Method and system for evaluating comprehensive performance of automatic emergency braking system of automobile
By constructing a two-level dynamic weight architecture and an adaptive evaluation model with five-dimensional indicators, the problem of performance evaluation deviation of the AEB system in complex traffic scenarios is solved, accurate performance evaluation under multiple working conditions is achieved, and the scientific nature and engineering applicability of the evaluation are improved.
Patent Information
- Application Number
- CN202510786837.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-13
- Publication Date
- 2025-09-26
AI Technical Summary
Existing performance evaluation methods for automotive automatic emergency braking (AEB) systems suffer from poor reliability and deviations between evaluation results and actual performance in complex traffic scenarios. In particular, they are unable to accurately reflect the system's dynamic response characteristics in severe weather, high-speed driving, and when vision is obstructed. Traditional evaluation frameworks are disconnected from experimental verification, affecting evaluation accuracy and engineering applicability.
A comprehensive performance evaluation method for the AEB system is constructed using a two-level dynamic weight architecture. Through a two-variable weight function of vehicle speed and road adhesion coefficient, five-dimensional indicators such as braking distance and collision speed are integrated to establish a multi-condition adaptive evaluation model. The weights and condition parameters are dynamically mapped to achieve dynamic adjustment of performance evaluation.
It effectively solves the problems of insufficient static weight evaluation in AEB system performance evaluation and the disconnection between traditional evaluation framework and experimental verification, realizes accurate performance evaluation in complex scenarios, and improves the scientific nature and engineering applicability of the evaluation.
Smart Images
Figure CN120705013A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of brake system detection, and in particular to a method and system for evaluating the comprehensive performance of an automobile automatic emergency brake system. Background Art
[0002] With the rapid development of technologies like artificial intelligence and big data, intelligent and connected vehicles (ICVs) are becoming a focal point in the global automotive industry. Autonomous Emergency Braking (AEB), a core active safety technology in ICVs, has a direct impact on road safety. Research by the Insurance Institute for Highway Safety (IIHS) shows that vehicles equipped with AEB systems can reduce the accident rate by 27%. AEB reduces the risk of fatalities in pedestrian accidents by 84% to 87% and the risk of serious injuries (MAIS 3+) by 83% to 87%. AEB has proven highly effective, preventing 83% of rear-end collisions.
[0003] However, in complex traffic scenarios (such as pedestrians crossing urban intersections and strong light interference on highways), the reliability of AEB systems varies across different driving environments and operating conditions. Reliability is poor in adverse weather conditions, at high speeds, and when vision is obstructed. This poses significant challenges to the scientific nature of evaluation methods. First, current mainstream evaluation methods (such as Euro NCAP and C-NCAP) employ a static weight distribution mechanism, which struggles to accurately reflect the dynamic response characteristics of the system under multiple coupled operating conditions. Second, traditional evaluation frameworks often suffer from a disconnect between theoretical models and experimental verification, leading to deviations between evaluation results and actual road performance. With the widespread adoption of L2+ / L3 autonomous driving technology, AEB systems must cope with increasingly complex scenarios (such as pedestrians crossing the road and multi-vehicle interactions). These two key flaws hinder the accuracy and engineering applicability of AEB system performance evaluations.
[0004] Currently, most system performance tests are based on simulations and real-world road tests. Simulations, conducted under idealized scenarios, are highly subjective and rely primarily on the construction of control models and the setting of key parameters, so they should not be considered a primary reference. Real-world road tests provide accurate results that reflect real-world driving conditions, comprehensively considering parameters such as the environment, road conditions, and driver characteristics. However, they present significant safety risks, a limited number of reproducible scenarios, and low efficiency. Summary of the Invention
[0005] To solve the above problems, the present invention provides a method and system for evaluating the comprehensive performance of an automobile automatic emergency braking system.
[0006] The technical solutions of the present invention are as follows:
[0007] A comprehensive performance evaluation method for an automobile automatic emergency braking system, comprising:
[0008] Constructing an AEB system comprehensive performance evaluation model, comprising a target layer, a criterion layer, and a solution layer; the target layer is the AEB system comprehensive performance score; the criterion layer is the AEB system performance standard; and the solution layer is the AEB system evaluation index;
[0009] Based on the performance standards of the AEB system in the criterion layer, the priorities of each performance standard in different driving scenarios are divided, and the weight function of the criterion layer relative to the target layer is established based on the vehicle speed and road adhesion coefficient in different driving scenarios;
[0010] Compare each performance criterion of the criterion layer with the weight function of the target layer in pairs, construct the importance relationship table of the criterion layer relative to the target layer, obtain the judgment matrix of the criterion layer, calculate the weight vector of the judgment matrix, and standardize it;
[0011] Compare each evaluation indicator of the solution layer with each performance standard of the criterion layer in pairs, build the importance relationship table of each evaluation indicator of the solution layer with respect to the criterion layer, obtain the judgment matrix of the solution layer and calculate the weight vector of the judgment matrix of the solution layer;
[0012] A transposed matrix is established based on the evaluation indicators. The final AEB system comprehensive performance evaluation model is determined based on the transposed matrix, the weight vector of the judgment matrix of the solution layer relative to the criterion layer, and the weight vector of the judgment matrix of the criterion layer relative to the target layer.
[0013] By inputting the evaluation index values, vehicle speed and road adhesion coefficient under actual driving conditions, the comprehensive performance score of the AEB system is obtained.
[0014] The final AEB system comprehensive performance evaluation model is:
[0015] EM tc =ω1×w B1 w t +ω2×w B2 w t +ω3×w B3 w t
[0016] Where ω1 represents the weight vector of AEB system safety relative to the comprehensive performance of the AEB system; ω2 represents the weight vector of AEB system reliability relative to the comprehensive performance of the AEB system; ω3 represents the weight vector of driving comfort relative to the comprehensive performance of the AEB system; w t is the transposed matrix of the evaluation index; w B1is the weight vector of the scheme-level evaluation index relative to the judgment matrix of the AEB system safety at the criterion level; w B2 is the weight vector of the judgment matrix of the scheme layer evaluation index relative to the criterion layer AEB system reliability; w B3 is the weight vector of the judgment matrix of the scheme-level evaluation index relative to the criterion-level driving comfort.
[0017] The weight vectors of the judgment matrix of the solution layer relative to the criterion layer are:
[0018]
[0019] The formula of the weight function of the criterion layer relative to the target layer is:
[0020] S ij (v,μ)=a i v+b i μ+c i (≥0),
[0021] Among them, a i is the speed correspondence layer weight S i The influence coefficient of b i is the road adhesion coefficient criterion layer weight S i The influence coefficient of i is the base constant.
[0022] The performance standards considered in the comprehensive performance evaluation of the AEB system include AEB system safety, AEB system reliability and driving comfort.
[0023] The evaluation indicators of the AEB system include braking distance d br , braking deceleration a br , collision speed v c , AEB system braking intervention time t br_in , acceleration change rate j.
[0024] After constructing the importance relationship table of the criterion layer relative to the target layer to obtain the judgment matrix of the criterion layer, it also includes combining data of different vehicle speeds and different road adhesion coefficients in pairs, substituting the combined data into the judgment matrix, calculating the weight vector of each judgment matrix, and performing consistency test.
[0025] A comprehensive performance evaluation system for an automatic emergency braking system of an automobile, for implementing the comprehensive performance evaluation method for an automatic emergency braking system of an automobile as described above, comprising:
[0026] An initial model construction module is used to construct an AEB system comprehensive performance evaluation model, which includes a target layer, a criterion layer, and a solution layer; the target layer is the AEB system comprehensive performance score; the criterion layer is the AEB system performance standard; and the solution layer is the AEB system evaluation index;
[0027] A dynamic weighting module prioritizes the performance criteria of the AEB system in different driving scenarios based on the criteria layer, and establishes a weighting function for the criteria layer relative to the target layer based on the vehicle speed and road adhesion coefficient in different driving scenarios.
[0028] The criterion layer weight comparison module is used to compare the weight functions of the performance criteria of the criterion layer with those of the target layer, construct a table of the importance of the criterion layer relative to the target layer, obtain the judgment matrix of the criterion layer, calculate the weight vector of the judgment matrix, and standardize it;
[0029] The scheme-level weight comparison module is used to compare each evaluation indicator of the scheme level with each performance standard of the criterion level, construct the importance relationship table of each evaluation indicator of the scheme level with respect to the criterion level, obtain the judgment matrix of the scheme level, and calculate the weight vector of the judgment matrix of the scheme level;
[0030] An evaluation model generation module is used to establish a transposed matrix based on the evaluation indicators and determine the final AEB system comprehensive performance evaluation model based on the transposed matrix and the weight vector of the judgment matrix of the solution layer relative to the criterion layer, as well as the weight vector of the judgment matrix of the criterion layer relative to the target layer;
[0031] The scoring calculation module is used to obtain the comprehensive performance score of the AEB system by inputting the evaluation index values, vehicle speed and road adhesion coefficient under actual driving conditions.
[0032] A comprehensive performance evaluation device for an automobile automatic emergency braking system includes a processor and a memory, wherein the processor implements the above-mentioned comprehensive performance evaluation method for an automobile automatic emergency braking system when executing a computer program stored in the memory.
[0033] A computer-readable storage medium is used to store a computer program, wherein when the computer program is executed by a processor, the method for evaluating the comprehensive performance of an automatic emergency braking system of an automobile as described above is implemented.
[0034] Beneficial Effects: This invention provides a comprehensive performance evaluation method for automotive automatic emergency braking systems. This method employs a two-level dynamic weighting architecture: the criterion layer introduces a two-variable weighting function for vehicle speed and adhesion coefficient to achieve dynamic coupled modeling of road environment parameters. The solution layer integrates five-dimensional indicators, such as braking distance and collision speed, and establishes a multi-condition adaptive evaluation model using a nonlinear weighting algorithm. Compared to the traditional analytic hierarchy process (AHP), the dynamic weighted AHP effectively addresses the shortcomings of AEB system performance evaluation techniques, such as the lack of adaptability to working conditions due to static weight evaluation, and the disconnection between traditional evaluation frameworks and experimental verification, by establishing a dynamic mapping relationship between weighting functions and working condition parameters. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] By reading the detailed description of the preferred embodiment below, the solutions and advantages of the present application will become clear to those skilled in the art. The accompanying drawings are only for illustrating the preferred embodiment and are not to be considered as limiting the present invention.
[0036] In the attached figure:
[0037] Figure 1 This is a flow chart of the comprehensive performance evaluation method for the automobile automatic emergency braking system;
[0038] Figure 2 This is a schematic diagram of the structure of the AEB system performance test bench;
[0039] Figure 3 These are the test results of the AEB system performance test bench under different working conditions, where a is the test result at a vehicle speed of 20 km / h; b is the test result at a vehicle speed of 30 km / h; c is the test result at a vehicle speed of 40 km / h;
[0040] 1. Front roller assembly frame; 2. Roller assembly; 3. T-type reducer; 4. Retractable drive shaft; 5. Rear roller assembly frame; 6. Mechanical flywheel; 7. Main and auxiliary roller synchronization chain; 8. AC electric dynamometer or eddy current dynamometer; 9. Moving guide rail; 10. Axle load meter; 11. HBM speed and torque sensor; 12. Protective device. DETAILED DESCRIPTION
[0041] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings.
[0042] Example 1
[0043] See also Figure 1 This embodiment provides a method for evaluating the comprehensive performance of an automatic emergency braking system of an automobile, and the specific steps are as follows:
[0044] Step 1: Construct an AEB system comprehensive performance evaluation model, which includes a target layer, a criterion layer, and a solution layer; the target layer is the AEB system comprehensive performance score; the criterion layer is the AEB system performance standard; and the solution layer is the AEB system evaluation index;
[0045] Among them, the performance standards considered in the comprehensive performance evaluation of the AEB system include AEB system safety, AEB system reliability and driving comfort.
[0046] The evaluation indicators of the AEB system include braking distance d br , braking deceleration a br , collision speed v c , AEB system braking intervention time t br_in , acceleration change rate j.
[0047] Step 2: Based on the performance criteria of the AEB system in the criterion layer, the priorities of each performance criterion in different driving scenarios are divided, and the weight function of the criterion layer relative to the target layer is established based on the vehicle speed and road adhesion coefficient in different driving scenarios;
[0048] Based on the performance standards considered in the comprehensive performance evaluation of the AEB system, the driving scenario priorities corresponding to each performance standard are determined, as shown in Table 1.
[0049] Table 1 Priority ranking of AEB systems in different driving scenarios
[0050]
[0051] In low-road adhesion and high-speed driving scenarios, due to insufficient vehicle stability and high collision impact forces, ensuring that the AEB system can successfully avoid obstacles or reduce collision speed after activation is the top priority. Therefore, the safety of the AEB system in this scenario is given high priority.
[0052] In driving scenarios where sensors have limited perception in inclement weather or where the vehicle's field of view is restricted on complex urban roads, it is crucial that the AEB system's environmental perception detects the presence of potential dangers and initiates braking intervention within the control threshold. Therefore, AEB system reliability is a high priority in such scenarios.
[0053] In urban roads, where traffic participants are complex and there are many traffic lights, frequent following and braking is required; or on dry asphalt roads, where the braking deceleration is large and the braking intensity is high, driving comfort is affected. Therefore, driving comfort is a high priority in this scenario.
[0054] According to the vehicle speed and road adhesion coefficient under different driving conditions, the weight function of the criterion layer relative to the target layer is established, and its formula is:
[0055] S ij (v,μ)=a i v+b i μ+c i (≥0),
[0056] Among them, a i is the speed correspondence layer weight S i The influence coefficient of b i is the road adhesion coefficient criterion layer weight S i The influence coefficient of i is the base constant.
[0057] Step 3: Compare the weight functions of the performance criteria of the criterion layer with those of the target layer in pairs, construct a table of the importance of the criterion layer relative to the target layer, obtain the judgment matrix of the criterion layer, calculate the weight vector of the judgment matrix, and standardize it;
[0058] In this embodiment, based on the AEB system comprehensive performance evaluation at the criterion level, the AEB system safety, AEB system reliability, and driving comfort are considered. Corresponding to the priority driving conditions in Table 1, and according to the different vehicle speeds and road adhesion coefficients in each driving condition, the weight function of each performance standard relative to the target level can be given as follows:
[0059] S 12 (v,μ)=3v+(1-μ)+2
[0060] S 13 (v,μ)=5v+3(1-μ)+2
[0061] S 23 (v,μ)=v+3(1-μ)+1
[0062] Among them, μ∈[0.1, 0.9]; v=0~120km / h, which is linearly normalized to [0, 1].
[0063] The weight functions of the performance criteria of the criterion layer relative to the target layer are compared pairwise, and a table of the importance relationship between the criterion layer and the target layer is constructed, as shown in Table 2.
[0064] Table 2 The importance relationship between the criterion layer and the target layer
[0065]
[0066] According to the importance relationship table of the criterion layer relative to the target layer, the judgment matrix A is obtained as follows:
[0067]
[0068] The weight vector of the judgment matrix A is calculated and normalized. The weight vector of the judgment matrix A is:
[0069]
[0070] Among them, ω1 represents the weight vector of AEB system safety relative to the comprehensive performance of the AEB system; ω2 represents the weight vector of AEB system reliability relative to the comprehensive performance of the AEB system; ω3 represents the weight vector of driving comfort relative to the comprehensive performance of the AEB system.
[0071] In addition, after constructing the importance relationship table of the criterion layer relative to the target layer to obtain the judgment matrix of the criterion layer, the method also includes combining data of different vehicle speeds and different road adhesion coefficients in pairs, substituting the combined data into the judgment matrix, and calculating the weight vector of each judgment matrix to perform consistency test;
[0072] The validity of the dynamic weighting function was verified by discretizing the parameters and sampling key points. The factors were vehicle speed and road adhesion coefficient (number of factors m = 2). The vehicle speed level (S) primarily considered low speed (S1) and high speed (S2), while the road adhesion coefficient level (H) primarily considered low road adhesion coefficient (H1) and high road adhesion coefficient (H2). Therefore, the number of factor levels was 2 (n = 2). In summary, the experimental plan was designed, as shown in Table 3.
[0073] Table 3 Sampling inspection test plan design
[0074]
[0075] Among them, S1 is taken as 0.25 (v=30km / h); S2 is taken as 1 (v=120km / h); H1 is taken as 0.2 (μ=0.26); H1 is taken as 0.8 (μ=0.74).
[0076] Substitute the above test plans and data into the judgment matrix A. The four test plans are recorded as judgment matrices A1, A2, A3, and A4, and consistency tests are performed on them respectively.
[0077]
[0078] Calculate the weight vector of each matrix and normalize it, and calculate the maximum eigenvalue λ max The constructed matrix was checked for consistency.
[0079]
[0080] Since the judgment matrix is of third order, by searching for random one-time indicators, RI is taken as 0.58. In summary, the consistency test results of each matrix are shown in Table 4.
[0081] Table 4 Consistency test results
[0082]
[0083] As shown in Table 4, the CRs are all less than 0.1, and the consistency test is passed.
[0084] Step 4: Compare each evaluation indicator of the solution layer with each performance standard of the criterion layer in pairs, construct the importance relationship table of each evaluation indicator of the solution layer with respect to the criterion layer, obtain the judgment matrix of the solution layer, and calculate the weight vector of the judgment matrix of the solution layer;
[0085] According to the five evaluation indicators of the AEB system at the solution level, the solution level braking distance d is constructed respectively. br , braking deceleration a br , collision speed v c , AEB system braking intervention time t br_in , the importance relationship table of acceleration change rate j relative to AEB system safety, AEB system reliability and driving comfort of the criterion layer, the importance relationship table is shown in Table 5-7.
[0086] Table 5. Relationship between the importance of solution layers and AEB system safety
[0087]
[0088] Table 6 Relationship between the importance of solution layers and AEB system reliability
[0089]
[0090] Table 7 Relationship between the importance of solution layers and driving comfort
[0091]
[0092] According to the importance relationship table, the judgment matrix of the solution layer relative to the criterion layer is:
[0093]
[0094] The weight vectors of the judgment matrices B1, B2, and B3 are calculated respectively, and the weight vectors are:
[0095]
[0096] After constructing the judgment matrix and weight vector of the solution layer relative to the criterion layer, the judgment matrices B1, B2, and B3 and their weight vectors are subjected to consistency testing to verify their validity. The consistency test results of the judgment matrices of the solution layer relative to the criterion layer are shown in Table 8.
[0097] Table 8 Consistency test results of the importance relationship between the solution layer and the criterion layer
[0098]
[0099] As shown in Table 8, the CRs are all less than 0.1, and the consistency test passes, proving that the importance relationship of each evaluation index in the target layer relative to each performance in the criterion layer is valid.
[0100] Step 5: Build a transposed matrix based on the evaluation indicators. Determine the final AEB system comprehensive performance evaluation model based on the transposed matrix, the weight vector of the judgment matrix of the solution layer relative to the criterion layer, and the weight vector of the judgment matrix of the criterion layer relative to the target layer.
[0101] The transposed matrix is established by five types of evaluation indicators, and the transposed matrix w t =(d br ,a br ,v c ,t br_in ,j) T .
[0102] The final AEB system comprehensive performance evaluation model is determined based on the transposed matrix and the weight vector of the judgment matrix of the criterion layer relative to the target layer, as well as the weight vector of the judgment matrix of the solution layer relative to the criterion layer; the final AEB system comprehensive performance evaluation model is:
[0103] EM tc =ω1×w B1 w t +ω2×w B2 w t +ω3×w B3 w t ,
[0104] The braking distance range is 0 to 100m, and the braking deceleration range is 1 to 10m / s. 2 The collision speed range is 0-120km / h, the braking intervention time is 0-5s, and the average acceleration change rate range is 1-10m / s 3 .
[0105] To facilitate testing and evaluation, the five evaluation indicators are standardized to [0, 1]. Among them, the acceleration change rate mainly reflects the driving comfort, and the value is generally 1 to 3 m / s in the comfortable scene. 3 In emergency situations, the value is generally 5 to 10 m / s. 3Therefore, the smaller the value, the higher the comfort; the braking distance, collision speed and acceleration change rate are considered to be as small as possible, so they are calculated according to (1-standardized value of evaluation index); the longer the braking intervention time, the greater the possibility of collision avoidance, so the larger the value, the better, and the normal standardized value is used during calculation.
[0106] Step 6: Obtain a comprehensive performance score for the AEB system by inputting the evaluation index values, vehicle speed, and road adhesion coefficient under actual driving conditions.
[0107] Based on the data of braking distance, braking deceleration, collision speed, AEB system braking intervention time and acceleration change rate under actual working conditions, the transposed matrix w is obtained. t , substitute the vehicle speed and road adhesion coefficient into the judgment matrix of the criterion layer relative to the target layer to calculate the weight vector w A , based on the final AEB system comprehensive performance evaluation model EM tc The comprehensive performance score of the AEB system is calculated, which can intuitively reflect the comprehensive performance of the AEB system in actual working conditions and realize the evaluation of the comprehensive performance of the AEB system.
[0108] In addition, this embodiment also provides a comprehensive performance evaluation system for an automobile automatic emergency braking system, including:
[0109] An initial model construction module is used to construct an AEB system comprehensive performance evaluation model, which includes a target layer, a criterion layer, and a solution layer; the target layer is the AEB system comprehensive performance score; the criterion layer is the AEB system performance standard; and the solution layer is the AEB system evaluation index;
[0110] A dynamic weighting module prioritizes the performance criteria of the AEB system in different driving scenarios based on the criteria layer, and establishes a weighting function for the criteria layer relative to the target layer based on the vehicle speed and road adhesion coefficient in different driving scenarios.
[0111] The criterion layer weight comparison module is used to compare the weight functions of the performance criteria of the criterion layer with those of the target layer, construct a table of the importance of the criterion layer relative to the target layer, obtain the judgment matrix of the criterion layer, calculate the weight vector of the judgment matrix, and standardize it;
[0112] The scheme-level weight comparison module is used to compare each evaluation indicator of the scheme level with each performance standard of the criterion level, construct the importance relationship table of each evaluation indicator of the scheme level with respect to the criterion level, obtain the judgment matrix of the scheme level, and calculate the weight vector of the judgment matrix of the scheme level;
[0113] An evaluation model generation module is used to establish a transposed matrix based on the evaluation indicators and determine the final AEB system comprehensive performance evaluation model based on the transposed matrix and the weight vector of the judgment matrix of the solution layer relative to the criterion layer, as well as the weight vector of the judgment matrix of the criterion layer relative to the target layer;
[0114] The scoring calculation module is used to obtain the comprehensive performance score of the AEB system by inputting the evaluation index values, vehicle speed and road adhesion coefficient under actual driving conditions.
[0115] Secondly, this embodiment also provides a comprehensive performance evaluation device for an automobile automatic emergency braking system, including a processor and a memory, wherein the processor implements the above-mentioned comprehensive performance evaluation method for an automobile automatic emergency braking system when executing a computer program stored in the memory.
[0116] This embodiment further provides a computer-readable storage medium for storing a computer program, wherein when the computer program is executed by a processor, the method for evaluating the comprehensive performance of an automatic emergency braking system of an automobile as described above is implemented.
[0117] Example 2
[0118] In combination with the various steps in Example 1, in order to realize the performance test of the AEB system of the intelligent connected vehicle, as shown in FIG. Figure 2 As shown, this embodiment uses an AEB system performance test bench to test the operating data of a vehicle's AEB system. By loading and storing the comprehensive performance evaluation method for an automatic emergency braking system into a host program of the AEB system performance test bench, the host computer can be used to operate an intelligent connected vehicle for testing, thereby intuitively obtaining test data for the AEB system performance test.
[0119] This example pre-determines the vehicle parameters of the vehicle under test, including wheelbase, drive mode, braking system, maximum power, AEB system safety configuration, AEB system sensor type, number of sensors, sensor detection range, and detection distance. A target dummy is placed at the front of the AEB system performance test bench. The target dummy's surface resembles the appearance of a real pedestrian, enabling it to be sensed by the sensors and meeting various AEB system test scenarios. The test scenarios and test conditions are then constructed, as shown in Table 9.
[0120] Table 9 Test condition settings
[0121]
[0122] When testing a vehicle on an AEB system performance test bench, the test bench's rotational inertia simulates the vehicle's inertial drag while driving on the road. The core principle is kinetic energy equivalence (translational mass is equal to mechanical inertia). Therefore, to ensure test accuracy and reliability, the vehicle's curb mass (m) (gross vehicle mass + driver weight) must be converted to the rotational inertia of the test bench's roller and flywheel before testing.
[0123] The vehicle under test was weighed to have a mass of 1864 kg. The test bench's fixed mechanical inertia equivalent mass is 1021 kg, and the test bench's three independently controllable mechanical flywheels each have an equivalent mass of 220 kg. The AC dynamometer (which continuously controls its output torque, enabling precise simulation of the test bench's translational inertia) can simulate a translational inertia equivalent mass range of -220 kg to 220 kg. Therefore, during the test preparation phase, it was necessary to engage the three flywheels (with a total equivalent mass of 660 kg) and compensate for the remaining equivalent mass (163 kg) with electrical inertia to achieve an equivalent match between the vehicle's translational mass and the test bench's mechanical inertia.
[0124] Once the test begins, the AEB system performance test bench reads real-time vehicle speed, wheel speed, braking deceleration, braking time, and braking distance data, providing the basis for scoring the braking distance, braking deceleration, collision speed, AEB system braking intervention time, and acceleration rate of change at the solution level. Braking deceleration is output as MFDD to reflect the average level and stability of vehicle deceleration during braking. A larger MFDD indicates a greater braking deceleration during braking. The average acceleration rate of change describes the severity of acceleration changes over time, critically assessing the smoothness or comfort of the motion. A smaller jerk indicates a smoother acceleration change, resulting in less perceived "bumping" or "shoving" sensation. Therefore, the average acceleration rate of change is output. During the test, the entire vehicle is recorded to observe the vehicle warning and braking moments, as well as real-time vehicle speed changes, to ensure the accuracy of the collision speed and AEB system braking intervention time results.
[0125] To ensure the reliability of the data, each working condition was tested twice in each test scenario, and the test data and test results were extracted and analyzed. The test results under each working condition are shown in Table 10.
[0126] Table 10 Analysis of test results under various test scenarios and working conditions
[0127]
[0128] like Figure 3 The following table shows the test bench diagrams of the AEB system performance test bench under different working conditions. Figure 3Data analysis revealed that MFDD and average acceleration rate of change increased with increasing vehicle speed, resulting in reduced braking smoothness and poor ride comfort. At speeds of 20 km / h and 30 km / h, the AEB system intervened relatively early, accompanied by higher braking intensity, effectively avoiding a collision. However, in an emergency situation involving a pedestrian suddenly crossing the road, when the vehicle speed increased to 40 km / h, the AEB system's intervention time, after detecting a potential collision risk, was shortened to 0.81 seconds, insufficient time to complete the collision avoidance maneuver. The final calculated collision speed was 22.7 km / h.
[0129] Based on the bench test data of the AEB system performance test bench, the test data was substituted into the AEB system comprehensive performance evaluation model. By standardizing each evaluation index, the standardized data of each test condition and evaluation index are shown in Table 11.
[0130] Table 11 Standardized results of each evaluation index
[0131]
[0132] Substituting the above data into the comprehensive performance evaluation model of the AEB system, the scores of the solution layer relative to the criterion layer are obtained, as shown in Table 12.
[0133] Table 12 Score of solution layer relative to criterion layer
[0134]
[0135] According to the vehicle speed and road adhesion coefficient under different driving conditions, the judgment matrix A and weight vector of each performance standard of the criterion layer are calculated respectively, which are:
[0136] When (v,μ) is (0.167, 0.875), the judgment matrix A is:
[0137]
[0138] After calculating and normalizing the weight vector of the judgment matrix A, it is recorded as:
[0139]
[0140] When (v,μ) is (0.250, 0.875), the judgment matrix A is:
[0141]
[0142] After calculating and normalizing the weight vector of the judgment matrix A, it is recorded as:
[0143]
[0144] When (v, μ) is (0.333, 0.875), the judgment matrix A is:
[0145]
[0146] After calculating and normalizing the weight vector of the judgment matrix A, it is recorded as:
[0147]
[0148] Based on the weight vectors of each judgment matrix under different driving conditions, the score of the criterion layer relative to the target layer is obtained, as shown in Table 13. The comprehensive performance score of the AEB system under different tests is calculated based on the weight vector wA and the data of braking distance, braking deceleration, collision speed, AEB system braking intervention time and acceleration change rate, as shown in Table 14.
[0149] Table 13 Scores of the criterion layer relative to the target layer
[0150]
[0151] Table 14 Comprehensive performance scores of AEB system under different tests
[0152]
Claims
1. A comprehensive performance evaluation method for an automobile automatic emergency braking system, characterized in that: include: Constructing an AEB system comprehensive performance evaluation model, comprising a target layer, a criterion layer, and a solution layer; the target layer is the AEB system comprehensive performance score; the criterion layer is the AEB system performance standard; and the solution layer is the AEB system evaluation index; Based on the performance standards of the AEB system in the criterion layer, the priorities of each performance standard in different driving scenarios are divided, and the weight function of the criterion layer relative to the target layer is established based on the vehicle speed and road adhesion coefficient in different driving scenarios; Compare each performance criterion of the criterion layer with the weight function of the target layer in pairs, construct the importance relationship table of the criterion layer relative to the target layer, obtain the judgment matrix of the criterion layer, calculate the weight vector of the judgment matrix, and standardize it; Compare each evaluation indicator of the solution layer with each performance standard of the criterion layer in pairs, build the importance relationship table of each evaluation indicator of the solution layer with respect to the criterion layer, obtain the judgment matrix of the solution layer and calculate the weight vector of the judgment matrix of the solution layer; A transposed matrix is established based on the evaluation indicators. The final AEB system comprehensive performance evaluation model is determined based on the transposed matrix, the weight vector of the judgment matrix of the solution layer relative to the criterion layer, and the weight vector of the judgment matrix of the criterion layer relative to the target layer. By inputting the evaluation index values, vehicle speed and road adhesion coefficient under actual driving conditions, the comprehensive performance score of the AEB system is obtained.
2. The comprehensive performance evaluation method of an automobile automatic emergency braking system according to claim 1, characterized in that: The final AEB system comprehensive performance evaluation model is: EM tc =ω1·w B1 ·In t +ω2·w B2 ·In t +ω3·w B3 ·In t , Where ω1 represents the weight vector of AEB system safety relative to the comprehensive performance of the AEB system; ω2 represents the weight vector of AEB system reliability relative to the comprehensive performance of the AEB system; ω3 represents the weight vector of driving comfort relative to the comprehensive performance of the AEB system; w t is the transposed matrix of the evaluation index; w B1 is the weight vector of the scheme-level evaluation index relative to the judgment matrix of the AEB system safety at the criterion level; w B2 is the weight vector of the scheme-level evaluation index relative to the judgment matrix of the AEB system reliability at the criterion level; w B3 is the weight vector of the judgment matrix of the scheme-level evaluation index relative to the criterion-level driving comfort.
3. The comprehensive performance evaluation method of an automobile automatic emergency braking system according to claim 1, characterized in that: The formula of the weight function of the criterion layer relative to the target layer is: S ij (v,μ)=a i v+b i m+c i (≥0), Among them, a i is the speed correspondence layer weight S i The influence coefficient of b i is the road adhesion coefficient criterion layer weight S i The influence coefficient of i is the base constant.
4. The comprehensive performance evaluation method of an automobile automatic emergency braking system according to claim 1, characterized in that: The performance standards considered in the comprehensive performance evaluation of the AEB system include AEB system safety, AEB system reliability and driving comfort.
5. The comprehensive performance evaluation method of an automobile automatic emergency braking system according to claim 1, characterized in that: The evaluation index of the AEB system includes braking distance d br , braking deceleration a br , collision speed v c , AEB system braking intervention time t br_in , acceleration change rate j.
6. The comprehensive performance evaluation method of an automobile automatic emergency braking system according to claim 2, characterized in that: The weight vectors of the judgment matrix of the solution layer relative to the criterion layer are:
7. The comprehensive performance evaluation method of an automobile automatic emergency braking system according to claim 1, characterized in that: After constructing the importance relationship table of the criterion layer relative to the target layer to obtain the judgment matrix of the criterion layer, the method also includes combining data of different vehicle speeds and different road adhesion coefficients in pairs, substituting the combined data into the judgment matrix, calculating the weight vector of each judgment matrix, and performing consistency test.
8. A comprehensive performance evaluation system for an automobile automatic emergency braking system, used to implement the comprehensive performance evaluation method for an automobile automatic emergency braking system according to claim 1, characterized in that: include: An initial model construction module is used to construct an AEB system comprehensive performance evaluation model, which includes a target layer, a criterion layer, and a solution layer; the target layer is the AEB system comprehensive performance score; the criterion layer is the AEB system performance standard; and the solution layer is the AEB system evaluation index; A dynamic weighting module prioritizes the performance criteria of the AEB system in different driving scenarios based on the criteria layer, and establishes a weighting function for the criteria layer relative to the target layer based on the vehicle speed and road adhesion coefficient in different driving scenarios. The criterion layer weight comparison module is used to compare the weight functions of the performance criteria of the criterion layer with those of the target layer, construct a table of the importance of the criterion layer relative to the target layer, obtain the judgment matrix of the criterion layer, calculate the weight vector of the judgment matrix, and standardize it; The scheme-level weight comparison module is used to compare each evaluation indicator of the scheme level with each performance standard of the criterion level, construct the importance relationship table of each evaluation indicator of the scheme level with respect to the criterion level, obtain the judgment matrix of the scheme level, and calculate the weight vector of the judgment matrix of the scheme level; An evaluation model generation module is used to establish a transposed matrix based on the evaluation indicators and determine the final AEB system comprehensive performance evaluation model based on the transposed matrix and the weight vector of the judgment matrix of the solution layer relative to the criterion layer, as well as the weight vector of the judgment matrix of the criterion layer relative to the target layer; The scoring calculation module is used to obtain the comprehensive performance score of the AEB system by inputting the evaluation index values, vehicle speed and road adhesion coefficient under actual driving conditions.
9. A comprehensive performance evaluation device for an automobile automatic emergency braking system, characterized in that: The method comprises a processor and a memory, wherein when the processor executes the computer program stored in the memory, the method for evaluating the comprehensive performance of an automatic emergency braking system of an automobile as claimed in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that Used to store a computer program, wherein when the computer program is executed by a processor, a comprehensive performance evaluation method for an automatic emergency braking system of an automobile as described in any one of claims 1 to 7 is implemented.