Vehicle infrastructure cooperation scene test evaluation method and system

Abstract

The invention discloses a vehicle-road cooperation scene test evaluation method and system, and relates to the field of intelligent traffic, and the method comprises the steps: capturing the dynamic information of vehicles, pedestrians and other traffic participants through the deployment of a sensor network, obtaining the driving state data through a sensor of a vehicle, and carrying out the preprocessing; performing feature extraction and fusion on the image and the laser radar data; constructing an initial test scene library based on the fused data, evaluating the difficulty level and the coverage range of the scene by using a machine learning algorithm, and dynamically adjusting and optimizing the test scene; executing a test according to the test plan, recording data and generating an analysis report; and finally evaluating vehicle performance based on a test result and adjusting the test scene library. A dynamic adjustment mechanism ensures the diversity and representativeness of a test scene, and the comprehensiveness and effectiveness of the test are improved. And evaluation is carried out through a machine learning algorithm, so that the construction of the test scene is more intelligent and automatic.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to a vehicle-road collaboration scenario testing and evaluation method and system. Background Art

[0002] In recent years, with the rapid development of autonomous driving technology, intelligent transportation systems have gradually become a research hotspot. The widespread use of autonomous vehicles in urban transportation has put forward higher requirements for vehicle safety and reliability. At present, the testing of autonomous vehicles mainly relies on traditional laboratory simulation tests and open road tests. Laboratory simulation tests usually use virtual simulation environments, which can easily create various traffic scenarios, but lack the support of real environmental perception data and are difficult to fully reflect actual road conditions. Existing testing methods often ignore the ability to dynamically adjust test scenarios, resulting in a single test scenario that cannot fully cover various complex traffic conditions. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides a vehicle-road collaborative scenario test evaluation method and system to solve the problems of incomplete feature extraction, single test scenario and difficulty in dynamically adjusting test difficulty in existing autonomous driving test methods.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a vehicle-road collaboration scenario test and evaluation method and system, which includes deploying a sensor network to capture dynamic information of vehicles, pedestrians and other traffic participants, obtaining driving status data of the vehicle through sensors carried by the vehicle, and preprocessing the collected data;

[0007] Extract features from image data and lidar data and fuse them;

[0008] Build an initial test scenario library based on the fused data, evaluate the difficulty and coverage of each scenario in the initial scenario library, dynamically adjust the difficulty and complexity of the test scenario according to the evaluation results, and add the optimized scenario to the scenario library;

[0009] According to the test plan, select scenarios from the optimized scenario library to perform tests, record and analyze the data during the test, and generate a test result analysis report;

[0010] Evaluate vehicle performance and adjust the test scenario library based on the test result analysis report.

[0011] As a preferred solution of the vehicle-road collaboration scenario test and evaluation method and system of the present invention, wherein:

[0012] The sensor network includes cameras, lidars, and millimeter-wave radars;

[0013] The cameras cover major intersections and crossings to obtain visual information. The lidars obtain three-dimensional information of the vehicle's surrounding environment, and the millimeter-wave radars obtain the dynamic information of other traffic participants around the vehicle. The dynamic information includes the distance and speed of other traffic participants around the vehicle;

[0014] The driving state data of the vehicle includes speed, acceleration, steering angle, and braking state;

[0015] Perform data cleaning, fill in missing data, and remove high-frequency noise on the collected data.

[0016] As a preferred solution of the vehicle-road cooperation scenario test and evaluation method and system of the present invention, wherein: extract features from the image data and lidar data and perform fusion, specifically including the following steps,

[0017] Load the image file through image processing software, convert the image into a two-dimensional matrix, and obtain the image pixel intensity value at each position in the matrix;

[0018] Extract the scene features of the image data and lidar data;

[0019] Fuse the features of the image data and lidar data.

[0020] As a preferred solution of the vehicle-road cooperation scenario test and evaluation method and system of the present invention, wherein: construct an initial test scenario library based on the fused data, and use machine learning algorithms to evaluate the difficulty level and coverage of each scenario in the initial scenario library.

[0021] As a preferred solution of the vehicle-road cooperation scenario test and evaluation method and system of the present invention, wherein: combine the difficulty level and coverage of each scenario for comprehensive evaluation;

[0022] Set a site scoring threshold, and dynamically adjust the difficulty and complexity of the test scenario according to the comprehensive evaluation result;

[0023] Determine the scenarios selected into the optimized test scenario library according to the comprehensive score.

[0024] As a preferred solution of the vehicle-road cooperation scenario test and evaluation method and system of the present invention, wherein: select scenarios from the optimized scenario library according to the test plan for execution testing. For each selected test scenario, record the data during the test, including vehicle status information, environmental information, and other test-related data;

[0025] After each test scenario ends, record the key performance indicators, abnormal situations, and any failure points to generate a test result analysis report.

[0026] As a preferred solution of the vehicle-road collaborative scenario test evaluation method and system described in the present invention, wherein: evaluate the vehicle performance based on the test result analysis report and adjust the test scenario library;

[0027] Evaluate the overall performance of the vehicle according to the performance indicators in the test result analysis report, and determine the performance indicators and the changing trend of the performance of the vehicle under different test scenarios;

[0028] Evaluate the effectiveness of the scenarios in the current test scenario library according to the test results, and adjust the scenarios in the test scenario library based on the test result analysis report;

[0029] Feed back the test results and the adjusted test scenarios to the relevant teams for the next round of testing. Based on the results of each test, continuously adjust and optimize the test scenario library.

[0030] In a second aspect, the present invention provides a vehicle-road collaborative scenario test evaluation system, including,

[0031] A data acquisition and preprocessing module, which is responsible for collecting the dynamic information of vehicles, pedestrians, and other traffic participants and preprocessing the data;

[0032] A feature extraction and fusion module, which is responsible for extracting features from image and lidar data and performing fusion processing;

[0033] A scenario library construction and evaluation module, which constructs an initial test scenario library, evaluates the difficulty level and coverage of the scenarios, and dynamically adjusts the scenario library;

[0034] A test plan execution and analysis module, which executes the test plan, records the test data, and generates a test result analysis report;

[0035] A performance evaluation and scenario library adjustment module, which evaluates the vehicle performance and adjusts the test scenario library according to the test results.

[0036] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and wherein: when the computer program is executed by the processor, it implements any step of the vehicle-road collaborative scenario test evaluation method and system described in the first aspect of the present invention.

[0037] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and wherein: when the computer program is executed by the processor, it implements any step of the vehicle-road collaborative scenario test evaluation method and system described in the first aspect of the present invention.

[0038] The beneficial effects of the present invention are as follows: By using the Gaussian kernel function to extract and fuse the image data and lidar data, the key information in the traffic scene can be captured more accurately. The fused feature vectors contain complementary information of the image and lidar data, improving the comprehensiveness and description ability of the features, and providing richer data support for the subsequent construction of the test scenario library. By using machine learning algorithms to evaluate the difficulty level and coverage of each scenario in the initial test scenario library and dynamically adjusting the test according to the evaluation results, the test scenario library can be constructed more scientifically. The dynamic adjustment mechanism ensures the diversity and representativeness of the test scenarios, improving the comprehensiveness and effectiveness of the test. Through the evaluation of machine learning algorithms, the construction of the test scenarios becomes more intelligent and automated. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0040] Figure 1 It is a flowchart of the vehicle-road collaborative scenario test evaluation method and system in Embodiment 1.

[0041] Figure 2 It is a determination diagram for dynamically adjusting the scenario difficulty and complexity based on the comprehensive evaluation results of the site in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be made with reference to the accompanying drawings of the specification.

[0043] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0044] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selectively exclusive embodiment from other embodiments.

[0045] Embodiment 1, refer to Figure 1 and Figure 2, which is the first embodiment of the present invention. This embodiment provides a method and system for testing and evaluating a vehicle-road collaborative scenario, including the following steps:

[0046] S1. The sensor network includes cameras, lidars, and millimeter-wave radars;

[0047] The cameras cover major intersections and crossroads, obtain visual information. The cameras are mainly used to capture image information of static and dynamic objects, providing rich visual data for subsequent image processing and feature extraction;

[0048] The lidars obtain three-dimensional information of the vehicle's surrounding environment. By emitting laser beams and receiving reflected signals, the lidars can accurately measure the distances in the surrounding environment and the positions of objects, providing three-dimensional point cloud data.

[0049] The millimeter-wave radars obtain dynamic information of other traffic participants around the vehicle. The dynamic information includes the distances and speeds of other traffic participants around the vehicle. By emitting and receiving millimeter-wave signals, the millimeter-wave radars can real-time monitor the distances and relative speeds of surrounding objects, and can still maintain a high detection accuracy especially under adverse weather conditions.

[0050] S1.1. The driving state data of the vehicle includes speed, acceleration, steering angle, and braking state;

[0051] Further explanation: The current speed information of the vehicle is used to evaluate the driving state of the vehicle in different scenarios. The acceleration information of the vehicle reflects the situation of the vehicle accelerating or decelerating, which helps to evaluate the power performance of the vehicle. The steering angle information of the vehicle is used to evaluate the behavior of the vehicle when turning. The braking state information of the vehicle includes whether the braking system is activated and the degree of braking, which is used to evaluate the braking performance of the vehicle.

[0052] Perform data cleaning, fill in missing data, and remove high-profile noise on the collected data. By data cleaning, filling in missing data, and removing high-profile noise, the quality and reliability of the data are improved, ensuring that subsequent feature extraction and fusion processing can be based on accurate data. And the robustness of the system is improved, reducing misjudgments and errors caused by data quality problems, making the test results more credible.

[0053] Generally speaking, multi-sensor fusion has many advantages. For example, it can provide comprehensive perception. By combining the use of cameras, lidar, and millimeter-wave radars, it can obtain environmental information around the vehicle from multiple angles and dimensions, providing more comprehensive and accurate perception data. Complementarity means that different sensors are complementary under different conditions. For example, cameras perform well in well-lit environments, while lidar and millimeter-wave radars can still provide stable data at night or in adverse weather conditions. Redundancy means that the configuration of multiple sensors increases the redundancy of the system. Even if a certain sensor fails, other sensors can still continue to provide necessary information, improving the reliability of the system.

[0054] Collecting vehicle driving state data is also an important step, which can achieve precise evaluation. By collecting key driving state data such as vehicle speed, acceleration, steering angle, and braking state, the behavior and performance of the vehicle in different test scenarios can be precisely evaluated. Comprehensive monitoring means that these data not only help evaluate the basic driving performance of the vehicle but also can be used to monitor the dynamic behavior of the vehicle in complex traffic environments, ensuring the comprehensiveness and effectiveness of the test.

[0055] S2. Extract features from the image data and lidar data and perform fusion;

[0056] Load the image file through image processing software, convert the image into a two-dimensional matrix. In this process, the position (x, y) of each pixel corresponds to an element in the matrix, and the value of this element is the image pixel intensity value. After converting the image into a two-dimensional matrix, it is convenient for subsequent mathematical processing and feature extraction. The image pixel intensity can accurately reflect the color or grayscale value of each pixel in the image, providing a basis for subsequent feature extraction.

[0057] S2.1. Extract features from the image data, and its expression is:

[0058]

[0059] Among them, F I represents the image feature, I(x, y) represents the image pixel intensity value. For grayscale images, this may be a value between 0 and 255; for color images, this may be an RGB triple. Ω represents the effective area of the image, that is, the set of all pixel points considered. μ x represents the abscissa of the image center, x represents the abscissa in the pixel coordinates, μ y represents the ordinate of the image center, y represents the ordinate in the pixel coordinates, σ represents the standard deviation of the Gaussian kernel, which is used to control the width of the Gaussian kernel, dx represents the integral of the small increment of x, and dy represents the integral of the small increment of y;

[0060] Furthermore, by performing weighted processing on the image using the Gaussian kernel function, the central region in the image can be highlighted, and the influence of the edge region can be suppressed, thereby better capturing the core features in the image. In addition, the integral operation ensures the continuity and smoothness of feature extraction and improves the robustness of the features.

[0061] S2.2. Feature extraction of the lidar data scene, and its expression is:

[0062]

[0063] Among them, F L represents the feature of the lidar, N represents the number of points in the lidar point cloud, d i represents the distance of the i-th point in the lidar point cloud, r i represents the reflection intensity of the i-th point in the lidar point cloud, represents the average value of the reflection intensities of all points, and σ r represents the standard deviation of the reflection intensity, which determines the width of the Gaussian function;

[0064] Furthermore, through the distance and reflection intensity information of the lidar point cloud, the feature vector F L is extracted. This feature extraction method can effectively capture the key information in the point cloud data, especially the distribution characteristics of the reflection intensity, which is very useful for identifying different types of obstacles. The weighted average operation makes the feature extraction more accurate.

[0065] S2.3. Fuse the features of the image data and the lidar data, and its expression is:

[0066]

[0067] Among them, G(F I , F L ) represents the fused feature vector, A represents the number of image features F I , M represents the number of lidar features F L , (F I ) k represents the k-th element in the image features, and (F L ) j represents the j-th element in the lidar features.

[0068] Furthermore, by fusing the features of the image data and the lidar data, more comprehensive and accurate environmental information can be obtained. The fused feature vector not only contains the visual information in the image but also contains the three-dimensional structure information in the lidar point cloud, thereby improving the description ability of the overall features.

[0069] S3. Build an initial test scenario library based on the fused data;

[0070] By fusing the image data features F I and F L , combined with the vehicle driving state data, an initial test scenario library is built; By fusing multiple sensor data, the diversity and richness of the test scenarios are ensured, enabling the test scenarios to more comprehensively reflect various situations in the actual traffic environment.

[0071] Use machine learning algorithms to evaluate the difficulty level and coverage of each scenario in the initial scenario library, and its expression is:

[0072]

[0073] R(S) = ∑ t∈T ω t ·f t (S);

[0074] Among them, C(S) represents the difficulty level of scenario S, γ represents the importance coefficient of the optimized feature vector, δ represents the importance coefficient of the driving state data, v represents the vehicle speed, max(v) represents the maximum vehicle speed in all scenarios, a represents the vehicle acceleration, max(a) represents the maximum vehicle acceleration in all scenarios, θ represents the vehicle steering angle, max(θ) represents the maximum vehicle steering angle in all scenarios, b represents the vehicle braking state, and the braking state can be a binary variable (for example, 0 represents not braking, 1 represents braking), dS represents the spatial microelement of scenario S, R(S) represents the coverage of scenario S, T represents the set of all possible traffic participant types, t represents an element of the traffic participant type, ω t represents the weight of type t, and f t (S) represents the occurrence frequency of type t in scenario S.

[0075] By evaluating the difficulty level and coverage, the characteristics of each scenario can be quantified, so as to more scientifically select representative test scenarios. This helps to improve the efficiency and accuracy of the test.

[0076] S3.1. Combine the difficulty level and coverage of the scenarios for comprehensive evaluation, and its expression is:

[0077]

[0078] Among them, E(S) represents the comprehensive evaluation of scenario S, max(C(S)) represents the maximum difficulty level in all scenarios, and max(R(S)) represents the maximum coverage in all scenarios;

[0079] Through comprehensive evaluation, the difficulty level and coverage of each scenario can be comprehensively measured to ensure that the selected test scenarios are both challenging and representative;

[0080] Set a venue scoring threshold and dynamically adjust the difficulty and complexity of the test scenarios according to the comprehensive evaluation results;

[0081] Further explanation, Scenario A:

[0082] There are only a small number of straight - running vehicles on the road, without pedestrians and bicycles.

[0083] Vehicle driving speed: 30 km / h.

[0084] Acceleration: 0 m / s 2 (Constant speed driving).

[0085] Steering angle: 0° (Straight driving).

[0086] Braking state: Not braked.

[0087] Comprehensive evaluation results, difficulty level C(A)=0.2, coverage R(A)=0.3;

[0088] Based on the comprehensive evaluation formula:

[0089] Assume that the set venue scoring threshold is 0.7. This means that any scenario with a comprehensive evaluation score lower than 0.7 needs to be adjusted to increase its difficulty or coverage; through dynamic adjustment, ensure that the difficulty and complexity of the test scenarios are appropriate, neither too simple nor too complex, so as to better evaluate the performance of the vehicle in different scenarios.

[0090] Determine the scenarios selected into the optimized test scenario library according to the comprehensive score, and its expression is:

[0091]

[0092] Among them, T represents the optimized test scenario library, Sc represents the set of all possible scenarios, S ′ represents the scenario traversed when calculating the sum of the comprehensive evaluation indicators of all scenarios, η represents the scenario importance coefficient, ζ represents the similarity influence coefficient, and sim(S,T) represents the average similarity of the current scenario S to all scenarios in the test scenario library T.

[0093] The value range of T is between [0,1], where a value close to 1 represents a higher possibility that the scenario is selected into the test scenario library, while a value close to 0 means a greater possibility that the scenario is excluded.

[0094] Further explanation: Through the optimized test scenario library, ensure that each scenario has a certain degree of uniqueness and representativeness, avoid repeated testing of the same type of scenarios, thereby improving the efficiency and effectiveness of testing.

[0095] S4. Select scenarios from the optimized scenario library according to the test plan for execution testing. For each selected test scenario, record the data during the test, including vehicle status information such as speed, acceleration, steering angle, braking status, environmental information such as information about traffic participants (such as pedestrians, other vehicles), road conditions (such as the degree of road surface slipperiness), weather conditions (such as sunny, rainy), and other test-related data, such as the internal system status of the vehicle, communication data, communication delay, sensor status;

[0096] After each test scenario ends, record the key performance indicators of driving performance (response time, throughput, error rate, vehicle safety distance control, emergency braking performance, vehicle smoothness, ride experience), abnormal situations, and any fault points;

[0097] Abnormal situations are to record any unexpected behaviors that occur during the test, such as the vehicle suddenly deviating from the lane or failing to correctly identify traffic signs.

[0098] Fault points are to record any hardware or software failures that occur during the test, such as sensor failures or control system malfunctions.

[0099] After completing all tests, generate a test result analysis report, which includes a test overview, summary of test data, analysis of key performance indicators, analysis of abnormal situations and fault points, conclusions and suggestions.

[0100] S5. Evaluate the vehicle performance based on the test result analysis report and adjust the test scenario library;

[0101] Evaluate the overall performance of the vehicle according to the performance indicators in the test result analysis report, and determine the performance indicators and the change trend of performance of the vehicle under different test scenarios;

[0102] Performance indicators include driving ability, decision-making ability, safety, and comfort.

[0103] The change trend of performance includes performance stability, modification effect, and anomaly detection;

[0104] Evaluate the effectiveness of the scenarios in the current test scenario library according to the test results, and adjust the scenarios in the test scenario library based on the test result analysis report;

[0105] Effectiveness includes scenario coverage, scenario difficulty, and scenario diversity. Scenario coverage refers to evaluating whether the test scenarios cover all key traffic situations and edge cases. Scenario difficulty refers to assessing whether the difficulty level of the test scenarios is appropriate and can effectively distinguish different performance levels of vehicles. Scenario diversity refers to evaluating the diversity of the test scenarios to ensure that different types of traffic participants and environmental conditions are fully considered.

[0106] Scenario adjustment includes adding new scenarios, adjusting existing scenarios, and deleting invalid scenarios. Adding new scenarios means that if certain key situations are not covered, new test scenarios can be designed and added. Adjusting existing scenarios means that if the difficulty level of the existing scenarios is inappropriate, the scenarios can be optimized by adjusting parameters (such as traffic density, weather conditions, etc.). Deleting invalid scenarios means that if certain scenarios do not bring significant test effects, these scenarios can be considered for deletion to reduce unnecessary test burdens.

[0107] Feed the test results and the adjusted test scenarios back to the relevant teams for the next round of testing. Based on the results of each test, continuously adjust and optimize the test scenario library.

[0108] Based on the adjusted test scenario library, organize the next round of testing to ensure that the improvement measures are effectively verified. According to the results of each test, continuously adjust and optimize the test scenario library to form a continuous improvement cycle.

[0109] This embodiment also provides a vehicle-road collaborative scenario test and evaluation system, including:

[0110] A data collection and preprocessing module, responsible for collecting the dynamic information of vehicles, pedestrians, and other traffic participants, and preprocessing the data;

[0111] A feature extraction and fusion module, responsible for extracting features from image and lidar data and performing fusion processing;

[0112] A scenario library construction and evaluation module, constructing an initial test scenario library, evaluating the difficulty level and coverage of the scenarios, and dynamically adjusting the scenario library;

[0113] A test plan execution and analysis module, executing the test plan, recording the test data, and generating a test result analysis report;

[0114] A performance evaluation and scenario library adjustment module, evaluating the vehicle performance and adjusting the test scenario library according to the test results.

[0115] This embodiment also provides a computer device, which is applicable to the situation of the vehicle-road collaborative scenario test and evaluation method and system, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the vehicle-road collaborative scenario test and evaluation method and system as proposed in the above embodiment.

[0116] This computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the outer shell of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0117] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the vehicle-road collaborative scenario test and evaluation method and system as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM for short), Electrically Erasable Programmable Read-Only Memory (EEPROM for short), Erasable Programmable Read-Only Memory (EPROM for short), Programmable Read-Only Memory (PROM for short), Read-Only Memory (ROM for short), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0118] In summary, the present invention extracts and fuses features from image data and lidar data through a Gaussian kernel function, enabling more accurate capture of key information in traffic scenarios. The fused feature vector contains complementary information from image and lidar data, improving the comprehensiveness and descriptive ability of the features and providing richer data support for subsequent construction of a test scenario library. By evaluating the difficulty level and coverage of each scenario in the initial test scenario library through a machine learning algorithm and dynamically adjusting the difficulty and complexity of the test scenarios according to the evaluation results, a test scenario library can be constructed more scientifically. The dynamic adjustment mechanism ensures the diversity and representativeness of the test scenarios, improving the comprehensiveness and effectiveness of the tests. Through the evaluation of the machine learning algorithm, the construction of the test scenarios becomes more intelligent and automated.

[0119] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A test and evaluation method for vehicle-road collaborative scenarios, characterized in that Including: Deploy a sensor network to capture the dynamic information of vehicles, pedestrians and other traffic participants, obtain the driving state data of the vehicle through the sensors carried by the vehicle, and preprocess the collected data; Extract features from the image data and lidar data and fuse them; Build an initial test scenario library based on the fused data, evaluate the difficulty level and coverage of each scenario in the initial scenario library, dynamically adjust the difficulty and complexity of the test scenarios according to the evaluation results, and add the optimized scenarios to the scenario library; Select scenarios from the optimized scenario library according to the test plan to execute the test, record the data during the test, and analyze it to generate a test result analysis report; Evaluate the vehicle performance and adjust the test scenario library based on the test result analysis report.

2. The vehicle-road collaborative scenario test and evaluation method according to claim 1, characterized in that, The sensor network includes cameras, lidar, and millimeter-wave radars; The cameras cover the main intersections and crossroads to obtain visual information, the lidar obtains the three-dimensional information of the vehicle's surrounding environment, and the millimeter-wave radar obtains the dynamic information of other traffic participants around the vehicle. The dynamic information includes the distance and speed of other traffic participants around the vehicle; The driving state data of the vehicle includes speed, acceleration, steering angle, and braking state; Clean the collected data, fill in the missing data, and remove the high-profile noise.

3. The vehicle-road collaborative scenario test and evaluation method according to claim 2, wherein, Extract features from the image data and lidar data and fuse them. The specific steps are as follows: Load the image file through image processing software, convert the image into a two-dimensional matrix, and obtain the image pixel intensity value at each position in the matrix; Extract the scene features of the image data and lidar data; Fuse the features of the image data and lidar data.

4. The vehicle-road collaborative scenario test and evaluation method according to claim 3, wherein, Build an initial test scenario library based on the fused data, and use machine learning algorithms to evaluate the difficulty level and coverage of each scenario in the initial scenario library.

5. The vehicle-road collaborative scenario test and evaluation method according to claim 4, wherein Combine the difficulty level and coverage of each scenario for comprehensive evaluation; Set a site scoring threshold, and dynamically adjust the difficulty and complexity of the test scenarios according to the comprehensive evaluation results; Determine the scenarios selected for the optimized test scenario library according to the comprehensive score.

6. The vehicle-road collaborative scenario test and evaluation method according to claim 5, wherein Select scenarios from the optimized scenario library according to the test plan to perform the test, and record the data during the test for each selected test scenario; Record the key performance indicators, abnormal situations, and any fault points after each test scenario ends, and generate a test result analysis report.

7. The vehicle-road collaborative scenario test and evaluation method according to claim 6, wherein Evaluate the vehicle performance and adjust the test scenario library based on the test result analysis report. The specific steps are as follows: Evaluate the overall performance of the vehicle according to the performance indicators in the test result analysis report, and determine the performance indicators and the changing trend of the performance of the vehicle under different test scenarios; Evaluate the effectiveness of the scenarios in the current test scenario library according to the test results, and adjust the scenarios in the test scenario library based on the test result analysis report; Feed back the test results and the adjusted test scenarios to the relevant teams for the next round of testing. Based on the results of each test, continuously adjust and optimize the test scenario library.

8. A vehicle-road collaborative scenario test and evaluation system, based on the vehicle-road collaborative scenario test and evaluation method and system according to any one of claims 1 to 7, characterized in that, Including: The data acquisition and preprocessing module is responsible for collecting the dynamic information of vehicles, pedestrians and other traffic participants and preprocessing the data; The feature extraction and fusion module is responsible for extracting features from image and lidar data and performing fusion processing; The scenario library construction and evaluation module constructs an initial test scenario library, evaluates the difficulty level and coverage of scenarios, and dynamically adjusts the scenario library; The test plan execution and analysis module executes the test plan, records test data, and generates a test result analysis report; The performance evaluation and scenario library adjustment module evaluates the vehicle performance and adjusts the test scenario library according to the test results.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the vehicle-road collaborative scenario test and evaluation method and system according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the vehicle-road collaborative scenario test and evaluation method and system according to any one of claims 1 to 7 are implemented.

Images