Intelligent networked automobile operation safety actual road evaluation method and device

Abstract

The invention relates to an actual road evaluation method and device for the running safety of an intelligent networked automobile, and relates to the field of actual road testing of the intelligent networked automobile. According to the technical scheme, the test road is evaluated, the test vehicle ODD is matched, the test path is precisely planned, a driver test score deduction mode is used for reference, an evaluation system integrating manual and automatic evaluation is constructed, evaluation is ensured to be comprehensive and objective, variables such as road difficulty and scene randomness are included in the scheme for solving the problem of comparability of the evaluation result, and the evaluation accuracy is improved. Differences are eliminated through normalization processing, and the pain points that actual road evaluation is large in randomness and results are difficult to compare are overcome. Under the condition, transverse comparison of results of different vehicles under different roads and environments is achieved, the fair evaluation requirement of a third-party mechanism is met, data support is provided for a management department to comprehensively analyze the test condition of the whole industry, and standardized development of intelligent network connection vehicle evaluation is assisted.

Description

Technical Field

[0001] This invention relates to the field of real-world road testing technology for intelligent connected vehicles, and particularly to a method and apparatus for evaluating the operational safety of intelligent connected vehicles on real-world roads. Background Technology

[0002] To ensure the operational safety of intelligent connected vehicles in the real-world application phase, the International Automobile Manufacturers Association (OICA) has proposed a "three-pillar" testing and certification system centered on simulation testing, closed-track testing, and real-world road testing. Real-world road testing specifically refers to driving tests of intelligent connected vehicles on open, real roads, verifying the system by recreating complex environments such as real road topology, traffic flow distribution, and weather conditions. Compared to simulation and closed-track testing, this testing mode has irreplaceable technical advantages such as high environmental fidelity and strong dynamic interactivity. However, it also presents practical challenges such as poor controllability of real traffic scenarios, high testing costs, and significant safety risks. Therefore, it is usually considered the final stage of performance verification for intelligent connected vehicles.

[0003] Currently, before officially deploying assisted driving systems (Level 2 driving automation systems), automakers are required to conduct targeted real-world road tests to verify the reliability of the system's functions and performance. Furthermore, pilot units participating in the road testing and demonstration applications of intelligent connected vehicles also conduct real-world road tests to iterate and optimize their autonomous driving systems after obtaining temporary license plates for their test vehicles (equipped with Level 3 or higher driving automation systems). However, existing real-world road tests largely focus on internal product upgrade needs within companies. Their test route selection, scenario settings, and evaluation systems are only applicable to a single model or brand, resulting in test results that cannot be effectively compared across different automakers and products. Summary of the Invention

[0004] The purpose of this invention is to provide a method and apparatus for evaluating the operational safety of intelligent connected vehicles on actual roads, so as to solve the problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for real-world road testing of the operational safety of intelligent connected vehicles, the method comprising: Data was collected and evaluated on the actual road conditions for the tests, and the actual roads were classified. Based on the test operating range (ODD) of the test vehicle, determine the road sections and test scenarios for conducting actual road tests, and set the test start point, end point, and driving route according to test requirements; Tests were conducted on selected actual roads, and vehicle data was collected during the test process using evaluation devices. The accompanying test personnel then conducted a qualitative evaluation of the test process and made a compliance assessment of the test scenario. The test difficulty coefficient is calculated based on the test process records, and the actual road test score of the test vehicle is calculated according to the evaluation results of the testers. By combining the tested road grade and the actual road evaluation score, the performance of the test vehicle on the actual road is comprehensively evaluated and normalized, so as to realize the comparison of test results of different test vehicles on different test roads and under different test environments.

[0006] In some implementations, the process of collecting and evaluating data on the actual road conditions under which the tests are conducted, and classifying the actual roads, includes: Based on the test road data, we extract lane, intersection, and bridge / tunnel information of the actual test road to construct a feature set of the actual test road. Based on the extracted road information, and combined with the statistical analysis of the test roads' challenges such as multi-lane narrowing, uphill and downhill sections, and obstructed views, the road static classification coefficient A is calculated. The formula for calculating the road static classification coefficient A is as follows: + ; Where, x i The weights of the classification coefficients for different road characteristics; n i y represents the number of times the corresponding road feature appears; j The weights of different levels of testing difficulty are assigned to m. j This corresponds to the number of times the test difficulty occurred; Based on the traffic flow distribution of road sections obtained from on-site measurements and the traffic dynamic information of accident-risk road sections identified by experts, the road dynamic classification coefficient B is calculated. By combining the road static grading coefficient A and the road dynamic grading coefficient B, a comprehensive grading coefficient C for the actual test road is formed, and the actual test road level is determined according to the preset coefficient range.

[0007] In some implementations, determining the road sections and test scenarios for actual road testing based on the test operating range (ODD) of the test vehicle, and setting the test start point, end point, and driving route according to test requirements, includes: Based on the test operating range (ODD) of the test vehicle, the types of road features that the test vehicle can test are determined, and feature elements corresponding to the testable road feature types are matched in the actual test road feature set. Among the testable road feature elements obtained through matching, specific road feature elements are extracted to ensure that the specific road feature elements achieve full coverage of the vehicle operating range ODD; Based on the extracted specific road feature elements, a complete test path that meets the test requirements is automatically planned, the starting point, ending point and mileage information of the test path are determined, and a basic test score α is assigned based on the comprehensive grading coefficient C of the actual test road. The formula for calculating the basic test score α is: α = 100C.

[0008] In some implementations, the step of conducting tests on selected actual roads, collecting vehicle data during the testing process using an evaluation device, and having on-board test personnel conduct a qualitative evaluation of the testing process and a compliance assessment of the test scenario's passability includes: A testing device is installed on the test vehicle based on the principle of minimum requirements. The testing device collects operational data during the test process. The operational data includes vehicle position, real-time speed, acceleration, environmental perception data, and road feature data. The control data of the test vehicle is collected through the vehicle bus interface or sensor acquisition method. The control data includes throttle opening, brake pedal travel, steering wheel angle and steering angular velocity. The evaluation device synchronizes the operating data and control data in time, realizes data conversion, synthesis and storage, and monitors the test process data in real time and reviews the test data after the test is completed. The on-board test personnel record the deduction points during the test process and make a compliance assessment of the test vehicle's performance in the test scenario; The evaluation device automatically calculates the deduction points of the test vehicle based on the collected vehicle data, and the test personnel then verify the results. The deduction value β of the test vehicle is calculated by combining the deduction points recorded manually and calculated by the evaluation device.

[0009] In some implementations, the step of calculating the test difficulty coefficient based on the evaluation process records and calculating the test vehicle's actual road test score according to the testers' evaluation results includes: Based on the collected test vehicle data and the test personnel's evaluation records, combined with environmental, weather, traffic flow and emergency impact factors, the test difficulty coefficient D is calculated; Based on the test base score α, the deduction score β, and the test difficulty coefficient D, the actual road test score θ of the test vehicle is calculated. The formula for calculating the test score θ is: θ=D*α-β.

[0010] In some implementations, the step of combining the tested road grade and the actual road evaluation score to comprehensively evaluate and normalize the performance of the test vehicle on actual roads includes: A reference score γ is constructed based on expert experience; The test score θ is normalized according to the reference score γ to obtain the final score λ. The formula for calculating the final score λ is: λ=100*(θ / γ), and the value range of the final score λ is between 0 and 100. A test score database T is constructed, and relevant data from each test is stored in the database. The reference score γ is then updated and iterated based on the database data.

[0011] Secondly, the present invention provides a real-world road testing device for the operational safety of intelligent connected vehicles. This testing device is applied to the aforementioned real-world road testing method for the operational safety of intelligent connected vehicles. The testing device includes: The data acquisition module is used to collect the operating data and control data of the test vehicle during actual road testing. The operating data includes vehicle position, real-time speed, acceleration, environmental perception data and road characteristic data. The control data includes throttle opening, brake pedal travel, steering wheel angle and steering angular velocity. The data processing module is used to perform time synchronization, conversion and synthesis processing on the collected operation data and control data, and to store the processed data. It also enables real-time monitoring of test process data and data review after test. The evaluation module is used to automatically calculate the deduction points of the test vehicle based on the collected vehicle data, and to calculate the test difficulty coefficient in combination with the test environment, weather, traffic flow and the impact of emergencies. The interactive module is used by on-board test personnel to record deduction points during the test process, evaluate the compliance of the test vehicle's performance, and verify and confirm the deduction points calculated by the evaluation module.

[0012] In some implementations, the data acquisition module includes: The positioning unit is used to collect vehicle position, real-time speed, and acceleration data via satellite positioning or inertial navigation. The environmental perception unit is used to collect environmental perception data and road feature data through cameras, lidar and millimeter-wave radar. The road feature data includes lane information, intersection information, bridge and tunnel information and test difficulty information. The vehicle bus interaction unit is used to interact with the test vehicle via the OBD interface or CAN bus interface to collect data such as throttle opening, brake pedal travel, steering wheel angle and steering angular velocity. The data aggregation unit is used to initially integrate the data collected by each unit and transmit it to the data processing module.

[0013] In some embodiments, the data processing module includes: The time synchronization unit is used to align the time of operational and control data based on GPS timestamps or local high-precision clocks to ensure time consistency of data from different sources. The data conversion unit is used to convert collected data in different formats into a unified data format, and to perform data cleaning and synthesis to generate a standardized test dataset. The storage unit uses a combination of local caching and cloud backup to store standardized test datasets and test process logs, supporting long-term data retention and fast retrieval. The monitoring unit is used to display the test data change curve and vehicle operating status in real time. When abnormal data is detected or the vehicle performs dangerous operations, it triggers an early warning. It also supports viewing historical test data by going back through the timeline.

[0014] In some implementations, the evaluation module includes; The deduction calculation unit has a built-in test scoring rule database. Based on the collected vehicle operation data and control data, it automatically identifies violations and calculates deduction values ​​by comparing the scoring rules. The difficulty coefficient calculation unit is used to import the comprehensive grading coefficient of the test road, combine it with real-time collected environmental weather data, traffic flow data and emergency records, and calculate the difficulty coefficient of this test through a preset algorithm. The data output unit integrates the automatically calculated deduction values ​​and difficulty coefficients with the manual evaluation results of the on-board testers to generate preliminary evaluation data containing basic scores, deduction values, and difficulty coefficients, and transmits it to an external scoring system to complete the final score calculation and normalization process.

[0015] In some implementations, the interaction module includes: The manual recording unit is used for testers to enter subjective deduction points and scenario passability evaluations during the testing process; The result verification unit is used to verify, correct and confirm the deduction points automatically calculated by the evaluation module; The information display unit is used to present the test progress, vehicle status, and preliminary evaluation results in real time. The instruction input unit is used by testers to trigger operations such as data saving and pausing the test.

[0016] The beneficial effects of the technical solution provided by this invention include at least the following: This technical solution precisely plans test routes by evaluating test roads and matching test vehicle ODDs. Drawing inspiration from driver's license exam point deduction models, it constructs an evaluation system that integrates manual and automated assessments, ensuring comprehensive and objective evaluation. Addressing the challenge of comparability of test results, this solution incorporates variables such as road difficulty and scenario randomness, eliminating discrepancies through normalization. This overcomes the pain point of high randomness and difficulty in comparing results in real-world road testing. In this context, it enables horizontal comparison of results from different vehicles under varying road conditions, meeting the impartial testing needs of third-party organizations and providing data support for management departments to comprehensively analyze industry-wide testing, thus contributing to the standardization of intelligent connected vehicle testing. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0018] Figure 1 The diagram shows a flowchart of a method for evaluating the operational safety of intelligent connected vehicles on actual roads, provided by an exemplary embodiment of the present invention.

[0019] Figure 2 The diagram shows a structural block diagram of a real-road testing device for the operational safety of intelligent connected vehicles provided by an exemplary embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0022] Figure 1 The diagram illustrates a flowchart of a method for evaluating the operational safety of intelligent connected vehicles on real-road, according to an exemplary embodiment of the present invention. This method includes: Step 101: Collect and evaluate data on the actual road conditions for the test, and classify the actual roads.

[0023] In some embodiments, data collection and evaluation are performed on the actual road conditions under which the tests are conducted, and the actual roads are classified, including: Based on the test road data, we extract lane, intersection, and bridge / tunnel information of the actual test road to construct a feature set of the actual test road. Based on the extracted road information, and combined with the statistical analysis of the test roads' challenges such as multi-lane narrowing, inclines and declines, and obstructed views, the road static grading coefficient A is calculated. The formula for calculating the road static grading coefficient A is as follows: + ; Where, x i The weights of the classification coefficients for different road characteristics; n i y represents the number of times the corresponding road feature appears; j The weights of different levels of testing difficulty are assigned to m. j This corresponds to the number of times the test difficulty occurred; Based on the traffic flow distribution of road sections obtained from on-site measurements and the traffic dynamic information of accident-risk road sections identified by experts, the road dynamic classification coefficient B is calculated. By combining the static road classification coefficient A and the dynamic road classification coefficient B, a comprehensive classification coefficient C for the actual test road is formed, and the actual test road level is determined according to the preset coefficient range.

[0024] In this embodiment, a dataset is constructed by extracting features such as lanes and intersections, making road attributes visible and quantifiable. Static coefficient A focuses on inherent road challenges, while dynamic coefficient B incorporates real-time variables such as traffic flow and accident risk. The combined coefficient C achieves a comprehensive measurement of both fixed road difficulty attributes and dynamic scenarios. This grading method ensures that test sections are matched to vehicle capabilities, and clearly defining road levels avoids distortion of evaluation results due to differences in road conditions.

[0025] Step 102: Based on the test operating range (ODD) of the test vehicle, determine the road sections and test scenarios for conducting actual road tests, and set the test start point, end point, and driving route according to test requirements.

[0026] In some embodiments, based on the test operating range (ODD) of the test vehicle, the road segments and test scenarios for conducting actual road tests are determined, and the test start point, end point, and driving route are set according to test requirements, including: Based on the test operating range (ODD) of the test vehicle, the types of road features that the test vehicle can test are determined, and feature elements corresponding to the testable road feature types are matched in the actual test road feature set. Among the testable road feature elements obtained through matching, specific road feature elements are extracted to ensure that the specific road feature elements achieve full coverage of the vehicle operating range ODD; Based on the extracted specific road feature elements, the system automatically plans a complete test path that meets the test requirements, determines the start point, end point, and mileage information of the test path, and assigns a basic test score α based on the comprehensive grading coefficient C of the actual test road. The formula for calculating the basic test score α is: α = 100C.

[0027] In this embodiment, by clearly defining the road features that the vehicle can adapt to, matching elements are selected from the constructed road feature set to avoid test scenarios that are too difficult for the vehicle's capabilities or too easy to verify performance. When extracting feature elements, full coverage of the ODD (Optical Distribution Controller) is emphasized to ensure that the performance of the vehicle's core operating range can be evaluated, eliminating test blind spots. Automatic path planning improves testing efficiency and standardization, avoiding the subjectivity of manual route selection. Furthermore, assigning a base score α based on a road classification coefficient directly links road difficulty to the scoring system, making test scores on different roads comparable.

[0028] Step 103: Conduct tests on selected actual roads, collect vehicle data during the test using the evaluation device, and have the accompanying test personnel conduct a qualitative evaluation of the test process and make a compliance assessment of the test scenario.

[0029] In some embodiments, tests are conducted on selected actual roads, vehicle data is collected during the test using an evaluation device, and on-board test personnel conduct a qualitative evaluation of the test process and a compliance assessment of the test scenario's passability, including: Testing devices are installed on the test vehicle based on the principle of minimum requirements. The testing devices collect operational data during the test process. The operational data includes vehicle position, real-time speed, acceleration, environmental perception data, and road feature data. The control data of the test vehicle is collected through the vehicle bus interface or sensor acquisition method. The control data includes throttle opening, brake pedal travel, steering wheel angle and steering angular velocity. The testing device synchronizes the operation data and control data in time, realizes data conversion, synthesis and storage, and monitors the test process data in real time and reviews the test data after the test is completed. The on-board test personnel record the deduction points during the test process and make a compliance assessment of the test vehicle's performance in the test scenario; The testing device automatically calculates the deduction points for the test vehicle based on the collected vehicle data, and the test personnel then verify the results. The deduction value β of the test vehicle is calculated by combining the deduction points recorded manually and calculated by the testing device.

[0030] In this embodiment, the principle of minimizing the installation of testing devices is followed to avoid excessive modifications that could interfere with the vehicle's original performance. Comprehensive collection of operational and control data, coupled with time-synchronized processing, constructs a complete data chain of the vehicle's dynamic performance. The testing device's real-time monitoring and post-event review functions provide a basis for verifying disputed points and reduce judgment bias. The testing device objectively calculates and quantifies deduction points, avoiding subjective bias; testers compensate for equipment limitations by completing qualitative evaluations such as scenario passability and deduction verification, ensuring both accuracy and comprehensiveness in the judgment. The final synthesized deduction value β transforms the vehicle's test performance into a quantitative indicator.

[0031] Step 104: Calculate the test difficulty coefficient based on the evaluation process record, and calculate the actual road test score of the test vehicle according to the evaluation results of the testers.

[0032] In some embodiments, the test difficulty coefficient is calculated based on the evaluation process record, and the test vehicle's actual road test score is calculated according to the testers' evaluation results, including: Based on the collected test vehicle data and the test personnel's evaluation records, combined with environmental, weather, traffic flow and emergency impact factors, the test difficulty coefficient D is calculated; Combining the base score α, the deduction score β, and the test difficulty coefficient D, the actual road test score θ of the test vehicle is calculated. The formula for calculating the test score θ is: θ=D*α-β.

[0033] In this embodiment, a dynamic difficulty adjustment mechanism is introduced to make the vehicle test performance score more closely reflect the complexity of real-world scenarios. The difficulty coefficient D is not considered from a single dimension, but rather integrates dynamic variables such as environment and weather with random factors such as traffic flow and unexpected events. This overcomes the limitations of relying solely on road classification (base score α) to measure test difficulty, accurately reproducing the actual challenges during testing. The evaluation score θ adjusts the base score using the difficulty coefficient D, making the base score more differentiated in complex scenarios; then, deductions β, reflecting practical vehicle operation issues, are deducted, achieving a dual calibration of scenario difficulty adaptation and performance quantification. In this case, the one-sidedness of simple scoring is avoided, allowing the score θ to reflect the vehicle's true capabilities in specific complex scenarios.

[0034] Step 105: Combining the tested road grade and the actual road evaluation score, comprehensively evaluate and normalize the performance of the test vehicle on the actual road, so as to achieve a comparison of the test results of different test vehicles on different test roads and under different test environments.

[0035] In some embodiments, the performance of the test vehicle on actual roads is comprehensively evaluated and normalized by combining the tested road grade and the actual road assessment score, including: A reference score γ is constructed based on expert experience; The test score θ is normalized according to the reference score γ to obtain the final score λ. The formula for calculating the final score λ is: λ=100*(θ / γ), and the value range of the final score λ is between 0 and 100. Construct a test score database T, store relevant data from each test into the database, and update and iterate the reference score γ based on the database data.

[0036] In this embodiment, the evaluation value is maximized through comprehensive evaluation and normalization. Anchored to road grade and preliminary evaluation score, it breaks down the scoring barriers caused by different test scenarios and road difficulties. The reference score γ, constructed in conjunction with expert experience, provides an authoritative benchmark for score calibration, while the final score λ unifies the diverse preliminary scores to a standard dimension of 0-100 points, providing a common language for direct comparison of different vehicle performances. More importantly, the construction of the test database T forms a dynamic optimization closed loop. Each input of test data provides empirical support for the iteration of γ, ensuring the scoring standard continuously improves with technological advancements. This not only meets the needs of impartial third-party evaluations but also makes industry-specific analyses by management departments more scientific and timely.

[0037] Figure 2 This diagram illustrates a structural block diagram of a real-road testing device for the operational safety of intelligent connected vehicles, provided by an exemplary embodiment of the present invention. This device is applied in the aforementioned real-road testing method for the operational safety of intelligent connected vehicles. The testing device includes: The data acquisition module 201 is used to collect the operating data and control data of the test vehicle during actual road testing. The operating data includes vehicle position, real-time speed, acceleration, environmental perception data and road characteristic data. The control data includes throttle opening, brake pedal travel, steering wheel angle and steering angular velocity. The data processing module 202 is used to perform time synchronization, conversion and synthesis processing on the collected running data and control data, and to store the processed data. It also enables real-time monitoring of test process data and data playback after test. The evaluation module 203 is used to automatically calculate the deduction points of the test vehicle based on the collected vehicle data, and to calculate the test difficulty coefficient in combination with the test environment, weather, traffic flow and the impact of emergencies. The interaction module 204 is used for accompanying test personnel to record deduction points during the test process, evaluate the compliance of the test vehicle's performance, and verify and confirm the deduction points calculated by the evaluation module.

[0038] In some embodiments, the data acquisition module 201 includes: The positioning unit 2011 is used to collect vehicle position, real-time speed, and acceleration data via satellite positioning or inertial navigation. The environmental perception unit 2012 is used to collect environmental perception data and road feature data through cameras, lidar and millimeter-wave radar. The road feature data includes lane information, intersection information, bridge and tunnel information and information on testing difficulties. The vehicle bus interaction unit 2013 is used to interact with the test vehicle via the OBD interface or CAN bus interface to collect data such as throttle opening, brake pedal travel, steering wheel angle and steering angular velocity. The data aggregation unit 2014 is used to initially integrate the data collected by each unit and transmit it to the data processing module.

[0039] In some embodiments, the data processing module 202 includes: The time synchronization unit 2021 is used to time-align operational and control data based on GPS timestamps or local high-precision clocks to ensure time consistency of data from different sources. The data conversion unit 2022 is used to convert collected data in different formats into a unified data format, and to perform data cleaning and synthesis to generate a standardized test dataset. Storage Unit 2023 uses a combination of local caching and cloud backup to store standardized test datasets and test process logs, supporting long-term data retention and fast retrieval. The monitoring unit 2024 is used to display the test data change curve and vehicle operating status in real time. When abnormal data is detected or the vehicle performs dangerous operations, it triggers an early warning prompt and supports viewing historical test data through the timeline.

[0040] In some embodiments, the evaluation module 203 includes; The deduction calculation unit 2031 has a built-in test scoring rule database. Based on the collected vehicle operation data and control data, it automatically identifies violations and calculates deduction values ​​by comparing the scoring rules. The difficulty coefficient calculation unit 2032 is used to import the comprehensive classification coefficient of the test road, and combine it with real-time collected environmental weather data, traffic flow data and emergency records to calculate the difficulty coefficient of this test through a preset algorithm. The data output unit 2033 is used to integrate the automatically calculated deduction value and difficulty coefficient with the manual evaluation results of the on-board test personnel to generate preliminary evaluation data containing the basic score, deduction value and difficulty coefficient, and transmit it to the external scoring system to complete the final score calculation and normalization processing.

[0041] In some embodiments, the interaction module 204 includes: The manual recording unit 2041 is used for testers to enter subjective deduction points and scenario passability evaluations during the test process; The result verification unit 2042 is used to verify, correct and confirm the deduction points automatically calculated by the evaluation module 203. Information display unit 2043 is used to present test progress, vehicle status and preliminary evaluation results in real time; The instruction input unit 2044 is used by testers to trigger operations such as data saving and pausing the test.

[0042] It should be noted that the intelligent connected vehicle operation safety real-road evaluation device provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0043] It is understood that the specific examples in this document are only intended to help those skilled in the art better understand this disclosure, and are not intended to limit the scope of the invention.

[0044] It is understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of this disclosure.

[0045] It is understood that the various implementation methods described in this specification can be implemented individually or in combination, and this disclosure does not limit them.

[0046] Unless otherwise stated, all technical and scientific terms used in this disclosure have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this specification. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items. The singular forms "a," "the," and "the" as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0047] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this specification.

[0048] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method implementation, and will not be repeated here.

[0049] In the several embodiments provided in this specification, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the embodiments of the apparatus described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0050] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0051] In addition, the functional units in the various embodiments of this specification can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0052] The above description is merely a specific embodiment of this specification, but the scope of protection of this invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this specification should be included within the scope of protection of this specification. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for measuring and evaluating the actual road operation safety of intelligent networked vehicles, characterized in that, The method comprises: Data collection and evaluation of actual road conditions for testing, and grading of actual roads; According to the test operating range ODD of the test vehicle, determining the road sections and test scenarios for actual road testing, and setting the test starting point, ending point and driving path according to the test requirements; Carrying out the test on the selected actual road, collecting vehicle data during the test by the evaluation device, and qualitatively evaluating the test process by the on-board test personnel, and judging the compliance of the test scenario passability; According to the test difficulty coefficient calculated from the evaluation process record, and the test vehicle score calculated according to the test personnel's evaluation results; Combining the tested road grade and the actual road evaluation score, the performance of the test vehicle on the actual road is comprehensively evaluated and normalized, and the test results of different test vehicles on different test roads and in different test environments are compared. 2.The intelligent networked vehicle operation safety actual road test evaluation method according to claim 1, characterized in that, The data collection and evaluation of actual road conditions for testing, and grading of actual roads, comprises: Based on the test road data, the lane, intersection and bridge and tunnel road information of the actual test road are extracted, and the actual test road feature set is constructed; According to the extracted road information, combined with the statistical test road test difficulties such as multi-lane narrowing, uphill and downhill, and visual obstruction, the road static grading coefficient A is calculated, and the calculation formula of the road static grading coefficient A is: + ； wherein x i is a grading coefficient weight for different road features; n i is the number of occurrences of the corresponding road feature; y j is a grading influence weight for different test difficulties, m j is the number of occurrences of the corresponding test difficulty; According to the traffic flow distribution obtained by field measurement and the dynamic information of the accident risk road section identified by experts, the road dynamic grading coefficient B is calculated; Combining the road static grading coefficient A and the road dynamic grading coefficient B, the actual test road comprehensive grading coefficient C is formed, and the actual test road level is determined according to the preset coefficient interval. 3.The intelligent networked vehicle operation safety real road test evaluation method according to claim 2, characterized in that, According to the test operating range ODD of the test vehicle, the road feature type that can be tested by the test vehicle is determined, and the feature elements corresponding to the testable road feature type in the actual test road feature set are matched; In the matched testable road feature elements, specific road feature elements are extracted to ensure that the specific road feature elements realize full coverage of the vehicle operating range ODD; According to the extracted specific road feature elements, a complete test path meeting the test requirements is automatically planned, the starting point, ending point and mileage information of the test path are determined, and a test basic score α is given based on the actual test road comprehensive grading coefficient C, and the calculation formula of the test basic score α is: α = 100C. The test on the selected actual road, collecting vehicle data during the test by the evaluation device, and qualitatively evaluating the test process by the on-board test personnel, and judging the compliance of the test scenario passability, comprises: 4.The method of claim 3, wherein, The evaluation device is installed on the test vehicle based on the minimum principle, and the running data during the test is collected by the evaluation device, including vehicle position, real-time speed, acceleration, environmental perception data and road feature data; ​ The control data of the test vehicle is collected through a vehicle bus interface or a sensor collection mode, and the control data includes an accelerator opening degree, a brake pedal stroke, a steering wheel rotation angle, and a steering angular velocity; The operation data and the control data are time-synchronized by the evaluation device, data conversion, synthesis, and storage are realized, the test process data is monitored in real time by the evaluation device, and the test data is reviewed after the test is completed; The test personnel on the vehicle record the deduction points in the test process, and the performance of the test vehicle in the test scene is judged for compliance; The evaluation device automatically calculates the deduction points of the test vehicle according to the collected vehicle data, and the deduction points are reviewed by the test personnel; The deduction value β of the test vehicle is calculated by combining the manually recorded points and the deduction points calculated by the evaluation device. 5.The intelligent networked vehicle operation safety real road test evaluation method according to claim 4, characterized in that, The test difficulty coefficient is calculated according to the evaluation process record, and the actual road evaluation score of the test vehicle is calculated according to the judgment result of the test personnel, including: The test difficulty coefficient D is calculated according to the collected test vehicle data and the evaluation record of the test personnel, combined with environmental, weather, traffic flow, and unexpected event influencing factors; The test vehicle actual road evaluation score θ is calculated by combining the test base score α, the deduction value β, and the test difficulty coefficient D, and the calculation formula of the evaluation score θ is: θ=D*α-β. 6.The intelligent networked vehicle operation safety real road test evaluation method according to claim 5, characterized in that, The performance of the test vehicle on the actual road is comprehensively judged and normalized by combining the tested road level and the actual road evaluation score, including: The reference score γ is constructed according to expert experience; The final score λ is obtained by normalizing the evaluation score θ of this test according to the reference score γ, and the calculation formula of the final score λ is: λ=100*(θ / γ), and the numerical range of the final score λ is between 0 and 100; The test score database T is constructed, the related data of each test is stored in the database, and the reference score γ is updated and iterated based on the database data.

7. An intelligent networked vehicle operation safety actual road evaluation device, characterized in that, The evaluation device is applied to the intelligent connected vehicle operation safety actual road evaluation method in any one of claims 1 to 6, and the evaluation device includes: The data acquisition module is used to collect the operation data and the control data of the test vehicle during the actual road test, and the operation data includes vehicle position, real-time speed, acceleration, environmental perception data, and road feature data, and the control data includes accelerator opening degree, brake pedal stroke, steering wheel rotation angle, and steering angular velocity; The data processing module is used to time-synchronize, convert, and synthesize the collected operation data and control data, store the processed data, and realize real-time monitoring of the test process data and data review after the test is completed; The evaluation module is used to automatically calculate the deduction points of the test vehicle according to the collected vehicle data, and calculate the test difficulty coefficient combined with the test environment, weather, traffic flow, and unexpected event influencing factors; The interactive module is used for the test personnel on the vehicle to record the deduction points in the test process, judge the performance of the test vehicle for compliance, and review and confirm the deduction points calculated by the evaluation module. 8.The device for measuring and evaluating the running safety of intelligent networked vehicles on actual roads according to claim 7, characterized in that, The data acquisition module includes: A positioning unit is configured to collect vehicle position, real-time speed and acceleration data through satellite positioning or inertial navigation; An environment perception unit is configured to collect environment perception data and road feature data through cameras, laser radars and millimeter wave radars, wherein the road feature data includes lane information, intersection information, bridge and tunnel information and test difficulty information; A vehicle bus interaction unit is configured to interact with a test vehicle through an OBD interface or a CAN bus interface to collect throttle opening, brake pedal stroke, steering wheel rotation angle and steering angle speed data; A data aggregation unit is configured to preliminarily integrate the data collected by each unit and transmit the data to a data processing module. 9.The device for measuring and evaluating the running safety of intelligent networked vehicles on actual roads according to claim 7, characterized in that, The data processing module includes: A time synchronization unit is configured to perform time alignment on the operation data and the control data based on a GPS timestamp or a local high-precision clock to ensure the time consistency of data from different sources; A data conversion unit is configured to convert the collected data in different formats into a unified data format, perform data cleaning and synthesis, and generate a standardized test data set; A storage unit is configured to store the standardized test data set and a test process log in a manner of combining local caching with cloud backup, and support long-term storage and quick retrieval of data; A monitoring unit is configured to display a test data change curve and a vehicle operation state in real time, trigger a warning prompt when detecting data anomalies or dangerous operations of the vehicle, and support historical test data review through a time axis. 10.The device for measuring and evaluating the running safety of intelligent networked vehicles on actual roads according to claim 7, characterized in that, The evaluation module includes: A deduction calculation unit is configured to have a test score rule database built-in, automatically identify rule violations and calculate deduction values based on the collected vehicle operation data and control data in comparison with the score rules; A difficulty coefficient calculation unit is configured to import a comprehensive classification coefficient of a test road, combine real-time collected environment weather data, traffic flow data and emergency event records, and calculate a difficulty coefficient of the current test through a preset algorithm; A data output unit is configured to integrate the automatically calculated deduction values and difficulty coefficients with manual evaluation results of a test personnel on board, generate preliminary evaluation data including a basic score value, a deduction value and a difficulty coefficient, and transmit the data to an external scoring system to complete final score calculation and normalization processing; The interaction module includes: An artificial record unit is configured to allow the test personnel to input subjective deduction points and scene passability evaluations in a test process; A result review unit is configured to check, correct and confirm the deduction points automatically calculated by the evaluation module; An information display unit is configured to display a test progress, a vehicle state and a preliminary evaluation result in real time; An instruction input unit is configured to allow the test personnel to trigger data saving, test pausing and other operations.

Images