. Illustratively, the highway runs a scene.
The laser radar is arranged in the intelligent network-connected automobile, so that the application scene of the laser radar simulation model is required to meet the operation scene specified by the operation specification of the intelligent network-connected automobile, that is, the application scene of the laser radar simulation model is acquired when the application scene of the laser radar simulation model to be evaluated meets the operation scene specified by the operation specification of the intelligent network-connected automobile.
The application scene of the laser radar simulation model determines the stage of intelligent automobile simulation, and further determines the simulation degree requirement on the laser radar simulation model.
Step S103, based on the application scene and modeling level of the laser radar simulation model, importance assessment is carried out on the laser radar simulation model, and an importance assessment result of the laser radar simulation model is obtained.
After the application scene of the laser radar simulation model and the modeling grade of the laser radar simulation model are obtained, importance evaluation is carried out on the laser radar simulation model to be evaluated.
In this embodiment, the importance evaluation of the lidar simulation model is essentially to obtain the simulation degree requirement of the intelligent automobile on the lidar simulation model in the application scenario.
Step S104, determining working conditions required for evaluating the laser radar simulation model based on the importance evaluation result of the laser radar simulation model.
After obtaining the simulation degree requirement of the laser radar simulation model, determining the working condition required by evaluating the laser radar simulation model according to the simulation degree requirement of the laser radar simulation model.
Step S105, collecting first test data of a laser radar simulation model under working conditions and second test data of a real laser radar under working conditions.
It can be understood that after the working conditions required for evaluating the laser radar simulation model are determined, working condition preparation is performed, and after the working conditions are prepared, test data of the laser radar simulation model and the real laser radar under the same working conditions are obtained so as to perform simulation evaluation on the laser radar simulation model later.
And step S106, performing simulation evaluation on the laser radar simulation model based on the first test data and the second test data to obtain a simulation evaluation result of the laser radar simulation model.
After test data of the laser radar simulation model and the real laser radar under the same test working condition are obtained, performing simulation degree evaluation on the laser radar simulation model according to the test data of the laser radar simulation model and the test data of the real laser radar, and obtaining a simulation degree evaluation result of the laser radar simulation model.
The simulation degree of the laser radar simulation model refers to the degree of approach of the laser radar simulation model to data acquired by a real laser radar under the same working condition.
According to the vehicle-mounted laser radar simulation model evaluation method, importance evaluation is carried out on the laser radar simulation model based on the application scene and modeling level of the laser radar simulation model, the evaluation dimension covers the actual application scene of the laser radar sensor, different requirements of the intelligent automobile on the simulation degree of the laser radar simulation model in the simulation stage are considered, different test working conditions are selected according to the different requirements of the simulation degree of the laser radar model, simulation degree evaluation of the laser radar simulation model is carried out, and the current situation of research, development and verification requirements of an intelligent driving system is met.
In this embodiment, a vehicle-mounted laser radar simulation model evaluation method is provided, which may be used in the above mobile terminal, such as a central processing unit, a server, etc., fig. 2 is a flowchart of a laser radar simulation model evaluation method according to an embodiment of the present invention, and as shown in fig. 2, the flowchart includes the following steps:
step S201, determining a modeling level of the lidar simulation model based on modeling information of the lidar simulation model.
Specifically, the step S201 includes:
step S2011, determining the modeling level of the laser radar simulation model based on the modeling mode and the model output of the laser radar simulation model.
Exemplary modeling approaches include: maxwell's equations, ray tracing, black box models, phenomenon models, truth lists, etc. The model outputs may be native data, probe lists, target lists, and the like.
And determining the modeling grade of the laser radar simulation model according to the modeling mode and the model output of the laser radar simulation model to be evaluated.
In an alternative embodiment, step S2011 includes:
and a step a1, wherein when the modeling mode of the laser radar simulation model is Maxwell's equation set and the model output is the original data, the modeling grade of the laser radar simulation model is determined to be high-grade modeling.
When the modeling mode of the laser radar simulation model is ray tracing and the model output is the original data, the modeling level of the laser radar simulation model is determined to be high-level modeling.
And a step a2, wherein when the modeling mode of the laser radar simulation model is a black box model and the model output is a detection list, the modeling level of the laser radar simulation model is determined to be medium-level modeling.
It should be noted that, when the modeling mode of the lidar simulation model is a phenomenon model and the model output is a detection list, it is determined that the modeling level of the lidar simulation model is a medium level modeling.
And a step a3, determining the modeling level of the laser radar simulation model as low-level modeling under the condition that the modeling mode of the laser radar simulation model is a true value list and the model output is a target list.
Table 1 shows the lidar simulation model modeling level assessment requirements.
TABLE 1
It should be noted that, because of limitation in determining the modeling level based on the modeling manner, the present embodiment mainly uses the model output type to evaluate the modeling level. Specifically, under the condition that the model output of the laser radar simulation model is the original data, determining the modeling level of the laser radar simulation model as high-level modeling; under the condition that the model output of the laser radar simulation model is a detection list, determining the modeling level of the laser radar simulation model as medium level modeling; and under the condition that the model output of the laser radar simulation model is a target list, determining the modeling level of the laser radar simulation model as low-level modeling.
Step S202, acquiring an application scene of a laser radar simulation model based on the intelligent automobile design operation range.
Please refer to step S102 in the embodiment shown in fig. 1 in detail, which is not described herein.
Step S203, importance assessment is carried out on the laser radar simulation model based on the application scene and the modeling level of the laser radar simulation model, and an importance assessment result of the laser radar simulation model is obtained.
Specifically, as shown in fig. 3, the step S203 includes:
step S2031, determining the influence degree of the application scene of the laser radar simulation model on the simulation test result according to the application scene of the laser radar simulation model.
It should be noted that, the application scenario of the laser radar simulation model determines that the laser radar simulation model is in the simulation stage of the intelligent network-connected automobile.
The influence degree of the application scene of the laser radar simulation model on the simulation test result is evaluated and set by technicians. Specifically, technicians divide the influence degree of different application scenes on simulation test results into three levels, namely low influence, medium influence and high influence.
Step S2032, inputting the modeling level of the laser radar simulation model and the influence degree of the application scene of the laser radar simulation model on the simulation test result into an importance evaluation matrix to obtain the importance evaluation result of the laser radar simulation model.
Under the condition that the modeling level of the laser radar simulation model and the influence degree of the application scene of the laser radar simulation model on the simulation test result are obtained, the importance evaluation matrix is input, and the importance evaluation result of the laser radar simulation model is obtained.
As shown in fig. 4, when the modeling level of the lidar simulation model is low-level modeling and the degree of influence of the application scenario of the lidar simulation model on the simulation test result is low, the importance evaluation result is a low-fidelity requirement.
And under the condition that the modeling level of the laser radar simulation model is medium modeling and the influence degree of the application scene of the laser radar simulation model on the simulation test result is medium influence, the importance evaluation result is a requirement of the simulation degree and the like.
Under the condition that the modeling level of the laser radar simulation model is high-level modeling and the influence degree of the application scene of the laser radar simulation model on the simulation test result is high, the importance evaluation result is a high-fidelity requirement.
It should be noted that, under the condition that the modeling level of the laser radar simulation model and the influence degree of the application scene of the laser radar simulation model on the simulation test result are not at the same level, the importance evaluation result is obtained by taking the higher level of the modeling level and the influence degree of the application scene of the laser radar simulation model as the standard.
For example, in the case that the modeling level of the lidar simulation model is low-level modeling, and the influence degree of the application scene of the lidar simulation model on the simulation test result is a medium influence, the importance evaluation result is a requirement of reality degree or the like.
Step S204, determining working conditions required for evaluating the laser radar simulation model based on the importance evaluation result of the laser radar simulation model.
Please refer to step S104 in the embodiment shown in fig. 1 in detail, which is not described herein.
In an alternative embodiment, step S204 includes:
and b1, determining working conditions required by evaluating the laser radar simulation model as laser radar performance dimension working conditions, point cloud quality dimension working conditions and application scene performance dimension working conditions under the condition that the importance evaluation result of the laser radar simulation model is high in simulation degree.
And b2, determining the working condition required by evaluating the laser radar simulation model as the laser radar performance dimension working condition under the condition that the importance evaluation result of the laser radar simulation model is the requirement of the degree of similarity.
And b3, determining that the laser radar simulation model does not need to be subjected to the simulation degree evaluation under the condition that the importance evaluation result of the laser radar simulation model is low in simulation degree requirement.
It should be noted that, the importance evaluation result of the laser radar simulation model is a low-fidelity requirement, which is modeled in a modeling manner of a truth value list, and the similarity evaluation result of the default laser radar simulation model is high.
Specifically, table 2 shows different dimensional conditions of the lidar simulation model, as shown in table 2:
TABLE 2
Step S205, collecting first test data of a laser radar simulation model under working conditions and second test data of a real laser radar under working conditions.
Under the condition that the test working condition required by evaluating the laser radar simulation model is determined, the test working condition is prepared, after the test working condition is prepared, the test is performed under the required test working condition, and the first test data of the laser radar simulation model under the required working condition and the second test data of the real laser radar under the same working condition are collected.
In order to describe the calculation process of the evaluation index in detail, taking the application scene as an example of a stationary scene of the front vehicle in the automatic emergency braking system test method, the evaluation process using the point cloud result index is described. The laser radar measuring directions are all selected from nine test areas with equal dividing areas in a view field plane perpendicular to the normal reflection direction of the laser radar, as shown in fig. 12.
Working condition one: detection accuracy evaluation under laser radar performance dimension
1. Test scenario: as shown in FIG. 5, the test road is a long straight road at least comprising two lanes, the laser radar, the precise turntable and the reflecting plate are positioned at the middle position of the right road, the distance between the laser radar and the reflecting plate is 5m,10m,20m,50m,80m,100m, more than 100m and gradually increased at a step distance of 40 m.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) Fixing the laser radar on a precise turntable, and adjusting the relative position of the laser radar and the reflecting plate to enable the plane of the reflecting plate to be perpendicular to the normal direction (light emitting direction) of the laser radar, so that the light emitting direction of the laser radar is opposite to the central position of the reflecting plate;
2) Measuring a reference distance on a laser radar position point by using a reference range finder, wherein the reference distance refers to the distance from the laser radar position point to the center of the reflecting plate;
3) Starting a laser radar, enabling enough heat engine time to reach a stable working temperature, and selecting a measuring point in a nine-grid according to the method of FIG. 12;
4) Continuously recording the measurement point cloud to obtain data of not less than 100 frames, wherein the data comprises: the distance of the laser radar from the measuring point;
5) Repeating steps 1) to 4) at each measurement distance and measurement point.
b. And (3) a simulation laser radar testing step:
1) According to the relative distance between the laser radar and the reflecting plate measured in the real laser radar testing step 1), placing the laser radar and the reflecting plate at the same relative distance in a simulation scene;
2) Starting a simulation test system, and selecting a measuring point in a nine-grid according to the method of FIG. 12;
3) Continuously recording the measurement point cloud to obtain data of not less than 100 frames, wherein the data comprises: the distance of the laser radar from the measuring point;
4) Repeating steps 1) to 3) at each measurement distance and measurement point.
3. Judgment mode:
wherein: mu 1 represents the accuracy of the detection distance,represents the average value of the distance measurement, and d1 represents the single measurement value of the distance.
Working condition II: blind area evaluation under laser radar performance dimension:
1. test scenario: as shown in fig. 6, the test road is a long straight road at least comprising two lanes, the laser radar, the precise turntable and the reflecting plate are positioned at the middle position of the right road, and the relative distance between the laser radar and the reflecting plate is 1m or more and the distance between the laser radar and the reflecting plate and the clear and complete point cloud data appear in the corresponding point cloud area.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) The laser radar to be measured is fixed on a precise turntable, the rotation central axis of the precise turntable is coincident with the vertical central axis of the laser radar, the relative position of the laser radar and the reflecting plate is adjusted, the plane of the reflecting plate is perpendicular to the normal direction (light emitting direction) of the laser radar, and the light emitting direction of the laser radar is opposite to the horizontal central position of the reflecting plate. The distance between the reflecting plate and the laser radar is smaller than the nominal minimum range of the laser radar;
2) Starting a laser radar, enabling enough heat engine time to reach stable working temperature, selecting 50cm in the middle of a reflecting plate as an effective area to collect point cloud, gradually translating the reflecting plate away from the laser radar until 1m or more distance from the reflecting plate to the point cloud area corresponding to the reflecting plate to generate clear and complete point cloud data, and recording all dynamic distances;
3) Measuring a reference distance at a laser radar position point with a reference range finder;
4) Rotating the laser radar to enable points in the effective area to sequentially enter the reflecting plate;
5) Repeating steps 1) to 4) in each test direction measuring distance.
b. And (3) a simulation laser radar testing step:
1) According to the dynamic distance measured in the real laser radar testing step 2), placing the laser radar and the reflecting plate to the same relative distance in a simulation scene;
2) Starting a simulation test system, and selecting 50cm cmX cm in the middle of the reflecting plate as an effective area acquisition point cloud;
3) Continuously recording the measurement point cloud to obtain data of not less than 100 frames;
4) Repeating steps 1) to 3) at each measurement distance and measurement point.
3. Judgment mode:
and the minimum detection distance is within 50cm cmX cm of the center of the reflecting plate, the accuracy requirement is met, and the detection probability is not less than 50%.
And (3) working condition III: evaluation of angle accuracy in the laser radar performance dimension:
1. test scenario: as shown in FIG. 7, the test road is a long straight road at least comprising two lanes, and the laser radar, the precise turntable and the reflecting plate are positioned at the middle position of the right road, and the distance is 10m+0.02m.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) Placing the laser radar above the precise turntable, and ensuring that the rotation center of the precise turntable and the optical center of the laser radar coincide;
2) Placing a laser radar at an origin, placing a reflecting plate in 10m+0.02m right in front of the laser radar, turning on the laser radar and preheating, collecting at least 30 frames of data at an angle position i on the reflecting plate, and calculating a point cloud angle average value
3) The precise turntable is rotated for a certain angle (such as 5 degrees, 10 degrees and 15 degrees), and the degree is recordedThe laser radar reaches the position j at the moment;
4) Collecting at least 30 frames of data at position j, calculating a point cloud angle average
5) Calculating the angle accuracy of the laser radar in the direction;
6) Testing 9 areas in the nine-grid, and taking an average value as a final result;
7) Repeating steps 3) to 6) for 5 times
b. And (3) a simulation laser radar testing step:
1) According to the angle i in the real laser radar testing step 2), placing the laser radar and the reflecting plate to the same angle in a simulation scene;
2) Starting a simulation test system, and selecting a measuring point in a nine-grid according to the method of FIG. 12;
3) Continuously recording the measurement point cloud to obtain data of not less than 30 frames;
4) Repeating steps 1) to 3).
3. Judgment mode:
wherein delta is θ Is the angular accuracy of the lidar.
And (4) working condition four: angular resolution evaluation under lidar performance dimension:
1. test scenario: as shown in FIG. 8, the test road is a long straight road at least comprising two lanes, and the laser radar, the precise turntable and the reflecting plate are positioned at the middle position of the right road, and the distance is 10m+0.02m.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) Placing the laser radar on the precise turntable, ensuring that the laser radar is coaxial with the precise turntable, and opening and preheating the laser radar;
2) Placing a reflecting plate at 10m+0.02m right in front of the laser radar;
3) Rotating the laser radar, observing the point cloud until the point cloud corresponding to the edge of the reflecting plate moves from a first line/point to an adjacent next line/point, as shown in fig. 8, moving from a line/point 2 to a line/point 3, changing the precise turntable from a position a to a position b, wherein the angle rotated by the precise turntable is the angular resolution of the current two channels;
4) Traversing all channels of the laser radar, and measuring the angular resolution between every two adjacent lines/points;
5) The angle resolution in the vertical direction is consistent with the method, the rotation direction of the precise turntable is changed into the vertical direction, and the reflecting plate is changed into the vertical direction.
b. And (3) a simulation laser radar testing step:
1) Placing a reflecting plate at a position of 10m+0.02m right in front of the laser radar simulation model;
2) Starting a simulation test system, continuously recording a measurement point cloud, and obtaining data of not less than 30 frames;
3) Rotating the laser radar, observing the point cloud until the point cloud corresponding to the edge of the reflecting plate moves from a first line/point to an adjacent next line/point, as shown in fig. 8, moving from a line/point 2 to a line/point 3, changing the precise turntable from a position c to a position d, wherein the angle rotated by the turntable is the angular resolution of the current two channels;
4) Traversing all channels of the laser radar, and measuring the angular resolution between every two adjacent lines/points;
5) The angle resolution in the vertical direction is consistent with the method, the rotating direction of the turntable is changed into the vertical direction, and the reflecting plate is changed into the vertical direction.
3. Judgment mode:
angle gamma rotated by precision turntable θ The angular resolution of the current two channels is the same.
Working condition five: evaluation of the field angle under the performance dimension of the laser radar:
1. test scenario: as shown in FIG. 9, the test road is a long straight road at least comprising two lanes, and the laser radar, the precise turntable and the reflecting plate are positioned at the middle position of the right road, and the distance is 10m+0.02m.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) Placing the laser radar above the precise turntable, and ensuring that the rotation center of the precise turntable is coincident with the center of the laser radar;
2) Placing a reflecting plate at about 10m+0.02m in front of the lidar as shown in fig. 9;
3) Placing a laser radar at an origin, opening the laser radar, preheating, and starting to collect point cloud;
4) When the edge of the reflecting plate is not observed, the angle of the precise turntable is recorded, and the average value of the multiple measurement is recorded as
5) Resetting the laser radar, reversely rotating the precise turntable, recording the angle of the precise turntable when the other edge of the reflecting plate is not observed, and recording the average value of the multiple measurement values as
6) And (5) replacing the laser radar position and the rotation direction of the precise turntable, and taking an average value through multiple measurements.
b. And (3) a simulation laser radar testing step:
1) Placing a reflecting plate at the position about 10m+0.02m in front of the laser radar simulation model;
2) When the edge of the reflecting plate is not observed, the angle of the precise turntable is recorded, and the average value of the multiple measurement is recorded as/>
3) Resetting the laser radar, reversely rotating the precise turntable, recording the angle of the precise turntable when the other edge of the reflecting plate is not observed, and recording the average value of the multiple measurement values as
4) And (5) replacing the laser radar position and the rotation direction of the precise turntable, and taking an average value through multiple measurements.
3. Judgment mode:
wherein beta is 02 For stretching the reflecting plate relative to the originThe angle, FOV, is the field angle of the lidar.
Working condition six: vehicle distance accuracy evaluation under laser radar point cloud quality dimension:
1. test scenario: as shown in fig. 10, the test road is a long straight road including at least two lanes, the laser radar-loaded test Vehicle (VUT) and the target Vehicle (VT) are located at the middle position of the right road, and the distances are 5m,10m,20m,50m,80m,100m, more than 100m, and are increased by 40m steps.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) The relative position of a test vehicle equipped with the laser radar and a target vehicle is adjusted, so that the tail plane of the target vehicle is perpendicular to the normal direction (light emitting direction) of the vehicle-mounted laser radar, and the light emitting direction of the laser radar is opposite to the center position of a rear axle of the target vehicle;
2) Starting a test vehicle equipped with a laser radar, enabling enough heat engine time to reach the stable working temperature of the laser radar, and selecting a measuring point in a nine-grid according to the method of FIG. 12;
3) Continuously recording the measurement point cloud to obtain data of not less than 100 frames;
4) Repeating steps 1) to 3) at each measured distance.
b. And (3) a simulation laser radar testing step:
1) According to the relative distance between the laser radar and the target vehicle measured in the real laser radar testing step 1), assembling a laser radar simulation model on the simulation test vehicle in a simulation scene, and positioning the laser radar simulation model at the same relative distance with the target vehicle simulation model;
2) Starting a simulation test system, and selecting a measuring point in a nine-grid according to the method of FIG. 12;
3) Continuously recording the measurement point cloud to obtain data of not less than 100 frames;
4) Repeating steps 1) to 3) at each measured distance
3 judgment modes:
wherein: μ2 represents the vehicle distance accuracy,represents the average of the vehicle distance measurements, and d2 represents a single vehicle distance measurement.
Working condition seven: european clustering result evaluation under laser radar point cloud quality dimension:
1. test scenario: as shown in fig. 10, the test road is a long straight road including at least two lanes, the test vehicle and the target vehicle carrying the laser radar are located at the middle position of the right road, the distances are 5m,10m,20m,50m,80m,100m, more than 100m, and the steps are increased by 40 m.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) The relative position of a test vehicle equipped with the laser radar and a target vehicle is adjusted, so that the vehicle tail plane of the target vehicle is perpendicular to the normal direction (light emitting direction) of the vehicle-mounted laser radar, and the light emitting direction of the laser radar is opposite to the center position of a rear axle of the target vehicle;
2) Continuously measuring detection target bounding box (bounding box) data of not less than 100 frames;
3) Repeating steps 1) to 2) at each measured distance.
b. And (3) a simulation laser radar testing step:
1) According to the relative distance between the laser radar and the target vehicle measured in the real laser radar testing step 1), assembling a laser radar simulation model on the simulation test vehicle in a simulation scene, and positioning the laser radar simulation model at the same relative distance with the target vehicle simulation model;
2) Starting a simulation test system, and continuously measuring the bounding box data of the detection target object with the frame number not less than 100;
3) Repeating steps 1) to 2) at each measured distance.
3. Judgment mode:
A d ＝|A r -A s |
wherein A is d Detection of target objects for real lidarArea difference of bounding box and simulated laser radar detection target bounding box, A r For real laser radar detection of the area of a target object bounding box, A s The area of the target bounding box is detected for the simulated lidar.
And calculating the area difference of the real bounding box and the simulation bounding box through the European clustering output detection target bounding box.
Working condition eight: and evaluating a point cloud result in a laser radar application scene:
1. test scenario: taking a stationary scene of a front vehicle to be handled by an automatic emergency braking system as an example, as shown in fig. 11, a test road is a long straight road at least comprising two lanes, a laser radar is mounted on the test vehicle, and a stationary target vehicle exists 300 meters in front. The test vehicle approaches the front stationary target vehicle at a speed of 10km/h, 30 km/h.
2. The testing method comprises the following steps:
a. the real laser radar testing step:
1) The method comprises the steps of (1) adjusting the relative position relation between a vehicle and a front stationary target vehicle by a vehicle provided with a laser radar, so that the light emitting direction of the laser radar is opposite to the center of a rear axle of the target vehicle;
2) Measuring a reference distance on a laser radar position point by using a reference distance meter, and adjusting the distance between a target vehicle and the vehicle by 300 m;
3) Continuously recording the measurement point cloud, and stopping recording when the speed of the vehicle is equal to zero;
b. and (3) a simulation laser radar testing step:
1) According to the real laser radar test scene, a twin simulation scene, an environment model and a vehicle model are established according to the same size;
2) Testing the relative distance between the laser radar measured in the step 1) and the front target vehicle according to the real laser radar;
3) Starting a simulation test system, continuously recording the measurement point cloud, and stopping recording when the speed of the vehicle is equal to zero;
3. judgment mode:
wherein: c (C) P The correlation between the laser radar simulation model and the sensor point cloud is that m is the total number of laser radar wire harnesses, n m The point cloud data is the total point number on the m-th wire harness of the laser radar, x is the point cloud data of the simulation model of the laser radar, and y is the point cloud data of the real laser radar sensor.
Step S206, based on the first test data and the second test data, performing simulation evaluation on the laser radar simulation model to obtain a simulation evaluation result of the laser radar simulation model.
Specifically, the step S206 includes:
in step S2061, a deviation value between the first test data and the second test data is determined.
After data acquisition, the index deviation F under the working conditions of different dimensions can be calculated (i) The evaluation method is as follows:
wherein S is (i) R is an index result obtained by collecting a laser radar simulation model under the ith working condition (i) For the index result obtained by the real laser radar sensor under the ith working condition, if the index deviation is judged in the working condition, the step can be omitted.
Based on model application requirements and expert experience, different indexes can be weighted through a hierarchical analysis method, and the overall deviation value of the laser radar simulation model is given, when the overall deviation value accords with the expectation, namely the laser radar simulation model has good fidelity, otherwise, the laser radar simulation model does not have good fidelity, and the overall deviation value of the laser radar simulation model is calculated as follows:
wherein F is Lidar The integral deviation value of the laser radar simulation model is n, the total number of acquisition working conditions is w (i) And evaluating the index weight for the ith working condition.
In step S2062, in the case that the deviation value is within the preset range, it is determined that the simulation result of the laser radar simulation model is good.
Wherein the preset range is determined by a technician.
The deviation value of each index is within the preset range, and the simulation result of the laser radar simulation model is considered to be good in simulation degree, so that the laser radar simulation model can be applied to relevant test and verification work.
Of course, the overall deviation value may be within a preset range of the overall deviation value of the laser radar simulation model, so as to determine that the simulation result of the laser radar simulation model is good in simulation.
In step S2063, in the case that the deviation value is not within the preset range, it is determined that the result of the evaluation of the fidelity of the lidar simulation model is not up to the expectation.
And if the evaluation result of the simulation degree of the laser radar simulation model does not reach the expectation, correcting the laser radar simulation model according to the evaluation result.
Specifically, the laser radar simulation model under the same working condition is corrected based on the test data of the real laser radar under each working condition.
In this embodiment, a vehicle-mounted lidar simulation model evaluation device is further provided, and the device is used for implementing the foregoing embodiments and preferred embodiments, and is not described again. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
The present embodiment provides a vehicle-mounted lidar simulation model evaluation device, as shown in fig. 13, including:
the modeling level determining module 1301 is configured to determine a modeling level of the lidar simulation model based on modeling information of the lidar simulation model.
The application scene acquisition module 1302 is configured to acquire an application scene of the laser radar simulation model based on the intelligent automobile design operation range.
The importance evaluation result obtaining module 1303 is configured to perform importance evaluation on the lidar simulation model based on the application scenario and the modeling level of the lidar simulation model, so as to obtain an importance evaluation result of the lidar simulation model.
The working condition determining module 1304 is configured to determine a working condition required for evaluating the lidar simulation model based on an importance evaluation result of the lidar simulation model.
The test data acquisition module 1305 is used for acquiring first test data of the laser radar simulation model under the working condition and second test data of the real laser radar under the working condition.
The simulation evaluation result obtaining module 1306 is configured to perform a simulation evaluation on the laser radar simulation model based on the first test data and the second test data, so as to obtain a simulation evaluation result of the laser radar simulation model.
In some alternative embodiments, modeling level determination module 1301 includes:
the modeling level determining subunit is used for determining the modeling level of the laser radar simulation model based on the modeling mode and the model output of the laser radar simulation model.
In some alternative embodiments, the modeling level determination subunit includes:
the high-level modeling determining unit is used for determining that the modeling level of the laser radar simulation model is high-level modeling under the condition that the modeling mode of the laser radar simulation model is Maxwell equation set and the model output is primary data;
the medium-level modeling determining unit is used for determining that the modeling level of the laser radar simulation model is medium-level modeling under the condition that the modeling mode of the laser radar simulation model is a black box model and the model output is a detection list;
the low-level modeling determining unit is used for determining that the modeling level of the laser radar simulation model is low-level modeling under the condition that the modeling mode of the laser radar simulation model is a true value list and the model output is a target list.
In some alternative embodiments, the importance assessment result obtaining module 1303 includes:
the importance evaluation result acquisition subunit is used for determining the influence degree of the application scene of the laser radar simulation model on the simulation test result according to the application scene of the laser radar simulation model, inputting the modeling grade of the laser radar simulation model and the influence degree of the application scene of the laser radar simulation model on the simulation test result into the importance evaluation matrix, and obtaining the importance evaluation result of the laser radar simulation model.
In some alternative embodiments, the condition determination module 1304 includes:
the working condition determining subunit is used for determining that the working condition required by the laser radar simulation model is the laser radar performance dimension working condition, the point cloud quality dimension working condition and the application scene performance dimension working condition under the condition that the importance assessment result of the laser radar simulation model is the high-fidelity requirement, determining that the working condition required by the laser radar simulation model is the laser radar performance dimension working condition under the condition that the importance assessment result of the laser radar simulation model is the medium-fidelity requirement, and determining that the laser radar simulation model does not need to be subjected to the fidelity assessment under the condition that the importance assessment result of the laser radar simulation model is the low-fidelity requirement.
In some alternative embodiments, the simulation evaluation result acquisition module 1306 includes:
and the deviation value determining unit is used for determining a deviation value between the first test data and the second test data.
The first simulation evaluation result acquisition subunit is used for determining that the simulation evaluation result of the laser radar simulation model is good in simulation degree under the condition that the deviation value is in a preset range.
In some alternative embodiments, the simulation evaluation result acquisition module 1306 includes:
And the second simulation evaluation result acquisition subunit is used for determining that the simulation evaluation result of the laser radar simulation model does not reach the expected value under the condition that the deviation value is not in the preset range.
Further functional descriptions of the above respective modules and units are the same as those of the above corresponding embodiments, and are not repeated here.
The vehicle-mounted lidar simulation model evaluation apparatus in this embodiment is presented in the form of functional units, where the units refer to ASIC (Application Specific Integrated Circuit ) circuits, processors and memories executing one or more software or fixed programs, and/or other devices that can provide the above functions.
The embodiment of the invention also provides computer equipment, which is provided with the vehicle-mounted laser radar simulation model evaluation device shown in the figure 13.
Referring to fig. 14, fig. 14 is a schematic structural diagram of a computer device according to an alternative embodiment of the present invention, as shown in fig. 14, the computer device includes: one or more processors 1401, memory 1402, and interfaces for connecting the components, including a high-speed interface and a low-speed interface. The various components are communicatively coupled to each other using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions executing within the computer device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In some alternative embodiments, multiple processors and/or multiple buses may be used, if desired, along with multiple memories and multiple memories. Also, multiple computer devices may be connected, each providing a portion of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). One processor 1401 is illustrated in fig. 14.
The processor 1401 may be a central processor, a network processor, or a combination thereof. Wherein the processor 1401 may further comprise a hardware chip. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general-purpose array logic, or any combination thereof.
Wherein the memory 1402 stores instructions executable by the at least one processor 1401 to cause the at least one processor 1401 to perform the methods illustrated by implementing the embodiments described above.
Memory 1402 may include a storage program area that may store an operating system, at least one application program required for functions, and a storage data area; the storage data area may store data created according to the use of the computer device, etc. Further, memory 1402 can include high-speed random access memory, and can also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, memory 1402 optionally includes memory located remotely from processor 1401, which may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Memory 1402 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid state disk; memory 1402 may also include a combination of the above types of memory.
The computer device also includes a communication interface 1403 for the computer device to communicate with other devices or communication networks.
The embodiments of the present invention also provide a computer readable storage medium, and the method according to the embodiments of the present invention described above may be implemented in hardware, firmware, or as a computer code which may be recorded on a storage medium, or as original stored in a remote storage medium or a non-transitory machine readable storage medium downloaded through a network and to be stored in a local storage medium, so that the method described herein may be stored on such software process on a storage medium using a general purpose computer, a special purpose processor, or programmable or special purpose hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, a solid state disk or the like; further, the storage medium may also comprise a combination of memories of the kind described above. It will be appreciated that a computer, processor, microprocessor controller or programmable hardware includes a storage element that can store or receive software or computer code that, when accessed and executed by the computer, processor or hardware, implements the methods illustrated by the above embodiments.
Although embodiments of the present invention have been described in connection with the accompanying drawings, various modifications and variations may be made by those skilled in the art without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope of the invention as defined by the appended claims.
Claims (10)
1. A vehicle-mounted lidar simulation model evaluation method, the method comprising:
determining a modeling grade of the laser radar simulation model based on modeling information of the laser radar simulation model;
acquiring an application scene of the laser radar simulation model based on the intelligent automobile design operation range;
based on the application scene and modeling level of the laser radar simulation model, carrying out importance assessment on the laser radar simulation model to obtain an importance assessment result of the laser radar simulation model;
determining working conditions required by evaluating the laser radar simulation model based on an importance evaluation result of the laser radar simulation model;
collecting first test data of the laser radar simulation model under the working condition and second test data of the real laser radar under the working condition;
and performing simulation evaluation on the laser radar simulation model based on the first test data and the second test data to obtain a simulation evaluation result of the laser radar simulation model.
2. The method of claim 1, wherein determining a modeling level of the lidar simulation model based on modeling information of the lidar simulation model comprises:
and determining the modeling grade of the laser radar simulation model based on the modeling mode and the model output of the laser radar simulation model.
3. The method of claim 2, wherein the determining the modeling level of the lidar simulation model based on the modeling manner and the model output of the lidar simulation model comprises:
determining the modeling level of the laser radar simulation model as high-level modeling under the condition that the modeling mode of the laser radar simulation model is Maxwell equation set and the model output is primary data;
determining that the modeling level of the laser radar simulation model is medium modeling under the condition that the modeling mode of the laser radar simulation model is a black box model and the model output is a detection list;
and determining the modeling level of the laser radar simulation model as low-level modeling under the condition that the modeling mode of the laser radar simulation model is a true value list and the model output is a target list.
4. The method according to claim 1, wherein the performing importance assessment on the lidar simulation model based on the application scenario and the modeling level of the lidar simulation model to obtain the importance assessment result of the lidar simulation model includes:
determining the influence degree of the application scene of the laser radar simulation model on a simulation test result according to the application scene of the laser radar simulation model;
inputting the modeling grade of the laser radar simulation model and the influence degree of the application scene of the laser radar simulation model on the simulation test result into an importance evaluation matrix to obtain the importance evaluation result of the laser radar simulation model.
5. The method according to claim 1, wherein the determining, based on the importance evaluation result of the lidar simulation model, a working condition required for evaluating the lidar simulation model includes:
under the condition that the importance assessment result of the laser radar simulation model is high in simulation degree, determining working conditions required by assessing the laser radar simulation model to be laser radar performance dimension working conditions, point cloud quality dimension working conditions and application scene performance dimension working conditions;
Under the condition that the importance assessment result of the laser radar simulation model is a requirement of intermediate degree of reality, determining a working condition required by assessing the laser radar simulation model as a laser radar performance dimension working condition;
and under the condition that the importance evaluation result of the laser radar simulation model is that the simulation degree is low, determining that the laser radar simulation model does not need to be subjected to simulation degree evaluation.
6. The method according to claim 1, wherein performing the simulation evaluation on the lidar simulation model based on the first test data and the second test data to obtain a simulation evaluation result of the lidar simulation model includes:
determining a deviation value between the first test data and the second test data;
and under the condition that the deviation value is in a preset range, determining that the simulation degree evaluation result of the laser radar simulation model is good in simulation degree.
7. The method of claim 6, wherein the method further comprises:
and under the condition that the deviation value is not in a preset range, determining that the simulation evaluation result of the laser radar simulation model is not up to the expected value.
8. An on-vehicle lidar simulation model evaluation device, the device comprising:
The modeling grade determining module is used for determining the modeling grade of the laser radar simulation model based on the modeling information of the laser radar simulation model;
the application scene acquisition module is used for acquiring an application scene of the laser radar simulation model based on the intelligent automobile design operation range;
the importance evaluation result acquisition module is used for carrying out importance evaluation on the laser radar simulation model based on the application scene and the modeling level of the laser radar simulation model to obtain an importance evaluation result of the laser radar simulation model;
the working condition determining module is used for determining working conditions required by evaluating the laser radar simulation model based on the importance evaluation result of the laser radar simulation model;
the test data acquisition module is used for acquiring first test data of the laser radar simulation model under the working condition and second test data of the real laser radar under the working condition;
and the simulation degree evaluation result acquisition module is used for carrying out simulation degree evaluation on the laser radar simulation model based on the first test data and the second test data to obtain a simulation degree evaluation result of the laser radar simulation model.
9. A computer device, comprising:
a memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the method for evaluating a vehicle-mounted lidar simulation model of any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon computer instructions for causing a computer to execute the vehicle-mounted lidar simulation model evaluation method according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311694725.0A CN117763818A (en) | 2023-12-11 | 2023-12-11 | Vehicle-mounted laser radar simulation model evaluation method and device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311694725.0A CN117763818A (en) | 2023-12-11 | 2023-12-11 | Vehicle-mounted laser radar simulation model evaluation method and device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117763818A true CN117763818A (en) | 2024-03-26 |
Family
ID=90313550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311694725.0A Pending CN117763818A (en) | 2023-12-11 | 2023-12-11 | Vehicle-mounted laser radar simulation model evaluation method and device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117763818A (en) |
-
2023
- 2023-12-11 CN CN202311694725.0A patent/CN117763818A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109946701B (en) | Point cloud coordinate conversion method and device | |
| CN113484851B (en) | Simulation test system and method for vehicle-mounted laser radar and complete vehicle in-loop test system | |
| CN109671174A (en) | A kind of pylon method for inspecting and device | |
| CN113554643B (en) | Target detection method and device, electronic equipment and storage medium | |
| CN106370435A (en) | Road spectrum acquisition system, vehicle and road spectrum acquisition method | |
| CN112861833B (en) | Vehicle lane level positioning method and device, electronic equipment and computer readable medium | |
| US11636684B2 (en) | Behavior model of an environment sensor | |
| CN112036359B (en) | Method for obtaining topological information of lane line, electronic device and storage medium | |
| CN116702607A (en) | BIM-FEM-based bridge structure digital twin body and method | |
| CN116580271A (en) | Evaluation method, device, equipment and storage medium for perception fusion algorithm | |
| CN114384547A (en) | Radar sensor model-based fidelity detection evaluation method and system | |
| JP2020109602A (en) | Model generation device, vehicle simulation system, model generation method and computer program | |
| US20230281873A1 (en) | Method for calibrating a portable reference sensor system, portable reference sensor system and use of the portable reference sensor system | |
| CN109816736B (en) | Automatic calibration method and system for vehicle camera and vehicle-mounted control equipment | |
| CN117763818A (en) | Vehicle-mounted laser radar simulation model evaluation method and device and computer equipment | |
| CN113066185B (en) | Scene data creating method, device, medium and equipment of mine simulation system | |
| CN112710343B (en) | RT-based vehicle-mounted sensor performance test method | |
| CN114325662A (en) | External parameter calibration method, device, equipment and storage medium for vehicle-mounted radar | |
| CN114791282A (en) | Road facility coordinate calibration method and device based on vehicle high-precision positioning | |
| JP7365633B2 (en) | Vehicle simulation system, vehicle simulation method and computer program | |
| CN112729366A (en) | Test evaluation method and device for weather simulation equipment for automatic driving field test | |
| CN113702931A (en) | External parameter calibration method and device for vehicle-mounted radar and storage medium | |
| CN115267861B (en) | Automatic driving fusion positioning precision testing method and device and electronic equipment | |
| CN112985455B (en) | Precision evaluation method and device of positioning and attitude determination system and storage medium | |
| CN113762001B (en) | Target detection method and device, electronic equipment and storage medium |
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 |