Commercial vehicle safety performance evaluation method based on auxiliary driving

Abstract

The invention relates to the technical field of commercial vehicle safety performance testing, in particular to a commercial vehicle safety performance evaluation method based on auxiliary driving, and the method comprises the steps: determining an auxiliary driving function library of a commercial vehicle; determining the form weight and proportion of each auxiliary driving function; determining a test scene of each auxiliary driving function, and determining the weight of each test scene; determining an auxiliary driving function of the to-be-tested commercial vehicle based on an auxiliary driving function library; for any auxiliary driving function, testing in the corresponding test scenes to obtain a test result, and determining a scene score corresponding to the auxiliary driving function based on the test result and the test scene weight; determining a single function score of each auxiliary driving function of the to-be-tested commercial vehicle based on the form weight, the proportion and the scene score; and determining the score of the commercial vehicle to be tested based on the single function score of each auxiliary driving function so as to evaluate the safety of the commercial vehicle. The method has the effect of comprehensively and accurately evaluating the safe driving performance of the commercial vehicle.

Description

Technical Field

[0001] This application relates to the technical field of commercial vehicle safety performance testing, and particularly to a method for evaluating the safety performance of commercial vehicles based on assisted driving. Background Art

[0002] As an important part of intelligent vehicles, Advanced Driver Assistance Systems (ADAS) have developed rapidly in recent years. These systems use devices such as sensors and cameras to continuously sense the surrounding environment of the vehicle and process and analyze the sensed data through algorithms to provide assisted driving functions for drivers. Commercial vehicles, as the main force in road transportation, the development of their assisted driving functions is of great significance for improving transportation efficiency and ensuring driving safety. With the wide application of assisted driving systems in commercial vehicles, it is particularly important to accurately and comprehensively test and evaluate their safety performance. This can not only ensure the driving safety of vehicles in complex road environments but also provide optimized and improved assisted driving system solutions for vehicle manufacturers in the research and development of vehicle models.

[0003] Traditional commercial vehicle safety performance testing methods often rely on on-vehicle testing, which has problems such as high testing costs, long testing cycles, and limited testing environments. A method for testing and evaluating the safety performance of assisted driving vehicles based on scenarios disclosed in relevant documents includes: selecting multiple target assisted driving functions available in the current assisted driving vehicle and calculating the function configuration score of the current assisted driving vehicle; separately testing a single assisted driving function of the current assisted driving vehicle to obtain the single-item test performance score of the current assisted driving vehicle; simultaneously testing multiple assisted driving functions of the current assisted driving vehicle to obtain the overall vehicle test comprehensive score of the current assisted driving vehicle; calculating the final safety performance test score based on the function configuration score, single-item test performance score, and overall vehicle test comprehensive score of the current assisted driving vehicle.

[0004] Regarding the above solution, there is a problem that it needs to cooperate with comprehensive testing, and it is necessary to develop a more accurate and simple method for testing and evaluating the safety performance of commercial vehicles. Summary of the Invention

[0005] In order to comprehensively and accurately evaluate the safe driving performance of commercial vehicles, the purpose of this application is to provide a method for evaluating the safety performance of commercial vehicles based on assisted driving.

[0006] The method for evaluating the safety performance of commercial vehicles based on assisted driving provided by this application adopts the following technical solutions:

[0007] A method for evaluating the safety performance of commercial vehicles based on assisted driving includes the following steps:

[0008] Determine the assisted driving function library of a commercial vehicle, where the assisted driving function library includes multiple assisted driving functions;

[0009] Determine the form weights of each assisted driving function according to the impact of each assisted driving function on driving safety;

[0010] Determine the proportion of each assisted driving function according to the Delphi method, where the sum of the proportions of all assisted driving functions is 1;

[0011] Determine the test scenarios of each assisted driving function and determine the weights of each test scenario;

[0012] Determine the assisted driving functions of the commercial vehicle to be tested based on the assisted driving function library;

[0013] For any assisted driving function, test results are obtained under the corresponding test scenarios, and the scenario scores of the corresponding assisted driving function are determined based on the test results and the weights of the test scenarios;

[0014] Determine the single-function scores of each assisted driving function of the commercial vehicle to be tested based on the form weights, proportions, and scenario scores;

[0015] Determine the score of the commercial vehicle to be tested based on the single-function scores of each assisted driving function to evaluate its safety.

[0016] Optionally, the determination of the assisted driving function library includes:

[0017] Based on the proportion of commercial vehicle traffic accidents exceeding the first threshold, determine the types of commercial vehicle accidents;

[0018] Determine the assisted driving function corresponding to any type of commercial vehicle accident;

[0019] The assisted driving functions corresponding to all types of commercial vehicle accidents constitute the assisted driving function library.

[0020] Optionally, the determination of the assisted driving function corresponding to any type of commercial vehicle accident includes:

[0021] Based on the type of commercial vehicle accident and the effect of the assisted driving function, determine the set of assisted driving functions corresponding to the type of commercial vehicle accident;

[0022] Statistically calculate the configuration rate of the assisted driving functions of the accident entities corresponding to the type of commercial vehicle accident, and eliminate the assisted driving functions in the set of assisted driving functions whose configuration rate exceeds the second threshold. The remaining assisted driving functions constitute the assisted driving function corresponding to the type of commercial vehicle accident.

[0023] Optionally, the determination of the form weights of each assisted driving function according to the impact of each assisted driving function on driving safety includes:

[0024] Determine the action forms of each assisted driving function, where the action forms include control type, warning type, and assistance type;

[0025] Determine the form weights according to the action forms.

[0026] Optionally, the determining the proportion of each assisted driving function according to the Delphi method includes:

[0027] Determine at least 10 experts in the field of assisted driving;

[0028] Each expert gives the proportion according to the contribution degree of each assisted driving function to the driving safety;

[0029] For any assisted driving function, remove one maximum value, remove one minimum value, and average the mean of the remaining experts to obtain the proportion of the corresponding assisted driving function.

[0030] Optionally, the determining the test scenarios of each assisted driving function includes:

[0031] Scenario settings under normal working conditions and failure conditions.

[0032] Optionally, the test results include pass and fail, and the test scores for pass and fail are respectively recorded as 1 and 0;

[0033] The determining the scenario score of the corresponding assisted driving function based on the test results and the test scenario weights includes:

[0034] For any assisted driving function, calculate the product of each test scenario weight and the test score, and then sum to obtain the scenario score of the corresponding vehicle assisted driving function.

[0035] Optionally, the determining the score of each assisted driving function based on the form weight, proportion, and scenario score includes:

[0036] Calculate the product of the form weight, proportion, and scenario score corresponding to each assisted driving function to obtain the score of the corresponding assisted driving function.

[0037] Optionally, the determining the score of the commercial vehicle based on the first function score of each assisted driving function to evaluate its safety includes:

[0038] Calculate the product of the form weight and proportion of each assisted driving function in the assisted driving function library, and then sum, denoted as C 10 ;

[0039] Calculate the sum of the single function scores of each assisted driving function of the commercial vehicle to be tested, denoted as C 11 ;

[0040] Determine the score of the commercial vehicle to be tested, denoted as C = (C 11 * 100) / C10 。

[0041] Optionally, determining the score of a commercial vehicle based on the single - function scores of each assisted driving function to evaluate its safety includes:

[0042] Based on the accident type, calculate the product of the formal weight and the proportion of any assisted driving function in the assisted driving function library, and then sum them up to obtain the theoretical score of each accident type;

[0043] For any accident type in the assisted driving function library, calculate the sum of the product of the accident type proportion and the theoretical score of the corresponding accident type, and denote it as C 20 ;

[0044] Based on the accident type, group and sum the single - function scores of the commercial vehicle to be tested to obtain the actual score of each accident type;

[0045] For any accident type of the commercial vehicle to be tested, calculate the sum of the product of the accident type proportion and the actual score of the corresponding accident type, and denote it as C 21 ;

[0046] Determine the score of the commercial vehicle to be tested, denoted as C=(C 21 * 100) / C 20 。

[0047] In summary, the present application includes at least one of the following beneficial technical effects:

[0048] 1. By constructing an assisted driving function library through the cooperation of accident types and configuration rates, the construction accuracy of the assisted driving functions for evaluation is improved; for each assisted driving function, test scenarios are comprehensively considered, making the evaluation results closer to the real situation and being able to more accurately evaluate the safety performance of commercial vehicles;

[0049] Meanwhile, during the process of constructing the assisted driving function library, it can effectively identify which assisted driving functions perform poorly or have potential safety hazards in actual applications, thereby guiding commercial vehicle manufacturers to conduct targeted optimization and improvement.

[0050] 2. By evaluating the performance of each assisted driving function from aspects such as formal weight, proportion, and scenario score, a comprehensive and accurate evaluation of the assisted driving functions of commercial vehicles is carried out. While simplifying the evaluation steps, the evaluation accuracy is improved;

[0051] Furthermore, on the basis of evaluating the performance of each assisted driving function from aspects such as formal weight, proportion, and scenario score, in cooperation with accident types, a comprehensive evaluation of each assisted driving function is carried out. From single to comprehensive, the evaluation results are regulated, further improving the accuracy of the single - assisted - function test evaluation.

[0052] 3. In view of the difficult situation of obtaining insurance for commercial vehicles, test commercial vehicles with assisted driving functions to evaluate their safety performance, provide scientific and objective evaluation results of commercial vehicle safety performance for regulatory agencies, and provide strong evidence for formulating relevant regulations and standards; if the vehicle safety is relatively high, insurance companies can refer to the test evaluation results; if the compensation risk is relatively low, provide auxiliary technical opinions for insurance companies to make investment decisions on commercial vehicle insurance, appropriately reduce the insurance premium, promote a good cooperation relationship between vehicle manufacturers and insurance companies, promote the healthy development of the industry, and at the same time encourage automobile manufacturers to assemble and improve functions, reducing the incidence of traffic accidents. Brief Description of the Drawings

[0053] Figure 1 is a flowchart of the commercial vehicle safety performance evaluation in Embodiment 1 of the present application.

[0054] Figure 2 is a flowchart of determining the assisted driving function library of commercial vehicles in Embodiment 1 of the present application Detailed Description of the Embodiments

[0055] The following further describes the present application in detail with reference to the Figure 1 drawings.

[0056] Embodiment 1

[0057] Embodiment 1 of the present application discloses a method for evaluating the safety performance of commercial vehicles based on assisted driving. Referring to Figure 1 the drawings, the method for evaluating the safety performance of commercial vehicles based on assisted driving includes the following steps:

[0058] S1. Determine the assisted driving function library of commercial vehicles.

[0059] In one embodiment, referring to Figure 2 the drawings, determining the assisted driving function library of commercial vehicles includes:

[0060] S10. Based on the types of commercial vehicle accidents with a statistical proportion in commercial vehicle traffic accidents exceeding a first threshold.

[0061] In this embodiment, a commercial vehicle traffic accident database is constructed, which contains at least 500 commercial vehicle traffic accidents. Select at least 500 commercial vehicle traffic accident data from the commercial vehicle traffic accident database in the order from the most recent to the earliest, or randomly, or a combination of both.

[0062] Analyze the types of accidents corresponding to the above 500 commercial vehicle traffic accident data, count the number of each type of accident, and calculate the proportion of each type of accident.

[0063] In one specific embodiment, randomly select 500 commercial vehicle traffic accident data, analyze the number of accident types, and calculate the proportion of accident types as shown in Table 1 below.

[0064] Data Analysis of 500 Commercial Vehicle Traffic Accidents in Table 1

[0065] Accident type Quantity Proportion Front view 79 15.8％ Rear view 9 1.8％ Side view 28 5.6％ Delayed emergency braking 241 48.2％ Insufficient emergency steering 69 13.8％ Scratching during oncoming vehicle passing 25 5％ Lane change 24 4.8％ Scratching during reverse driving 7 1.4％ Driver's attention 18 3.6％

[0066] The first threshold is determined according to the actual situation. In one specific embodiment, the first threshold is set to 1%. Based on the above, it is determined that the commercial vehicle accident types include 9 accident types: forward vision, rear vision, side vision, untimely emergency braking, insufficient emergency steering, vehicle scratching during oncoming vehicle passing, lane change, vehicle scratching during reversing, and driver distraction.

[0067] S11. Determine the assisted driving function corresponding to any commercial vehicle accident type.

[0068] In this embodiment, determining the assisted driving function corresponding to any commercial vehicle accident type can be determined by matching according to the commercial vehicle accident type and the effect of the assisted driving function.

[0069] In one embodiment, determining the assisted driving function corresponding to any commercial vehicle accident type includes:

[0070] S110. Determine the assisted driving function set corresponding to the commercial vehicle accident type based on the commercial vehicle accident type and the effect of the assisted driving function.

[0071] In this embodiment, taking the accident type "forward vision" as an example, the assisted driving functions corresponding to its effects include Blind Spot Detection (BSD), Forward Collision Warning (FCW), Forward Cross-Traffic Braking (FCTB), Forward Cross-Traffic Alert (FCTA), Intelligent High Beam Control (IHBC), Traffic Light Recognition (TLR), Traffic Sign Recognition (TSR), Intelligent Speed Assistance (ISA), etc. The above-mentioned assisted driving functions constitute the assisted driving function set for the accident type "forward vision".

[0072] S111. Statistically calculate the configuration rate of the assisted driving functions of the accident subject corresponding to the commercial vehicle accident type, and eliminate the assisted driving functions in the assisted driving function set whose configuration rate exceeds the second threshold. The remaining assisted driving functions constitute the assisted driving functions corresponding to the corresponding commercial vehicle accident type.

[0073] In this embodiment, statistically calculate the assisted driving functions configured for the accident subject (the accident commercial vehicle) corresponding to the commercial vehicle accident type. In one embodiment, taking the accident type "forward vision" as an example, there are 79 pieces of traffic accident data. Statistically calculate the configuration rate of each assisted driving function corresponding to the commercial vehicle in these 79 pieces of traffic accident data and corresponding to the assisted driving functions in the assisted driving function set of the accident type "forward vision".

[0074] Eliminate the assisted driving functions with the configuration rate of the centralized assisted driving functions exceeding the second threshold, where the second threshold is determined according to specific circumstances and can be set to 60%. In one embodiment, the configuration rate of Intelligent Speed Assist (ISA) is 64.6, so Intelligent Speed Assist (ISA) is eliminated from the set of assisted driving functions for the accident type "front vision", and the remaining assisted driving functions constitute the assisted driving functions corresponding to the corresponding commercial vehicle accident type.

[0075] Taking the accident type "front vision" as an example, the corresponding assisted driving functions include Blind Spot Detection (BSD), Forward Collision Warning (FCW), Front Cross-Traffic Braking (FCTB), Front Cross-Traffic Alert (FCTA), Intelligent High Beam Control (IHBC), Traffic Light Recognition (TLR), Traffic Sign Recognition (TSR).

[0076] S12. The assisted driving functions corresponding to all commercial vehicle accident types constitute the assisted driving function library.

[0077] In one embodiment, all commercial vehicle accident types include 9 accident types: front vision, rear vision, side vision, insufficient emergency braking, insufficient emergency steering, side collision during passing, lane change, rear collision during reversing, and driver distraction. The assisted driving functions corresponding to these 9 accident types are shown in Table 2 below.

[0078] Table 2 Assisted Driving Function Library

[0079]

[0080] Exemplarily, the effects of the above 26 commonly used assisted driving functions for commercial vehicles include:

[0081] 1. Blind Spot Detection (BSD): Monitor the blind spots on the side and rear of the vehicle through radar or camera. When other vehicles enter the blind spot, the system will remind the driver through visual or auditory warnings to avoid the collision risk during lane change.

[0082] 2. Forward Collision Warning (FCW): Real-time monitor the conditions of the road ahead and vehicles through sensors such as radar or camera, predict the possible collision danger, and provide warnings to the driver to avoid traffic accidents.

[0083] 3. Front Cross-Traffic Braking (FCTB): Real-time monitor other road users approaching horizontally in the front of the vehicle, and automatically activate the vehicle braking system to decelerate the vehicle when there is a possible collision risk, so as to avoid collision or reduce the collision consequences.

[0084] 4. Front Cross-Traffic Alert (FCTA): Warn the driver of vehicles, pedestrians or cyclists crossing horizontally in front when the vehicle is moving forward at low speed.

[0085] 5. Intelligent High Beam Control (IHBC): Monitors the lights of the vehicle ahead through a camera, and automatically switches between high beams and low beams to avoid dazzling the vehicle ahead, while providing a better night vision when there is no oncoming vehicle.

[0086] 6. Traffic Light Recognition (TLR): Recognizes the status of traffic lights through a camera and provides information to the driver in a timely manner to ensure compliance with traffic signals and avoid violations.

[0087] 7. Traffic Sign Recognition (TSR): Recognizes common traffic signs (such as speed limits, stops, U-turns, etc.) through a front camera and gives instructions or warnings to the driver in a timely manner.

[0088] 8. Rear Cross-Traffic Braking (RCTB): Monitors other road users approaching laterally from behind the vehicle in real time, and automatically applies the brakes when a collision risk is judged to reduce the vehicle speed or avoid a collision.

[0089] 9. Rear Cross-Traffic Alert (RCTA): Warns the driver of oncoming vehicles on both sides when reversing.

[0090] 10. Door Opening Warning (DOW): When the vehicle is parked and about to open the door, a rear radar is used to monitor moving targets in the blind area on the side and rear of the vehicle in real time.

[0091] 11. Automatic Emergency Braking (AEB): Based on environmental perception sensors (such as millimeter-wave radar or vision camera), it senses the potential collision risk with vehicles, pedestrians or other traffic participants ahead, and automatically triggers the actuator (such as: Electronic Stability Program ) to apply the brakes to avoid a collision or reduce the severity of the collision, which is an active safety function.

[0092] 12. Intelligent Speed Assist (ISA): By recognizing speed limit information and providing prompts or actively controlling the vehicle speed when the driver is speeding, it can avoid speeding accidents or reduce the harm of accidents.

[0093] 13. Adaptive Cruise Control (ACC) with Stop & Go: Monitors the speed and distance of the vehicle ahead in real time through sensors such as radar or camera, and automatically adjusts the speed of the vehicle to maintain a set safe distance.

[0094] 14. Traffic Jam Assist (TJA): Helps the driver reduce driving stress in a congested traffic environment, can automatically adjust the vehicle speed during low-speed driving, and also incorporates a function of fine-tuning the steering wheel to keep the vehicle driving within the lane.

[0095] 15. Emergency Lane Keeping (ELK): Detect lane lines, lane edges, oncoming vehicles, and following vehicles through cameras and radars, and provide steering control for the driver in advance within the speed range of 60 km / h to 140 km / h (adjustable).

[0096] 16. Automatic Emergency Steering (AES): Monitor the driving environment in front of, on the side, and on the side of the vehicle in real time through sensors, and automatically control the vehicle's steering when there is a potential collision risk to avoid collisions or mitigate the consequences of collisions.

[0097] 17. Emergency Steering Assist (ESA): Assist the driver in steering operations when there is a potential collision risk and the driver has a clear steering intention to avoid collisions or mitigate the consequences of collisions.

[0098] 18. Integrated Cruise Assist (ICA): Monitor the driving environment in front of and in the adjacent lanes of the vehicle in real time through sensors, provide intelligent longitudinal and lateral driving assistance for the driver, reduce the driver's driving fatigue, and improve driving safety and comfort.

[0099] 19. Lane Centering Keep (LCK): Monitor the lane dividing line through the camera on the windshield, and automatically apply steering force to keep the vehicle in the center of the lane, thereby improving driving safety and comfort.

[0100] 20. Lane Departure Prevention (LDP): Monitor the relative position of the vehicle and the lane line in real time through sensors such as cameras or radars. When the vehicle is about to deviate from the lane, the system will automatically make a small steering adjustment to help the vehicle return to the center of the lane.

[0101] 21. Lane Departure Warning (LDW): Monitor the relative position of the vehicle and the lane line in real time through cameras or sensors. When the vehicle deviates from the lane without turning on the turn signal, the system will remind the driver through visual, auditory, or tactile warnings to avoid potential collision risks.

[0102] 22. Lane Change Assist (LCA): Monitor the blind spot behind and on the side of the vehicle, and remind the driver of the situation of following vehicles, thereby improving safety when changing lanes.

[0103] 23. Smart Dodge Assist (SDA): Assist the driver in automatically controlling the vehicle to slightly adjust in the direction away from large vehicles (such as trucks, lorries, buses, etc.) in the adjacent lane when overtaking, to maintain a safe distance, thereby improving driving safety.

[0104] 24. Direct Change Lane Control (DCLC): The driver issues a lane change command by toggling the turn signal lever, and the system will automatically detect the surrounding environment and perform the lane change operation after confirming safety, ensuring the safety and comfort of the lane change process

[0105] 25. Parking Distance Control (PDC): Detects the distance of obstacles around the vehicle through ultrasonic sensors and alerts the driver with audible and visual signals when necessary to enhance safety during parking and low-speed driving.

[0106] 26. Driver Monitoring System (DMS): Implements functions such as driver identification, driver fatigue monitoring, driver attention monitoring, and monitoring of dangerous driving behaviors.

[0107] Synchronization, in cooperation with the assisted driving function library, for each assisted driving function also includes corresponding technical parameters, performance indicators, and test data.

[0108] Taking Automatic Emergency Braking (AEB) as an example, the technical parameters can include sensor type, controller type, and actuator type. Exemplarily, the sensor type can be millimeter-wave radar and camera; the controller type can be integrated inside the sensor or borne by the domain controller; the actuator type can be one of the ESC system and the EHB system.

[0109] Taking Automatic Emergency Braking (AEB) as an example, the performance indicators can include the working speed range, collision avoidance success rate (the proportion of the system successfully avoiding collisions in specific test scenarios), system false alarm rate (the proportion of the system wrongly issuing warnings when there is no potential collision risk), etc. Among them, the AEB system has its working speed range, generally being able to effectively function between 5 - 80 km / h. At speeds below 5 km / h, the vehicle speed itself is low, and the driver can usually control the vehicle through their own operations; at speeds above 80 km / h, the AEB system may not be able to completely avoid collisions due to various factors (such as sensor performance limitations, too long braking distance, etc.), but can still play a certain role in decelerating to mitigate the consequences of the collision.

[0110] Taking Automatic Emergency Braking (AEB) as an example, the test data can include: vehicle speed (the distance traveled by the vehicle per unit time), collision time (the time it takes for the vehicle to travel normally until a collision occurs), etc.

[0111] S2. Determine the form weights of each assisted driving function based on the impact of each assisted driving function on driving safety.

[0112] In this embodiment, determine the action forms of each assisted driving function based on the impact of each assisted driving function on driving safety. The action forms include control type, warning type, and assistance type.

[0113] The control type of advanced driver assistance functions is defined as: intervening in the control of the vehicle's driving behavior under specific circumstances, assisting driving by controlling the vehicle's driving state, and helping the driver reduce the driving burden; determining the form weight according to the action form. Exemplarily, for the control type of the driver assistance position, the form weight is set to 0.6. In one embodiment, the control type includes automatic emergency braking (AEB), emergency lane keeping (ELK), emergency steering assist (ESA), automatic emergency steering (AES), lane centering keep (LCK), lane departure prevention (LDP), adaptive cruise control (ACC), integrated cruise assist (ICA), rear cross-traffic braking (RCTB), front cross-traffic braking (FCTB), traffic jam assist (TJA), intelligent evasion (SDA), digital clutchless lane change (DCLC), parking distance control (PDC).

[0114] The warning type of advanced driver assistance functions is defined as: perceiving and identifying the risk factors around the vehicle, and timely sending a warning to the driver to remind the driver of potential dangers, but not directly controlling the vehicle's driving state. Exemplarily, for the control type of the driver assistance position, the form weight is set to 0.3. In one embodiment, the warning type includes forward collision warning (FCW), lane departure warning (LDW), blind spot detection (BSD), door open warning (DOW), lane change assist (LCA), driver monitoring system (DMS), rear cross-traffic alert (RCTA), front cross-traffic alert (FCTA), intelligent speed assist (ISA).

[0115] The assistance type of advanced driver assistance functions is defined as: mainly providing the information and functions required by the driver, improving the driving convenience and comfort, and generally not directly controlling the vehicle's driving state. Exemplarily, for the control type of the driver assistance position, the form weight is set to 0.1. In one embodiment, the assistance type includes intelligent high beam control (IHBC), traffic sign recognition (TSR), traffic light recognition (TLR).

[0116] S3. Determine the proportion of each advanced driver assistance function according to the Delphi method, where the sum of the proportions of all advanced driver assistance functions is 1.

[0117] In this embodiment, the Delphi method is also known as the expert opinion method; determining the proportion of each advanced driver assistance function according to the Delphi method may include the following steps:

[0118] S31. Determine at least 10 experts in the field of advanced driver assistance. In one embodiment, invite 10 engineers in the automotive industry (including automotive electronics system engineers, autonomous driving system engineers, etc.), automotive safety experts, automotive test engineers, and research scholars in related fields to form an expert team. Among them, each expert should have rich knowledge and test experience in the ADAS system.

[0119] S32. Each expert gives a proportion according to the contribution degree of each assisted driving function to the driving safety. In one embodiment, for 25 common assisted driving functions in the function database, a detailed questionnaire is designed, allowing experts to give the proportion according to the contribution degree of each function to the driving safety, and at the same time give the scoring reasons.

[0120] S33. For any assisted driving function, remove one maximum value and one minimum value, and average the mean values of the remaining experts to obtain the proportion corresponding to the assisted driving function. In one embodiment, the proportion of each assisted driving function is determined by averaging the mean values of the remaining eight experts.

[0121] Exemplarily, the test results are shown in Table 3 below.

[0122] Table 3 Proportion of Assisted Driving Functions

[0123]

[0124]

[0125] S4. Determine the test scenarios for each assisted driving function.

[0126] In this embodiment, in combination with the actual traffic environment and regulatory requirements, test scenarios are formulated for each assisted driving function. In one embodiment, determining the test scenarios for each assisted driving function includes scenario settings under normal working conditions and failure conditions. Among them, for the scenario setting under normal working conditions, single-factor variables are adopted, covering different test conditions.

[0127] Exemplarily: Taking the automatic emergency braking system (AEB) as an example, verify whether the system can correctly identify road conditions and traffic conditions, and accordingly adjust the vehicle speed, lane position or perform emergency braking. Set 3 test scenarios as follows:

[0128] Test Scenario 1: Warning and activation performance under stationary target (day / night)

[0129] The test vehicle should drive straight towards the stationary target for at least 2 s before the start of the test; the deviation between the test vehicle and the center line of the target should not exceed 0.5 m.

[0130] The test should start when the test vehicle is driving at a speed of (80 ± 2) km / h and is at least 120 m away from the target. Vehicles with a maximum designed speed less than 80 km / h should conduct the test at the maximum speed.

[0131] Except for slightly adjusting the steering wheel to prevent the vehicle from deviating, the driver should not make any adjustments to the test vehicle from the start of the test until the collision point.

[0132] Condition for passing: The time setting of the collision warning mode shall comply with the following regulations:

[0133] 1) The test vehicle shall start warning in at least one of the acoustic, tactile and optical modes no later than the following time:

[0134] For vehicles with pneumatic braking systems, it is 1.4 s before the start of the emergency braking phase; for M-class vehicles with power-assisted hydraulic braking systems and N-class vehicles with a maximum design total mass less than or equal to 8 t, it is 0.8 s before the start of the emergency braking phase.

[0135] 2) The test vehicle shall start warning in at least two of the acoustic, tactile and optical modes no later than the following time:

[0136] For vehicles with pneumatic braking systems, it is 0.8 s before the start of the emergency braking phase; for M-class vehicles with power-assisted hydraulic braking systems and N-class vehicles with a maximum design total mass less than or equal to 8 t, it is before the start of the emergency braking phase.

[0137] 3) The speed reduction during the warning phase shall not exceed 15 km / h or 30% of the total speed reduction of the test vehicle, whichever is higher.

[0138] 4) The emergency braking phase shall start after the warning phase.

[0139] 5) The total speed reduction of the test vehicle when colliding with a stationary target shall not be less than 10 km / h.

[0140] 6) The emergency braking phase shall not start when the predicted collision time is less than or equal to 3 s.

[0141] 7) After excluding the interference of other factors, at least 3 out of 5 tests shall meet the above regulations.

[0142] Test scenario 2: Warning and activation performance under moving target conditions (day / night)

[0143] The test vehicle and the moving target shall travel in the same direction along a straight line for at least 2 s before the test; the deviation of the test vehicle from the center line of the target shall not exceed 0.5 m.

[0144] The test shall start when the test vehicle is traveling at a speed of (80 ± 2) km / h, the moving target for the test vehicle with a pneumatic braking system is traveling at a speed of (32 ± 2) km / h, the moving target for the M-class test vehicle with a power-assisted hydraulic braking system and the N-class test vehicle with a maximum design total mass less than or equal to 8 t is traveling at a speed of (67 ± 2) km / h, and when the distance between them is at least 120 m; vehicles with a maximum design speed less than 80 km / h shall be tested at their maximum speed.

[0145] Except for making minor adjustments to the steering wheel to prevent the vehicle from deviating, the driver should not make any adjustments to the test vehicle from the start of the test until the speed of the test vehicle is equal to the target speed.

[0146] Passing condition: The time setting of the collision warning mode shall comply with the provisions of GB / T 38186-2019 4.3.2.1.

[0147] The speed reduction during the warning stage shall not exceed 15 km / h or 30% of the total speed reduction of the test vehicle, whichever is higher.

[0148] After the collision warning stage, there shall be an emergency braking stage. The test vehicle shall not collide with the moving target during the emergency braking stage, and the emergency braking stage shall not start 3 s before the predicted collision time.

[0149] After excluding driver interference, at least 3 out of 5 tests shall meet the above provisions.

[0150] Test scenario 3: Warning signal after system failure

[0151] Simulate circuit failure by disconnecting the power supply of the AEBS component or the circuit connection between AEBS components. When simulating AEBS failure, the circuit connection of the driver warning signal specified in 4.2.1 and the AEBS manual shutdown control device specified in 4.5 shall not be cut off.

[0152] Start and gradually accelerate the test vehicle, observe and record the signal of the failure warning device, the vehicle speed and the time when the warning signal is first issued; after parking, turn off the ignition switch in the vehicle stationary state and then turn it on again to check whether the failure warning signal is immediately re-lit.

[0153] Passing condition:

[0154] The continuously illuminated optical warning signal that complies with the provisions of GB4094-2016 shall be activated at the latest when the vehicle is traveling at a speed greater than 15 km / h for 10 s; and as long as the failure still exists, after the ignition switch of the vehicle is turned off and restarted in the stationary state, it shall still maintain its failure warning state.

[0155] S5. Determine the weights of each test scenario based on historical data and / or the Delphi method.

[0156] In this embodiment, for the test scenarios of each assisted driving function, the following three methods can be used to determine the weights of each test scenario. Among them, in any method, for the same assisted driving function, the sum of the weights of each test scenario is 1.

[0157] Method 1: Determine the weights of each test scenario based on historical data. In this method, according to the historical test result data of each test scenario, comprehensively measure the relative importance of each test scenario, and determine the weights of each test scenario based on the relative importance.

[0158] Method 2: Determine the weights of each test scenario based on the Delphi method. In this method, a method of collecting opinions and determining weights through expert surveys is used. Invite at least ten experts in related fields to score and rank each test scenario, and then determine the weights according to the opinions of the experts.

[0159] Method 3: Determine the weights of each test scenario based on historical data and the Delphi method. Combine the weights determined by Method 1 and Method 2 to comprehensively determine the weights of each test scenario.

[0160] In one embodiment, taking the automatic emergency braking system (AEB) as an example, the score weight of test scenario 1 is 40%, the score weight of test scenario 2 is 40%, and the score weight of test scenario 3 is 20%.

[0161] S6. Determine the assisted driving functions of the commercial vehicle to be tested based on the assisted driving function library.

[0162] In this embodiment, determine the assisted driving functions of the commercial vehicle to be tested, compare the assisted driving functions of the commercial vehicle to be tested with the assisted driving functions in the assisted driving function library, and count the same assisted driving functions as the assisted driving functions of the commercial vehicle to be tested for participation in the test.

[0163] S7. For any assisted driving function, obtain the test results under the corresponding test scenarios.

[0164] In this embodiment, for each scenario, the test results are respectively passed and not passed, passed is recorded as 1, and not passed is recorded as 0. In one specific embodiment, taking the automatic emergency braking system (AEB) of the commercial vehicle to be tested as an example, test scenarios 1, 2, and 3 all pass, and are recorded as 1.

[0165] S7. Determine the scenario score of the corresponding assisted driving function based on the test results and the test scenario weights.

[0166] In this embodiment, it is set that the test scenario is n, n≥1; the weights of each test scenario are recorded as W1, W2,......, W n ; the scores of the test scenarios corresponding to each assisted driving function are recorded as S1, S2,......, S n , then the scenario score of this assisted driving function is recorded as:

[0167] α 场景得分 =S1*W1 + S2*W2 +...... + S n*W n

[0168] In one embodiment, taking the automatic emergency braking system (AEB) as an example, α 场景得分 = 1 * 0.4 + 1 * 0.4 + 1 * 0.2 = 1.

[0169] S8. Determine the single - function scores of each assisted driving function of the commercial vehicle to be tested based on the form weight, proportion, and scenario score.

[0170] In this embodiment, α 单一功能得分 = α 形式 *α 占比 *α 场景得分 , where α 单一功能得分 is the first - function score of the commercial vehicle to be tested for the corresponding assisted driving function, α 形式 is the form weight of the corresponding assisted driving function of the commercial vehicle to be tested, and α 场景得分 is the scenario score of the corresponding assisted driving function of the commercial vehicle to be tested.

[0171] In one embodiment, taking the automatic emergency braking system (AEB) as an example, α 单一功能得分 = 0.6 * 9.66% * 1 = 0.05976;

[0172] S9. Determine the score of the commercial vehicle to be tested based on the single - function scores of each assisted driving function to evaluate its safety.

[0173] In this embodiment, determining the score of the commercial vehicle to be tested based on the single - function scores of each assisted driving function to evaluate its safety includes the following steps:

[0174] S900. Calculate the product of the form weight and proportion of each assisted driving function in the assisted driving function library, and then sum them up, denoted as C 10 ;

[0175] C 10 = Σα 形式 *α 占比

[0176] In one specific embodiment, C 10 = 0.50731.

[0177] S901. Calculate the sum of the single - function scores of each assisted driving function of the commercial vehicle to be tested, denoted as C 11 .

[0178] In one specific embodiment,

[0179] S902. Determine the score of the commercial vehicle to be tested, denoted as C = (C 11 * 100) / C10 。

[0180] The score of the commercial vehicle to be tested is in the range of 0 - 100 points. In one embodiment, it can be divided into three levels: excellent (≥90), good (75 - 89), and to be improved (<75). Specifically, it can be adjusted and controlled according to the actual situation.

[0181] Embodiment 2

[0182] Embodiment 2 of the present application discloses a method for evaluating the safety performance of a commercial vehicle based on assisted driving. Similar to Embodiment 1, the difference between this Embodiment 2 and Embodiment 1 lies in step S9. Step S9: Determine the score of the commercial vehicle to be tested based on the single - function scores of each assisted - driving function to evaluate its safety.

[0183] In this embodiment, determining the score of the commercial vehicle to be tested based on the single - function scores of each assisted - driving function to evaluate its safety includes the following steps:

[0184] S910: Based on the accident type, calculate the product of the formal weight and the proportion for any assisted - driving function in the assisted - driving function library, and then sum them to obtain the theoretical score for each accident type.

[0185] In one specific embodiment, there are 9 types of accident types, and each type of accident type includes corresponding assisted - driving functions. See Table 2 for details. Taking the accident type "front - view field" as an example, it includes 7 assisted - driving functions: Blind Spot Detection (BSD), Forward Collision Warning (FCW), Front Cross - traffic Braking (FCTB), Front Cross - traffic Alert (FCTA), Intelligent High - beam Control (IHBC), Traffic Light Recognition (TLR), and Traffic Sign Recognition (TSR). Calculate the product of the formal weight and the proportion for each of these 7 assisted - driving functions respectively, and then sum them to obtain the theoretical score for each accident type. Specifically, taking the accident type "front - view field" as an example, the theoretical score of the accident type is 0.05743.

[0186] S911: For any accident type in the assisted - driving function library, calculate the sum of the product of the accident - type proportion and the theoretical score of the corresponding accident type, and denote it as C 20 。

[0187] In one embodiment, the accident - type proportion is as shown in Table 2. Taking the accident type "front - view field" as an example, the accident - type proportion is 15.8%, and the theoretical score of the accident type is 0.05743. Multiply the two. Sum the results of the above products for all accident types to obtain C 20 。

[0188] S912: Based on the accident type, group and sum the single - function scores of the commercial vehicle to be tested to obtain the actual score for each accident type.

[0189] In this embodiment, the assisted driving functions of the commercial vehicle to be tested participating in the test are grouped according to the accident type, and the single function scores of the assisted driving functions within each group are summed to obtain the actual accident type score of the commercial vehicle to be tested for this accident type.

[0190] S913. For any accident function score type of the commercial vehicle to be tested, calculate the sum of the products of the accident type proportion and the actual score of the corresponding accident type, and denote it as C 21 。

[0191] In this embodiment, for any accident type of the commercial vehicle to be tested, calculate the product of the accident type proportion (in one embodiment, refer to Table 2) and the actual score of the corresponding accident type, and then sum them, and denote it as C 21 。

[0192] S914. Determine the score of the commercial vehicle to be tested and denote it as C = (C 21 * 100) / C 20 。

[0193] The score of the commercial vehicle to be tested is in the range of 0 - 100 points. In one embodiment, it can be divided into three levels: excellent (≥90), good (75 - 89), and need improvement (<75), and specific adjustments and controls can be made according to the actual situation.

[0194] Embodiment 3

[0195] Embodiment 3 of the present application discloses a method for evaluating the safety performance of commercial vehicles based on assisted driving. The difference between this embodiment and Embodiment 1 is that it further includes:

[0196] 1) A cloud platform monitoring module, which is used to collect and analyze the operation data of commercial vehicles in actual traffic in real time, monitor the actual performance of assisted driving functions, and provide data support for model optimization. Specifically, it includes: in-vehicle data acquisition equipment, a data transmission module, and a data storage and management module.

[0197] Regarding the in-vehicle data acquisition equipment, taking the automatic emergency braking system (AEB) as an example: Professional data acquisition terminals are installed on various commercial vehicles (light, medium, and heavy commercial vehicles) participating in the test. These terminals have the following functions:

[0198] Sensor data recording: Connect millimeter-wave radars, cameras and other sensors in the automatic emergency braking system (AEB) already equipped on commercial vehicles, and record in real time the target information they detect. The target information includes target distance, speed, angle, type (vehicles, pedestrians, obstacles, etc.), and the working state parameters of the sensors themselves (such as the radar transmission power, camera frame rate, etc.).

[0199] Vehicle operation data recording: It is connected to the vehicle's CAN bus to obtain basic operation data such as vehicle speed, acceleration, steering angle, brake pedal status, and gear information. At the same time, it records the vehicle's positioning information (through GPS or Beidou positioning system), accurate to longitude and latitude as well as timestamp, so as to restore the vehicle's driving trajectory.

[0200] Environmental data recording: It is equipped with environmental sensors to collect external environmental information such as temperature, humidity, light intensity, and weather conditions (obtaining real-time weather categories such as rain, snow, sunny, etc. through docking with meteorological data interfaces). These data help analyze the impact of different environments on the performance of the AEB system.

[0201] Regarding the data transmission module, 4G / 5G wireless communication technology can be used to transmit the data collected by the on-vehicle data acquisition device to the cloud server in real time. To ensure the stability and integrity of data transmission, a data caching mechanism is set up. When the network signal is poor, the data is temporarily stored in the local cache first, and then batch uploaded after the network resumes.

[0202] Regarding the data storage and management module, a large database is built on the cloud server, and the uploaded data is classified and stored according to different dimensions such as commercial vehicle types, license plate numbers, date and time, etc., which is convenient for subsequent querying, screening, and analysis. At the same time, the data is encrypted to ensure the security and privacy of vehicle operation data.

[0203] 2) The model optimization module is used to further evaluate and optimize the commercial vehicle advanced driver assistance system (ADAS) by using the actual traffic data collected by the cloud platform, including:

[0204] Establish a long-term data feedback mechanism:

[0205] Continuously collect actual traffic data through the cloud platform, and re-evaluate the ADAS system in real time or regularly, because the actual traffic conditions are constantly changing (such as new road construction, traffic rule changes, etc.), and the system also needs to continuously adapt.

[0206] The embodiments of this specific implementation manner are all preferred embodiments of this application, and do not limit the protection scope of this application accordingly. The same components are represented by the same reference numerals. Therefore, all equivalent changes made according to the structure, shape, and principle of this application should be covered within the protection scope of this application.

Claims

1. A method for evaluating the safety performance of commercial vehicles based on assisted driving, characterized in that It includes the following steps: Determine the assisted driving function library of the commercial vehicle, where the assisted driving function library includes multiple assisted driving functions; Determine the form weights of each assisted driving function according to the impact of each assisted driving function on driving safety; Determine the proportion of each assisted driving function according to the Delphi method, where the sum of the proportions of all assisted driving functions is 1; Determine the test scenarios of each assisted driving function and determine the weights of each test scenario; Determine the assisted driving functions of the commercial vehicle to be tested based on the assisted driving function library; For any assisted driving function, test results are obtained under the corresponding test scenarios, and the scenario score of the corresponding assisted driving function is determined based on the test results and the test scenario weights; Determine the single function scores of each assisted driving function of the commercial vehicle to be tested based on the form weights, proportions, and scenario scores; Determine the score of the commercial vehicle to be tested based on the single function scores of each assisted driving function to evaluate its safety.

2. The safety performance evaluation method for commercial vehicles based on assisted driving according to claim 1, wherein The determination of the assisted driving function library includes: Based on the proportion of commercial vehicle traffic accidents, determine the types of commercial vehicle accidents exceeding the first threshold; Determine the assisted driving function corresponding to any type of commercial vehicle accident; The assisted driving functions corresponding to all types of commercial vehicle accidents constitute the assisted driving function library.

3. The method for evaluating the safety performance of commercial vehicles based on assisted driving according to claim 2, wherein The determination of the assisted driving function corresponding to any type of commercial vehicle accident includes: Based on the type of commercial vehicle accident and the effect of the assisted driving function, determine the set of assisted driving functions corresponding to the type of commercial vehicle accident; Statistically calculate the assisted driving function configuration rate of the accident subject corresponding to the type of commercial vehicle accident, and eliminate the assisted driving functions in the assisted driving function set whose assisted driving function configuration rate exceeds the second threshold. The remaining assisted driving functions constitute the assisted driving function corresponding to the type of commercial vehicle accident.

4. The method for evaluating the safety performance of commercial vehicles based on assisted driving according to claim 1, wherein The determination of the form weights of each assisted driving function according to the impact of each assisted driving function on driving safety includes: Determine the action forms of each assisted driving function, where the action forms include control type, warning type, and assistance type; Determine the form weights according to the action forms.

5. The safety performance evaluation method for commercial vehicles based on assisted driving according to claim 1, characterized in that The determination of the proportion of each assisted driving function according to the Delphi method includes: Determine at least 10 experts in the field of assisted driving; Each expert gives the proportion according to the contribution degree of each assisted driving function to driving safety; For any assisted driving function, remove one maximum value, remove one minimum value, and average the means of the remaining experts to obtain the proportion of the corresponding assisted driving function.

6. The method for evaluating the safety performance of commercial vehicles based on assisted driving according to claim 1, wherein, The determination of the test scenarios of each assisted driving function includes: Scenario settings under normal working conditions and failure conditions.

7. The method for evaluating the safety performance of commercial vehicles based on assisted driving according to claim 1, wherein The test results include pass and fail, and the test scores for pass and fail are recorded as 1 and 0 respectively; The determination of the scenario score of the corresponding assisted driving function based on the test results and the test scenario weights includes: For any assisted driving function, calculate the product of the weights of each test scenario and the test scores, and then sum them to obtain the scenario score of the corresponding vehicle assisted driving function.

8. The method for evaluating the safety performance of commercial vehicles based on assisted driving according to claim 1, wherein The determination of the scores of each assisted driving function based on the form weights, proportions, and scenario scores includes: Calculate the product of the form weight, proportion, and scenario score corresponding to each assisted driving function to obtain the score of the corresponding assisted driving function.

9. The method for evaluating the safety performance of commercial vehicles based on assisted driving according to claim 1, wherein Determining the score of a commercial vehicle based on the first functional scores of various assisted driving functions to evaluate its safety includes: Calculate the product of the formal weights and proportions of each assisted driving function in the assisted driving function library, and then sum them up, denoted as C 10 ; Calculate the sum of the single-function scores of each assisted driving function of the commercial vehicle to be measured, denoted as C 11 ; Determine that the score of the commercial vehicle to be measured is recorded as C = (C 11 * 100) / C 10 .

10. The method for evaluating the safety performance of a commercial vehicle based on assisted driving according to claim 1, wherein, Determining the score of a commercial vehicle based on the single functional scores of various assisted driving functions to evaluate its safety includes: Based on the accident type, calculate the product of the formal weight and the proportion for any assisted driving function in the assisted driving function library, and then sum them up to obtain the theoretical score for each accident type; For any accident type in the assisted driving function library, calculate the sum of the products of the proportion of the accident type and the theoretical score of the corresponding accident type, and denote it as C 20 ; Based on the accident type, group and sum the single functional scores of the commercial vehicle to be tested to obtain the actual score for each accident type; For any accident type of the commercial vehicle to be tested, calculate the sum of the products of the accident type proportion and the actual score of the corresponding accident type, and denote it as C 21 ; Determine the score of the commercial vehicle to be measured, denoted as C = (C 21 * 100) / C 20 .

Images