Vehicle ADAS function automatic evaluation method and device and industrial personal computer

Abstract

The invention discloses a vehicle ADAS function automatic evaluation method and device and an industrial personal computer. The method comprises the steps of obtaining real-time sensing data of a vehicle in an evaluation scene; processing the sensor data by using a sensing model to obtain truth value data; loading a configuration file corresponding to a to-be-tested ADAS function specified by a user, and obtaining a signal fusion rule and a threshold condition; based on the ADAS data and the truth value data, a composite signal is determined in combination with a signal fusion rule; and outputting an evaluation report of the ADAS function to be tested based on the composite signal and the dynamic threshold. According to the method, the ADAS data and the truth value data are utilized, the configuration file is combined, the corresponding composite signal and the dynamic threshold value are determined, evaluation of the ADAS function to be tested is achieved based on the composite signal and the dynamic threshold value, interference of human factors can be avoided, the hardware cost can be reduced, the evaluation process is simple and clear, and the evaluation efficiency of the ADAS function can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of ADAS (Advanced Driving Assistance System), and more particularly to an automated evaluation method, device, and industrial control computer for vehicle ADAS functions. Background Technology

[0002] Automotive ADAS utilizes various sensors installed in the vehicle (millimeter-wave radar, lidar, monocular / dual-lens cameras, and satellite navigation) to continuously sense the surrounding environment while the car is in motion, collect data, identify, detect, and track static and dynamic objects, and combine this data with navigation map data for system calculations and analysis. This allows the driver to anticipate potential dangers, effectively increasing driving comfort and safety. Current methods for evaluating automotive ADAS functions primarily include manual testing, simulation testing, and hardware-in-the-loop testing.

[0003] For manual evaluation methods, engineers verify ADAS functions by manually driving vehicles. While direct, this method is inefficient, costly, and susceptible to human error and safety hazards. Simulation evaluation methods utilize computer simulation software to simulate different driving environments and scenarios to verify ADAS system functionality. This method can cover a wide range of scenarios and reduce evaluation costs, but simulation results may deviate from actual road test results. Hardware-in-the-loop (HIL) evaluation methods connect the actual ECU (Electronic Control Unit) to a simulation environment and evaluate ADAS functions through real-time simulation. This method improves evaluation accuracy, but the equipment is complex and costly, and still requires verification through actual road testing. Summary of the Invention

[0004] This application provides an automated evaluation method, device, and industrial control computer for vehicle ADAS functions, with the aim of improving the evaluation efficiency of ADAS functions.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] An automated evaluation method for vehicle ADAS functions includes:

[0007] Obtain real-time perception data of the vehicle in the test scenario; the real-time perception data includes sensor data collected by sensors pre-installed on the vehicle, and ADAS data output by the vehicle's advanced driver assistance system.

[0008] The sensor data is processed using a perception model to obtain ground truth data; the ground truth data is used to reflect the targets and lane lines encountered by the vehicle during driving.

[0009] Load the configuration file corresponding to the ADAS function under test specified by the user, and obtain the corresponding signal fusion rules and threshold conditions; the signal fusion rules include the calculation expression of the composite signal related to the ADAS function under test; the threshold conditions include dynamic thresholds determined based on external input signals;

[0010] Based on the ADAS data and the true data, and in conjunction with the signal fusion rules, the corresponding composite signal is determined;

[0011] Based on the composite signal and the dynamic threshold, an evaluation report of the ADAS function under test is output.

[0012] Optionally, based on the ADAS data and the ground truth data, and in conjunction with the signal fusion rules, the corresponding composite signal is determined, including:

[0013] Based on the signal fusion rules, a calculation expression for the composite signal is determined; the calculation expression includes at least one calculation parameter.

[0014] Determine the first parameter value for each of the calculation parameters from the ADAS data;

[0015] Determine the second parameter value of each of the calculation parameters from the true value data;

[0016] Based on the first parameter values ​​of each of the calculation parameters, the first signal value of the composite signal is determined;

[0017] The second signal value of the composite signal is determined based on the second parameter values ​​of each of the calculated parameters.

[0018] Optionally, based on the composite signal and the dynamic threshold, an evaluation report of the ADAS function under test is output, including:

[0019] Determine the difference between the first signal value and the second signal value of the composite signal;

[0020] Based on the difference and the dynamic threshold, monitor whether the ADAS function under test is abnormal;

[0021] When the ADAS function under test is detected to have an abnormality, the corresponding first timestamp is recorded, until the ADAS function under test is detected to have recovered from the abnormality, the corresponding second timestamp is recorded.

[0022] An evaluation report for the ADAS function under test is generated based on the first timestamp and the second timestamp.

[0023] Optionally, based on the difference and the dynamic threshold, if the difference meets the dynamic threshold, it is determined that the ADAS function under test has experienced the abnormality; if the difference does not meet the dynamic threshold, it is determined that the ADAS function under test has not experienced the abnormality.

[0024] Optionally, if the ADAS function to be tested is an AEB function, the calculation expression for the corresponding composite signal is (△V+f(a))*g(θ)-h(v), where △V represents relative velocity, f, g, and h are custom signal conversion functions, a represents the vehicle's acceleration, θ represents the road gradient, and v represents the vehicle speed.

[0025] Optionally, if the ADAS function to be tested is ACC function, the corresponding composite signal includes S1 and S2; S1 is used to determine the relationship between vehicle distance and speed difference and the calculation expression of S1 is (ΔV / v_lim)+(D_min / D_current), where ΔV represents relative speed, v_lim represents the preset speed limit value, D_min represents the preset minimum safe vehicle distance, and D_current represents vehicle distance; S2 is used to determine the influence of road slope and the calculation expression of S2 is S2=θ / θ_max, where θ represents road slope and θ_max represents the preset maximum allowable slope.

[0026] Optionally, the configuration file includes a pre-defined relational rule table or rule formula, wherein the relational rule table includes multiple sample signal values ​​of the external input signal and the corresponding sample dynamic thresholds;

[0027] The process of determining the dynamic threshold based on the external input signal includes one of the following:

[0028] Based on the signal value of the external input signal, the relationship rule table is queried to determine the corresponding dynamic threshold;

[0029] The corresponding dynamic threshold is determined by substituting the signal value of the external input signal into the rule relationship.

[0030] An automated testing device for vehicle ADAS functions, comprising:

[0031] The real-time perception unit is used to obtain real-time perception data of the vehicle in the evaluation scenario; the real-time perception data includes sensor data collected by sensors pre-installed on the vehicle, and ADAS data output by the vehicle's advanced driver assistance system.

[0032] The data processing unit is used to process the sensor data using a perception model to obtain ground truth data; the ground truth data is used to reflect the targets and lane lines encountered by the vehicle during driving.

[0033] The file loading unit is used to load the configuration file corresponding to the ADAS function under test specified by the user, and obtain the corresponding signal fusion rules and threshold conditions; the signal fusion rules include the calculation expression of the composite signal related to the ADAS function under test; the threshold conditions include dynamic thresholds determined based on external input signals.

[0034] The signal determination unit is used to determine the corresponding composite signal based on the ADAS data and the truth data, combined with the signal fusion rules;

[0035] The report determination unit is used to output an evaluation report of the ADAS function under test based on the composite signal and the dynamic threshold.

[0036] A storage medium comprising a stored program, wherein the program is executed by a processor to perform the automated evaluation method for vehicle ADAS functions.

[0037] An industrial control computer includes: a processor, a memory, and a bus; the processor and the memory are connected via the bus.

[0038] The memory is used to store the program, and the processor is used to run the program, wherein the program is executed by the processor to perform the automated evaluation method for vehicle ADAS functions.

[0039] The technical solution provided in this application obtains real-time perception data of the vehicle in the testing scenario, processes the sensor data using a perception model to obtain ground truth data, loads the configuration file corresponding to the ADAS function to be tested specified by the user, and obtains the corresponding signal fusion rules and threshold conditions. Based on the ADAS data and ground truth data, combined with the signal fusion rules, the corresponding composite signal is determined. Based on the composite signal and dynamic threshold, the test report of the ADAS function to be tested is determined. This application utilizes ADAS data and ground truth data, combined with the configuration file, to determine the corresponding composite signal and dynamic threshold, and achieves the test of the ADAS function to be tested based on the composite signal and dynamic threshold. This avoids interference from human factors and reduces hardware costs, and the test process is relatively simple and clear, effectively improving the test efficiency of ADAS functions. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 A flowchart illustrating an automated evaluation method for vehicle ADAS functions provided in this application embodiment;

[0042] Figure 2 A flowchart illustrating another automated evaluation method for vehicle ADAS functions provided in this application embodiment;

[0043] Figure 3 A flowchart illustrating another automated evaluation method for vehicle ADAS functions provided in this application embodiment;

[0044] Figure 4 A schematic diagram of the architecture of an automated evaluation device for vehicle ADAS functions provided in this application embodiment;

[0045] Figure 5 A hardware connection diagram provided for an embodiment of this application;

[0046] Figure 6 This is a schematic diagram of an automated evaluation logic provided in an embodiment of this application. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0049] like Figure 1 The diagram shown is a flowchart of an automated evaluation method for vehicle ADAS functions provided in this application embodiment. It can be applied to an industrial control computer, which can be installed on a vehicle, and includes the following steps.

[0050] S101: Obtain real-time perception data of the vehicle in the evaluation scenario.

[0051] Real-time perception data includes sensor data collected by sensors pre-installed on the vehicle, as well as ADAS data output by the vehicle's advanced driver assistance systems.

[0052] In some examples, the evaluation scenarios include those designed to simulate and cover a variety of complex driving situations.

[0053] In some examples, the industrial control computer is connected to the signal output ports of both the sensors and the vehicle's ADAS. For specific hardware connection methods, please refer to [link / reference needed]. Figure 5 As shown.

[0054] In some examples, the number of sensors pre-installed in the vehicle is multiple, including but not limited to LiDAR, cameras, and combined inertial navigation sensors.

[0055] In some examples, ADAS data includes, but is not limited to, environmental perception data, vehicle status data, and driving behavior data.

[0056] In possible implementations, environmental perception data can be used to achieve target detection (e.g., identifying vehicles, pedestrians, bicycles, etc.), distance measurement (e.g., measuring the distance to vehicles or obstacles ahead), lane recognition (e.g., identifying lane line positions and relative vehicle positions), traffic sign recognition (e.g., identifying speed limit, parking, and other sign information), and road condition recognition (e.g., identifying slippery, snowy, and other road surfaces).

[0057] In possible implementations, vehicle status data includes, but is not limited to, vehicle speed (i.e., the current speed of the vehicle), acceleration (including longitudinal acceleration and lateral acceleration), steering angle (i.e., steering wheel angle), yaw rate (i.e., the speed at which the vehicle rotates about its vertical axis), and tilt angle (i.e., the degree of vehicle tilt).

[0058] In possible implementations, driving behavior data includes, but is not limited to, driver status (e.g., fatigue, distraction), steering wheel grip (e.g., driver grip strength), and pedal operation (including accelerator and brake usage).

[0059] It should be noted that the industrial control computer can run preset data acquisition software (e.g., Figure 5 The INTEWORK-ARS shown here collects sensor data and ADAS data in real time, thereby enabling real-time monitoring of perception data.

[0060] S102: Using a perception model, process sensor data to obtain true data.

[0061] The true data is used to reflect the objects and lane lines encountered by the vehicle during driving.

[0062] In some examples, the type of perception model can be a multimodal perception fusion model, such as the CLIP (Contrastive Language-Image Pre-Training) model. The CLIP model has high-precision multimodal fusion capabilities, which makes it possible to provide the semantic space of sensor data collected by multiple sensors.

[0063] In some examples, ground truth data includes, but is not limited to, the size, velocity, and acceleration of the target object, as well as the lane equation and lane position corresponding to the lane lines. Generally speaking, ground truth data can be regarded as an objective reference for evaluating ADAS functions.

[0064] In possible implementations, the type of target object includes, but is not limited to, vehicles, pedestrians, obstacles, etc.

[0065] S103: Load the configuration file corresponding to the ADAS function to be tested specified by the user, and obtain the corresponding signal fusion rules and threshold conditions.

[0066] The signal fusion rules include calculation expressions for composite signals related to the ADAS function under test, and the threshold conditions include dynamic thresholds determined based on external input signals.

[0067] In some examples, ADAS can provide multiple ADAS functions, including but not limited to AEB (Autonomous Emergency Braking), ACC (Adaptive Cruise Control), and LKA (Lane Keeping Assist).

[0068] It should be noted that each ADAS function has a pre-defined configuration file, which contains corresponding signal fusion rules and threshold conditions. Generally speaking, the configuration files for each ADAS function are different.

[0069] In some examples, signal fusion rules include fusion rules for composite signals (e.g., operations such as addition, subtraction, multiplication, division, differentiation, weighted averaging, logical combination, etc., on multiple computational parameters).

[0070] In some examples, the types of external input signals include, but are not limited to, vehicle speed and traffic flow density. It is understood that when a change in the detected external input signal is detected, the dynamic threshold indicated by the threshold condition will also change.

[0071] In a possible implementation, the dynamic threshold corresponding to the ADAS function under test is determined based on the vehicle speed. When the vehicle speed increases, the dynamic threshold increases accordingly, and when the vehicle speed decreases, the dynamic threshold decreases accordingly.

[0072] In some examples, the dynamic threshold corresponding to the external input signal can be determined using a relational rule table (which includes multiple sample signal values ​​of the external input signal and the corresponding sample dynamic thresholds).

[0073] In a possible implementation, when a change in the external input signal is detected, the corresponding dynamic threshold can be calculated according to the relational rule table shown in the configuration file, thereby constructing threshold conditions that better fit the actual needs and enabling more granular and flexible evaluation of the ADAS function under test.

[0074] In a possible implementation, a custom rule relationship for the dynamic threshold can be defined in the configuration file, and the dynamic threshold corresponding to the external input signal can be calculated using the rule relationship.

[0075] Optionally, the process of determining the dynamic threshold based on the external input signal includes the following: querying the relational rule table according to the signal value of the external input signal to determine the corresponding dynamic threshold; or substituting the signal value of the external input signal into the rule relational formula to determine the corresponding dynamic threshold.

[0076] In some examples, the ADAS function under test is the LKA function, and the dynamic threshold corresponding to the LKA function is the lane departure dynamic threshold. Generally speaking, when the vehicle speed increases, the allowable lane departure dynamic threshold needs to be more stringent (i.e., smaller deviation) to ensure safety. When the vehicle speed decreases, the allowable lane departure dynamic threshold is relaxed. Therefore, the rule relationship of the lane departure dynamic threshold can be found in formula (1):

[0077] Lane departure dynamic threshold = base threshold - k*v (1).

[0078] In formula (1), the base threshold is a custom threshold, k represents the adjustment factor, and v represents the vehicle speed. When the vehicle speed increases, the lane departure dynamic threshold automatically decreases, thereby achieving dynamic adjustment.

[0079] In some examples, the ADAS function under test is the AEB function, and the dynamic threshold corresponding to the AEB function is the AEB trigger distance threshold. When the vehicle is traveling at a higher speed on the highway, the trigger distance of AEB should be larger to ensure sufficient reaction time. Therefore, the rule relationship of the AEB trigger distance threshold can be found in formula (2):

[0080] AEB trigger distance threshold = D_base + α*v (2).

[0081] In formula (2), D_base is a custom base distance threshold, α represents the adjustment coefficient, and v represents the vehicle speed. When the vehicle speed increases, the AEB trigger distance threshold increases accordingly to cope with more urgent braking needs.

[0082] S104: Based on ADAS data and true data, combined with signal fusion rules, determine the corresponding composite signal.

[0083] Among them, the composite signal can be regarded as an objective reference for evaluating whether the ADAS function under test is normal.

[0084] Optionally, based on ADAS data and ground truth data, and combined with signal fusion rules, the implementation process of the corresponding composite signal can be determined. (See [link to relevant documentation]). Figure 2 The steps shown are accompanied by corresponding explanations.

[0085] S105: Based on composite signals and dynamic thresholds, outputs an evaluation report of the ADAS function under test.

[0086] After obtaining the composite signal and the dynamic threshold, the composite signal and the dynamic threshold can be compared to determine whether the ADAS function under test has malfunctioned, and the corresponding evaluation report can be determined based on the determination result.

[0087] Optionally, the implementation process for determining the evaluation report of the ADAS function under test based on composite signals and dynamic thresholds can be found in [reference needed]. Figure 3 The steps shown are accompanied by corresponding explanations.

[0088] In some examples, combined Figure 2 as well as Figure 3 The method shown can be simply summarized as follows: Figure 6 As shown, Figure 6 The ADAS evaluation logic shown can be regarded as a general application of S104 and S105.

[0089] Compared with existing technologies, the vehicle ADAS function automated evaluation method shown in this application has the following outstanding effects: (1) High efficiency: By using automated evaluation tools and platforms (which can be understood as industrial control computers), the efficiency of vehicle ADAS function evaluation is greatly improved, the evaluation cycle is shortened, and rapid iteration (which can be achieved by adjusting the configuration file) and large-scale verification (multiple ADAS functions under test can be supported at the same time); (2) Cost control: Industrial control computers are applicable to vehicles of any model, have good hardware reusability, reduce the dependence on a large number of manpower and hardware resources, and effectively reduce evaluation costs; (3) Comprehensive coverage: It can simulate and cover various complex driving scenarios, including extreme and rare working conditions, and comprehensively evaluate the performance and reliability of ADAS; (4) Consistency and objectivity: Through standardized evaluation processes and automated evaluation tools, human interference can be eliminated, ensuring the consistency and objectivity of evaluation results; (5) Reproducibility: It can record and reproduce the evaluation process, which is convenient for analyzing and verifying the evaluation results and improving the credibility of the evaluation.

[0090] The process described in S101-S105 above utilizes the vehicle's ADAS data and true data, combined with the configuration file corresponding to the ADAS function under test, to determine the corresponding composite signal and dynamic threshold. Based on the composite signal and dynamic threshold, the ADAS function under test is evaluated. This avoids interference from human factors and reduces hardware costs. Moreover, the evaluation process is relatively simple and clear, which can effectively improve the evaluation efficiency of ADAS functions.

[0091] like Figure 2 The diagram shown is a flowchart of another automated evaluation method for vehicle ADAS functions provided in this application, including the following steps.

[0092] S201: Determine the calculation expression of the composite signal based on the signal fusion rules.

[0093] The calculation expression includes at least one calculation parameter.

[0094] Optionally, the number of composite signals may include at least one.

[0095] Optionally, if the ADAS function under test is an AEB function, the calculation expression for the corresponding composite signal can be found in formula (3):

[0096] Composite signal = (△V + f(a)) * g(θ) - h(v) (3).

[0097] In formula (3), ΔV represents relative velocity (i.e., the relative velocity between the vehicle and the target in front), f, g, and h are user-defined signal conversion functions, a represents the vehicle's acceleration, θ represents the road gradient, and v represents the vehicle speed. Generally speaking, ΔV, a, θ, and v are all calculation parameters.

[0098] Optionally, if the ADAS function to be tested is the ACC function, the corresponding composite signal includes S1 and S2. The calculation expression of S1 can be found in formula (4), and the calculation expression of S2 can be found in formula (5).

[0099] S1=(ΔV / v_lim)+(D_min / D_current) (4).

[0100] In formula (4), S1 is used to determine the relationship between vehicle distance (i.e., the distance between the vehicle and the target in front) and speed difference, ΔV represents relative speed (i.e., the relative speed between the vehicle and the target in front), v_lim represents the preset speed limit value, D_min represents the preset minimum safe vehicle distance, and D_current represents the vehicle distance. Generally speaking, ΔV, v_lim, D_min, and D_current are all calculation parameters. When S1 exceeds the threshold, it means that the speed difference or vehicle distance is too tight.

[0101] S2=θ / θ_max (5).

[0102] In formula (5), S2 is used to determine the influence of road slope, θ represents road slope, and θ_max represents the preset maximum allowable slope.

[0103] It is understandable that the composite signals corresponding to the ACC function are S1 and S2. Therefore, it is necessary to superimpose S1 and S2 to determine whether the ACC function is abnormal. In a possible implementation, S1 corresponds to the dynamic threshold T1 and S2 corresponds to the dynamic threshold T2. When S1 meets the dynamic threshold T1 (e.g., S1 > T1) and S2 meets the dynamic threshold T2 (e.g., S2 > T2), it is determined that the ACC function is abnormal, indicating that the ACC performance is poor under the conditions of steep slope and excessively tight distance and speed difference.

[0104] It should be noted that the number of composite signals corresponding to the ADAS function under test can be multiple. By combining the dynamic threshold corresponding to each composite signal, more detailed judgment results that are closer to actual working conditions can be provided for the evaluation of the ADAS function under test, effectively improving the reliability of the evaluation of the ADAS function under test.

[0105] S202: Determine the first parameter value for each calculated parameter from the ADAS data.

[0106] ADAS data contains multiple parameters and their corresponding values, and the first parameter value of each calculated parameter can be determined from the ADAS data.

[0107] S203: Determine the second parameter value of each calculation parameter from the true value data.

[0108] The true value data contains multiple parameters and their corresponding values, and the second parameter value of each calculation parameter can be determined from the true value data.

[0109] S204: Determine the first signal value of the composite signal based on the first parameter value of each calculation parameter.

[0110] In this process, after determining the first parameter value of each calculation parameter, the first parameter value is substituted into the calculation expression of the composite signal to obtain the first signal value of the composite signal.

[0111] S205: Determine the second signal value of the composite signal based on the second parameter values ​​of each calculation parameter.

[0112] Specifically, after determining the second parameter values ​​of each calculation parameter, the second parameter values ​​are substituted into the calculation expression of the composite signal to obtain the second signal value of the composite signal.

[0113] The processes shown in S201-S205 above can determine the corresponding composite signal based on ADAS data and true data, combined with signal fusion rules, providing an objective reference for the evaluation of the ADAS function under test.

[0114] like Figure 3 The diagram shown is a flowchart of another automated evaluation method for vehicle ADAS functions provided in this application, including the following steps.

[0115] S301: Determine the difference between the first signal value and the second signal value of the composite signal.

[0116] The difference between the first signal value and the second signal value can be obtained by subtracting the first signal value from the second signal value of the composite signal.

[0117] S302: Based on the difference and dynamic threshold, monitor whether the ADAS function under test is abnormal.

[0118] After obtaining the difference between the first signal value and the second signal value, the difference can be compared with the dynamic threshold to determine in real time whether the ADAS function under test has malfunctioned.

[0119] Optionally, if the difference meets the dynamic threshold, it is determined that the ADAS function under test is abnormal; if the difference does not meet the dynamic threshold, it is determined that the ADAS function under test is not abnormal.

[0120] S303: When an abnormality is detected in the ADAS function under test, record the corresponding first timestamp until the ADAS function under test is detected to be out of the abnormality, and record the corresponding second timestamp.

[0121] Specifically, when an abnormality is detected in the ADAS function under test, a corresponding abnormality record can be generated to record the first timestamp. When the ADAS function under test is detected to be out of the abnormality, the corresponding second timestamp is recorded in the abnormality record.

[0122] It should be noted that when the vehicle is driving in the test scenario and the ADAS function under test has been designated for testing, if the difference is detected to meet the dynamic threshold, it is determined that the ADAS function under test has malfunctioned, and the timestamp of the malfunction is recorded as the first timestamp. As the vehicle drives, the difference changes continuously. When the changed difference is detected to no longer meet the dynamic threshold, it is determined that the ADAS function under test has moved out of the malfunction, and the timestamp of moving out of the malfunction is recorded as the second timestamp. The time period of the malfunction can be identified based on the first and second timestamps.

[0123] In some examples, when the detected difference meets the dynamic threshold, it is determined that the ADAS function under test is normal (understood as the function meets the expected requirements), and the corresponding monitoring timestamp is recorded at the same time.

[0124] S304: Generate an evaluation report for the ADAS function under test based on the first and second timestamps.

[0125] When an anomaly is detected in the ADAS function under test, a corresponding anomaly record can be generated, and a first timestamp and a second timestamp can be added to the anomaly record to generate an evaluation report for the ADAS function under test.

[0126] The process shown in S301-S304 above can use the difference between the first signal value and the second signal value of the composite signal, combined with a dynamic threshold, to monitor whether the ADAS function under test is abnormal. This enables the evaluation of the ADAS function under test using ADAS data and true data, ensuring the reliability of the evaluation results. Compared with existing technical means, the evaluation method of the ADAS function under test is simpler and clearer, and can effectively improve the evaluation efficiency.

[0127] like Figure 4 The diagram shown is a schematic of the architecture of an automated evaluation device for vehicle ADAS functions provided in an embodiment of this application, including the following units.

[0128] The real-time perception unit 100 is used to obtain real-time perception data of the vehicle in the evaluation scenario; the real-time perception data includes sensor data collected by sensors pre-installed on the vehicle, as well as ADAS data output by the vehicle's advanced driver assistance system.

[0129] The data processing unit 200 is used to process sensor data using a perception model to obtain ground truth data; the ground truth data is used to reflect the targets and lane lines encountered by the vehicle during driving.

[0130] The file loading unit 300 is used to load the configuration file corresponding to the ADAS function under test specified by the user, and obtain the corresponding signal fusion rules and threshold conditions. The signal fusion rules include the calculation expression of the composite signal related to the ADAS function under test. The threshold conditions include the dynamic threshold determined based on the external input signal.

[0131] Optionally, the configuration file includes a pre-defined relational rule table or rule formula. The relational rule table includes multiple sample signal values ​​of the external input signal and the corresponding sample dynamic thresholds. The file loading unit 300 is specifically used to: query the relational rule table based on the signal value of the external input signal to determine the corresponding dynamic threshold, or substitute the signal value of the external input signal into the rule formula to determine the corresponding dynamic threshold.

[0132] The signal determination unit 400 is used to determine the corresponding composite signal based on ADAS data and true data, combined with signal fusion rules.

[0133] Optionally, the signal determination unit 400 is specifically used to: determine a calculation expression for the composite signal based on signal fusion rules; the calculation expression includes at least one calculation parameter; determine a first parameter value for each calculation parameter from ADAS data; determine a second parameter value for each calculation parameter from truth data; determine a first signal value for the composite signal based on the first parameter value for each calculation parameter; and determine a second signal value for the composite signal based on the second parameter value for each calculation parameter.

[0134] Optionally, if the ADAS function to be tested is the AEB function, the calculation expression of the corresponding composite signal is (△V+f(a))*g(θ)-h(v), where △V represents the relative velocity, f, g and h are custom signal conversion functions, a represents the vehicle acceleration, θ represents the road slope, and v represents the vehicle speed.

[0135] Optionally, if the ADAS function under test is ACC function, the corresponding composite signal includes S1 and S2; S1 is used to determine the relationship between vehicle distance and speed difference and the calculation expression of S1 is (ΔV / v_lim)+(D_min / D_current), where ΔV represents relative speed, v_lim represents the preset speed limit value, D_min represents the preset minimum safe vehicle distance, and D_current represents vehicle distance; S2 is used to determine the influence of road slope and the calculation expression of S2 is S2=θ / θ_max, where θ represents road slope and θ_max represents the preset maximum allowable slope.

[0136] The report determination unit 500 is used to output an evaluation report of the ADAS function under test based on composite signals and dynamic thresholds.

[0137] Optionally, the report determination unit 500 is specifically used to: determine the difference between the first signal value and the second signal value of the composite signal; monitor whether the ADAS function under test is abnormal based on the difference and the dynamic threshold; when an abnormality is detected in the ADAS function under test, record the corresponding first timestamp until the ADAS function under test is detected to be out of the abnormality, and record the corresponding second timestamp; and generate an evaluation report of the ADAS function under test based on the first timestamp and the second timestamp.

[0138] Optionally, the report determination unit 500 is specifically used to: determine that the ADAS function under test is abnormal when the difference meets the dynamic threshold; and determine that the ADAS function under test is not abnormal when the difference does not meet the dynamic threshold.

[0139] The units described above utilize the vehicle's ADAS data and ground truth data, combined with the configuration file corresponding to the ADAS function under test, to determine the corresponding composite signal and dynamic threshold. Based on the composite signal and dynamic threshold, the ADAS function under test is evaluated. This avoids interference from human factors and reduces hardware costs. Moreover, the evaluation process is relatively simple and clear, which can effectively improve the evaluation efficiency of ADAS functions.

[0140] This application also provides a computer-readable storage medium including a stored program, wherein the program executes the automated evaluation method for vehicle ADAS functions provided in this application.

[0141] This application also provides an industrial control computer, including a processor, a memory, and a bus. The processor and the memory are connected via the bus. The memory is used to store programs, and the processor is used to run the programs. When the programs are running, they execute the automated evaluation method for vehicle ADAS functions provided in this application.

[0142] While several specific implementation details are included in the foregoing discussion, these should not be construed as limiting the scope of this application. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0143] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. An automated evaluation method for vehicle ADAS functions, characterized in that, include: Obtain real-time perception data of the vehicle in the testing scenario; The real-time perception data includes sensor data collected by sensors pre-installed on the vehicle, as well as ADAS data output by the vehicle's advanced driver assistance system. The sensor data is processed using a perception model to obtain ground truth data; the ground truth data is used to reflect the targets and lane lines encountered by the vehicle during driving. Load the configuration file corresponding to the ADAS function under test specified by the user, and obtain the corresponding signal fusion rules and threshold conditions; the signal fusion rules include the calculation expression of the composite signal related to the ADAS function under test; the threshold conditions include dynamic thresholds determined based on external input signals; Based on the ADAS data and the true data, and in conjunction with the signal fusion rules, the corresponding composite signal is determined; Based on the composite signal and the dynamic threshold, an evaluation report of the ADAS function under test is output.

2. The method according to claim 1, characterized in that, Based on the ADAS data and the ground truth data, and in conjunction with the signal fusion rules, the corresponding composite signal is determined, including: Based on the signal fusion rules, a calculation expression for the composite signal is determined; the calculation expression includes at least one calculation parameter. Determine the first parameter value for each of the calculation parameters from the ADAS data; Determine the second parameter value of each of the calculation parameters from the true value data; Based on the first parameter values ​​of each of the calculation parameters, the first signal value of the composite signal is determined; The second signal value of the composite signal is determined based on the second parameter values ​​of each of the calculated parameters.

3. The method according to claim 2, characterized in that, Based on the composite signal and the dynamic threshold, an evaluation report of the ADAS function under test is output, including: Determine the difference between the first signal value and the second signal value of the composite signal; Based on the difference and the dynamic threshold, monitor whether the ADAS function under test is abnormal; When the ADAS function under test is detected to have an abnormality, the corresponding first timestamp is recorded, until the ADAS function under test is detected to have recovered from the abnormality, the corresponding second timestamp is recorded. An evaluation report for the ADAS function under test is generated based on the first timestamp and the second timestamp.

4. The method according to claim 3, characterized in that, When the difference meets the dynamic threshold, it is determined that the ADAS function under test has experienced the abnormality; when the difference does not meet the dynamic threshold, it is determined that the ADAS function under test has not experienced the abnormality.

5. The method according to claim 2, characterized in that, If the ADAS function to be tested is the AEB function, the calculation expression of the corresponding composite signal is (△V+f(a))*g(θ)-h(v), where △V represents the relative velocity, f, g and h are custom signal conversion functions, a represents the acceleration of the vehicle, θ represents the road slope, and v represents the vehicle speed.

6. The method according to claim 2, characterized in that, If the ADAS function under test is ACC function, the corresponding composite signal includes S1 and S2; S1 is used to determine the relationship between vehicle distance and speed difference and the calculation expression of S1 is (ΔV / v_lim)+(D_min / D_current), where ΔV represents relative speed, v_lim represents the preset speed limit value, D_min represents the preset minimum safe vehicle distance, and D_current represents vehicle distance; S2 is used to determine the influence of road slope and the calculation expression of S2 is S2=θ / θ_max, where θ represents road slope and θ_max represents the preset maximum allowable slope.

7. The method according to claim 1, characterized in that, The configuration file includes a pre-defined relational rule table or rule formula, and the relational rule table includes multiple sample signal values ​​of the external input signal and the corresponding sample dynamic thresholds; The process of determining the dynamic threshold based on the external input signal includes one of the following: Based on the signal value of the external input signal, the relationship rule table is queried to determine the corresponding dynamic threshold; The corresponding dynamic threshold is determined by substituting the signal value of the external input signal into the rule relationship.

8. An automated testing device for vehicle ADAS functions, characterized in that, include: The real-time perception unit is used to obtain real-time perception data of the vehicle in the evaluation scenario; The real-time perception data includes sensor data collected by sensors pre-installed on the vehicle, as well as ADAS data output by the vehicle's advanced driver assistance system. The data processing unit is used to process the sensor data using a perception model to obtain ground truth data; the ground truth data is used to reflect the targets and lane lines encountered by the vehicle during driving. The file loading unit is used to load the configuration file corresponding to the ADAS function under test specified by the user, and obtain the corresponding signal fusion rules and threshold conditions; the signal fusion rules include the calculation expression of the composite signal related to the ADAS function under test; the threshold conditions include dynamic thresholds determined based on external input signals. The signal determination unit is used to determine the corresponding composite signal based on the ADAS data and the truth data, combined with the signal fusion rules; The report determination unit is used to output an evaluation report of the ADAS function under test based on the composite signal and the dynamic threshold.

9. A storage medium, characterized in that, The storage medium includes a stored program, wherein the program is executed by a processor to perform the automated evaluation method for vehicle ADAS functions according to any one of claims 1-7.

10. An industrial control computer, characterized in that, include: Processor, memory, and bus; The processor and the memory are connected via the bus; The memory is used to store a program, and the processor is used to run the program, wherein the program is executed by the processor to perform the vehicle ADAS function automated evaluation method according to any one of claims 1-7.

Images