Boundary performance testing method and device for automatic emergency braking

Abstract

The invention provides a boundary performance testing method and device for automatic emergency braking. Relates to the technical field of vehicles, a boundary test scene is set based on a standard scene in a vehicle evaluation procedure, a boundary test case is set based on a case of the vehicle evaluation procedure, and the boundary test case comprises test parameters for testing a to-be-tested vehicle sensing target object to trigger an automatic emergency braking system. The method comprises the following steps: acquiring a boundary test case corresponding to each test parameter in a boundary test scene, implementing the boundary test case corresponding to each test parameter in the boundary test scene, acquiring test data of a to-be-tested vehicle corresponding to the test parameters, and finally adjusting an algorithm identification threshold and / or a braking strategy threshold of the to-be-tested vehicle for triggering an emergency braking function based on the test data. Therefore, through the change of the test data of the to-be-tested vehicle corresponding to the plurality of test parameters, the algorithm identification threshold value and the optimization adjustment direction of the braking strategy are determined.

Description

Technical Field

[0001] The present application relates to the technical field of vehicles, and in particular, to a method and device for testing the boundary performance of autonomous emergency braking. Background Art

[0002] Equipped with an Autonomous Emergency Braking System (AEBS), it is an advanced driver assistance system that senses the risk of possible collision with vehicles, pedestrians, or other traffic participants ahead based on environmental perception sensors (such as millimeter-wave radar or cameras), and automatically triggers the actuator through the system to implement braking to avoid collision or reduce the degree of collision.

[0003] Currently, the test scenarios specified by regulations mainly propose performance requirements for the autonomous emergency braking function under specific test scenarios, and provide a direction for performance optimization of the autonomous emergency braking function through test scoring criteria, ensuring the performance and safety of AEBS to a certain extent. However, the regulatory test can only provide the test score as a result reference and cannot provide an optimization direction for the AEB function algorithm. Summary of the Invention

[0004] In view of this, the present application provides a method and device, aiming to solve the problem that the regulatory test can only provide the test score as a result reference and cannot provide an optimization direction for the AEB function algorithm.

[0005] In a first aspect, the present application provides a method for testing the boundary performance of autonomous emergency braking, including:

[0006] Setting a boundary test scenario based on the standard scenario in the vehicle evaluation procedure;

[0007] Setting boundary test cases based on the use cases of the vehicle evaluation procedure, where the boundary test cases include test parameters for testing the target object sensed by the vehicle under test to trigger the autonomous emergency braking system;

[0008] Implementing the boundary test cases corresponding to each test parameter in the boundary test scenario to obtain the test data of the vehicle under test corresponding to the test parameters;

[0009] Adjusting the algorithm recognition threshold and / or braking strategy threshold for the vehicle under test to trigger the emergency braking function based on the test data.

[0010] Optionally, setting a boundary test scenario based on the standard scenario in the vehicle evaluation procedure further includes:

[0011] Configuring weather conditions or road conditions that interfere with the recognition of the vehicle under test as the boundary test scenario based on the standard scenario in the vehicle evaluation procedure.

[0012] Optionally, the test data includes one or more of the boundary data for triggering and not triggering the automatic emergency braking system, the boundary data for not triggering and mis-triggering the automatic emergency braking system, the overall braking curve of the triggered automatic emergency braking system, and the TTC of the triggered automatic emergency braking system; the triggered automatic emergency braking system includes the automatic emergency braking of the triggered automatic emergency braking system, the collision warning of the triggered automatic emergency braking system, and the human-machine interaction alarm of the triggered automatic emergency braking system.

[0013] Optionally, the target is a pedestrian crossing the road, and the test parameters are multiple distance values;

[0014] The use cases based on the vehicle evaluation regulations, and the boundary test cases are set, including:

[0015] The vehicle under test and the pedestrian travel according to the corresponding use case regulations in the vehicle evaluation regulations, and the pedestrian stops when moving to the target distance from the vehicle driving central axis. It is detected whether the vehicle under test triggers the automatic emergency braking system, and the target distance parameter is one of the multiple distance values.

[0016] Optionally, the multiple distance values include multiple first distances decreasing in sequence and multiple second distances decreasing in sequence;

[0017] The obtaining of the test data of the vehicle under test corresponding to the test parameters includes:

[0018] Obtain the forward collision warning in the automatic emergency braking system not triggered by the vehicle under test corresponding to the first target distance, and obtain the forward collision warning in the automatic emergency braking system triggered by the vehicle under test corresponding to the second target distance adjacent to the first target distance; both the first target distance and the second target distance are used as the first boundary data, and the first target distance and the second target distance are two adjacent first distances among the multiple first distances decreasing in sequence;

[0019] Obtain that the vehicle under test does not trigger the automatic emergency braking in the automatic emergency braking system corresponding to the third target distance, and the fourth target distance adjacent to the third target distance triggers the automatic emergency braking in the automatic emergency braking system; both the third target distance and the fourth target distance are used as the second boundary data, and the third target distance and the fourth target distance are two adjacent second distances among the multiple second distances decreasing in sequence.

[0020] Optionally, based on the test data, adjusting one or more of the algorithm recognition threshold and the braking strategy threshold for the vehicle under test to trigger the emergency braking function includes:

[0021] If the first boundary data exceeds the first width threshold, both sides of the algorithm clipping area for algorithm recognition are translated and narrowed towards the central axis of the vehicle's travel. If the first boundary data does not exceed the first width threshold, both sides are translated and widened away from the central axis;

[0022] If the second boundary data exceeds the second width threshold, both sides of the collision warning area for algorithm recognition are translated and narrowed towards the central axis of the vehicle's travel. If the second boundary data does not exceed the second width threshold, both sides are translated and widened away from the central axis;

[0023] If the second boundary data is lower than the third width threshold, both sides of the collision warning area are translated and widened away from the central axis of the vehicle's travel, and the TTC is increased. The third width threshold is the distance at which the vehicle has a risk of scraping a pedestrian. The first width threshold is greater than the second width threshold, and the second width threshold is greater than the third width threshold;

[0024] If the second boundary data all exceeds the fourth width threshold, both sides of the collision warning area are translated and narrowed towards the central axis of the vehicle's travel, and the TTC is reduced. The fourth width threshold is the distance at which the vehicle accidentally triggers the automatic emergency braking system.

[0025] Optionally, the target object is a target vehicle traveling in front of the vehicle to be tested;

[0026] The test parameters include multiple groups. Each group of test parameters includes the first speed of the target vehicle's travel and the second speed of the vehicle to be tested. The second speeds in multiple groups of test parameters with the same first speed increase sequentially. The second speed is greater than the first speed;

[0027] The use cases based on the vehicle evaluation procedure set boundary test cases, including:

[0028] The vehicle to be tested and the target vehicle travel based on the use cases and test parameters of the corresponding vehicle evaluation procedure, and the test ends when any one of the target conditions is met. The target conditions include that the vehicle to be tested triggers the automatic emergency braking system and the speed of the vehicle to be tested is less than the speed of the target vehicle, the vehicle to be tested and the target vehicle come into contact, and the TTC of the vehicle to be tested is less than the target time threshold and the automatic emergency braking system is not triggered.

[0029] Optionally, obtaining the test data of the vehicle to be tested corresponding to the test parameters includes:

[0030] Obtain the test data of the vehicle to be tested corresponding to each group of test parameters. The test data includes the distance between the vehicle to be tested and the target vehicle when the vehicle to be tested triggers the automatic emergency braking system, the TTC for triggering the automatic emergency braking system, and the complete braking curve for triggering the automatic emergency braking system.

[0031] Optionally, the adjusting the algorithm recognition threshold and / or the braking strategy threshold for the vehicle to be tested to trigger the emergency braking function based on the test data includes:

[0032] If the distance between the vehicle to be tested and the target vehicle when the vehicle to be tested triggers the automatic emergency braking system exceeds the first distance threshold, reduce the TTC in the algorithm recognition threshold; if the distance between the vehicle to be tested and the target vehicle when the vehicle to be tested triggers the automatic emergency braking system does not exceed the first distance threshold, increase the TTC in the algorithm recognition threshold; if the vehicle speed change amount of the vehicle to be tested corresponding to each group of test parameters is greater than or equal to the vehicle speed difference between the vehicle to be tested and the target vehicle, reduce the TTC in the algorithm recognition threshold, and if at the end of the test, there is a vehicle speed change amount of the vehicle to be tested less than the vehicle speed difference between the vehicle to be tested and the target vehicle, increase the TTC in the algorithm recognition threshold;

[0033] If the relative speed between the vehicle to be tested and the target vehicle does not exceed the speed threshold, the vehicle to be tested brakes with the first deceleration, and when the TTC continues to decrease, increase the first deceleration to the second deceleration for braking; if the relative speed between the vehicle to be tested and the target vehicle exceeds the speed threshold, brake with the third deceleration, and the third deceleration is greater than or equal to the second deceleration; if the distance between the vehicle to be tested and the target vehicle when the vehicle to be tested triggers the automatic emergency braking system exceeds the first distance threshold and the TTC when triggering the automatic emergency braking system is greater than the preset value, increase the speed threshold, and vice versa, reduce the speed threshold.

[0034] In a second aspect, the present application also provides a boundary performance test device for automatic emergency braking, including:

[0035] A first processing unit for setting a boundary test scenario based on a standard scenario in a vehicle evaluation regulation;

[0036] A second processing unit for setting boundary test cases based on use cases of a vehicle evaluation regulation, where the boundary test cases include test parameters for testing the vehicle to be tested to sense a target object and trigger the automatic emergency braking system;

[0037] A test unit for implementing the boundary test cases corresponding to each test parameter in the boundary test scenario to obtain the test data of the vehicle to be tested corresponding to the test parameters;

[0038] An optimization unit for adjusting the algorithm recognition threshold and / or the braking strategy threshold for the vehicle to be tested to trigger the emergency braking function based on the test data.

[0039] In a third aspect, the present application provides a device, which includes a memory and a processor. The memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes an automatic emergency braking boundary performance testing method according to any one of the foregoing first aspects.

[0040] In a fourth aspect, the present application provides a computer storage medium, in which codes are stored. When the codes are run, the device running the codes implements an automatic emergency braking boundary performance testing method according to any one of the foregoing first aspects.

[0041] The present application provides an automatic emergency braking boundary performance testing method and device. When executing the method, first set a boundary test scenario based on the standard scenarios in the vehicle evaluation procedure, and then set boundary test cases based on the use cases of the vehicle evaluation procedure. The boundary test cases include test parameters for testing a vehicle under test to sense a target object and trigger the automatic emergency braking system. Then, implement the boundary test cases corresponding to each test parameter in the boundary test scenario to obtain test data of the vehicle under test corresponding to the test parameter. Finally, based on the test data, adjust the algorithm recognition threshold and / or braking strategy threshold for the vehicle under test to trigger the emergency braking function, so as to execute the boundary test cases based on the test parameters for the vehicle under test to sense a target object and trigger the automatic emergency braking system, and determine the test data for the vehicle under test to trigger the automatic emergency braking system. In this way, the algorithm recognition threshold and / or braking strategy threshold for the vehicle under test to trigger the emergency braking function are optimized and adjusted according to the test data for the vehicle under test to trigger the automatic emergency braking system. Thus, through multiple test parameters and implementing the boundary test cases corresponding to each test parameter, the test data of the vehicle under test corresponding to the test parameter is obtained. With the change of the test data corresponding to the change of multiple test parameters, the optimization and adjustment directions of the algorithm recognition threshold and braking strategy are better determined. Description of the Drawings

[0042] To more clearly illustrate the technical solutions in the embodiments or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0043] Figure 1 It is a schematic flowchart of an automatic emergency braking boundary performance testing method provided by an embodiment of the present application;

[0044] Figure 2Schematic diagram of boundary test for false triggering of a guardrail obstacle provided by an embodiment of the present application;

[0045] Figure 3 Schematic diagram of boundary performance test for automatic emergency braking of a pedestrian crossing the road provided by an embodiment of the present application;

[0046] Figure 4 Schematic diagram of a radar sensing area provided by an embodiment of the present application;

[0047] Figure 5 Schematic diagram of boundary performance test for automatic emergency braking of a vehicle moving straight ahead and a moving target vehicle in front in a test scenario provided by an embodiment of the present application;

[0048] Figure 6 Schematic diagram of the structure of a boundary performance test device for automatic emergency braking provided by an embodiment of the present application. Detailed implementation manners

[0049] Currently, the test scenarios specified by regulations mainly propose performance requirements for the automatic emergency braking function in specific test scenarios, and provide a direction for performance optimization of the automatic emergency braking function through a test scoring standard, ensuring the performance and safety of the automatic emergency braking system AEBS to a certain extent. However, there are also certain limitations in specific test scenarios and scoring standards. For example, the test scenarios in the test regulations cannot cover all usage scenarios of vehicles with automatic emergency braking functions, and the regulations test can only provide test scores as a result reference, and cannot provide an optimization direction for the automatic emergency braking AEB function algorithm.

[0050] Furthermore, the automatic emergency braking AEB function is mainly realized by the vehicle's sensors (perception), controller (algorithm), braking actuator (actuator), and instrument / car machine / smart cockpit (human-machine interaction). Among them, sensors and algorithms have a direct impact on the performance of the AEB function. When the sensor selection and performance are already determined, the threshold setting of algorithm recognition will directly affect the actual performance of the AEB function. And it is difficult to provide specific guidance for the rationality and optimization direction of the AEB function algorithm recognition threshold and the setting of braking strategies only by obtaining test scores through the test cases specified by regulations.

[0051] Based on the above problems, the present application provides a method and device for testing the boundary performance of automatic emergency braking. When executing the method, first set the boundary test scenario based on the standard scenarios in the vehicle evaluation regulations, and then set the boundary test cases based on the use cases of the vehicle evaluation regulations. The boundary test cases include test parameters for testing the automatic emergency braking system triggered by the target object sensed by the vehicle to be tested. Then, implement the boundary test cases corresponding to each test parameter in the boundary test scenario to obtain the test data of the vehicle to be tested corresponding to the test parameters. Finally, based on the test data, adjust the algorithm recognition threshold and / or braking strategy threshold for triggering the emergency braking function of the vehicle to be tested, so as to execute the boundary test cases based on the test parameters for testing the automatic emergency braking system triggered by the target object sensed by the vehicle to be tested, and determine the test data of the vehicle to be tested triggering the automatic emergency braking system. In this way, according to the test data triggering the automatic emergency braking system of the vehicle to be tested, optimize and adjust the algorithm recognition threshold and / or braking strategy threshold for triggering the emergency braking function of the vehicle to be tested. Thus, through multiple test parameters and implementing the boundary test cases corresponding to each test parameter, obtain the test data of the vehicle to be tested corresponding to the test parameters, and determine the optimization adjustment direction of the algorithm recognition threshold and braking strategy with the change of the corresponding test data as the multiple test parameters change.

[0052] In addition, based on the standard scenarios in the vehicle evaluation regulations, weather conditions or road conditions that interfere with the recognition of the vehicle to be tested can also be configured as the boundary test scenario, so that the test can cover more usage scenarios of vehicles with AEB functions.

[0053] In order to more comprehensively understand the features and technical content of the embodiments of the present disclosure, the implementation of the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for reference and illustration only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the sake of explanation, numerous details are provided to provide a thorough understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and devices may be shown in a simplified manner to simplify the drawings.

[0054] The terms "first", "second", etc. in the description and claims of the embodiments of the present disclosure and the above accompanying drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that such data may be interchanged under appropriate circumstances so as to implement the embodiments of the present disclosure described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion.

[0055] Unless otherwise specified, the term "plurality" means two or more.

[0056] In the embodiment of the present disclosure, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B indicates: A or B.

[0057] The term "and / or" is a description of the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.

[0058] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0059] See also Figure 1 , Figure 1 A flow chart of a method for testing the boundary performance of automatic emergency braking provided in an embodiment of the present application, a method for testing the boundary performance of automatic emergency braking, comprising:

[0060] S101. Setting a boundary test scenario based on a standard scenario in a vehicle evaluation procedure.

[0061] Optionally, the above-mentioned vehicle evaluation program may be C-NCAP (full name China-New Cars Assessment Program) issued by the China Automotive Technology and Research Center, which performs safety performance tests on vehicles, including occupant protection, pedestrian protection, active safety, etc.

[0062] At present, although the vehicle evaluation procedures have ensured the performance and safety of the AEB system to a certain extent, the vehicle evaluation procedures are also constantly updated and improved and there are differences in different countries and regions. In addition, the standard scenarios of the vehicle evaluation procedures cannot cover all usage scenarios of vehicles with AEB functions, such as the possible reduction in recognition performance under bad weather conditions (such as rain, snow, fog, haze, etc.) or special road conditions (such as curves, ramps, tunnels) that interfere with vehicle recognition, and the possible omissions and misjudgments of scene recognition algorithms that require a large amount of image or radar data processing at intersections and multi-lane scenes. Therefore, optionally, the present application can be combined with the standard scenarios in the vehicle evaluation procedures to set boundary test scenarios that affect, such as configuring weather conditions or road conditions that interfere with the recognition of the vehicle to be tested as boundary test scenarios.

[0063] S102. Based on the use cases of the vehicle evaluation procedure, a boundary test case is set, wherein the boundary test case includes test parameters for testing the automatic emergency braking system of the vehicle to be tested when sensing a target object.

[0064] Use cases of vehicle evaluation procedures put forward quantitative requirements for automatic emergency braking (AEB) test scenarios, such as the running speed, direction, and driving distance of the vehicle under test, the traveling speed, direction, and triggering position of the target vehicle or dummy, etc.; Executing the use cases of vehicle evaluation procedures can obtain vehicle scores, but it is difficult to provide an optimization direction for the recognition threshold of the optimization algorithm and reproduce the control process of the automatic emergency braking system to optimize the braking strategy only through the scores.

[0065] Therefore, this application configures multiple test parameters that increase sequentially, and then implements boundary test cases based on the test parameters, so that in the subsequent step S103, the test data of the vehicle under test corresponding to each test parameter can be obtained. With the change of the test data corresponding to the change of multiple test parameters, the optimization adjustment direction of the algorithm recognition threshold and / or braking strategy is determined.

[0066] S103. Implement the boundary test cases corresponding to each test parameter in the boundary test scenario, and obtain the test data of the vehicle under test corresponding to the test parameter.

[0067] The above test data may include one or more of the boundary data for triggering and not triggering the automatic emergency braking system, the boundary data for not triggering and mis-triggering the automatic emergency braking system, the overall braking curve for triggering the automatic emergency braking system, and the time to collision (TTC) for triggering the automatic emergency braking system; Triggering the automatic emergency braking system includes automatic emergency braking for triggering the automatic emergency braking system, forward collision warning for triggering the automatic emergency braking system, and human-machine interaction alarm for triggering the automatic emergency braking system.

[0068] The above TTC (Time To Collision), that is, the time to collision, refers to the time required for the vehicle to collide with the target object in front under the current speed and motion state.

[0069] The above-mentioned Autonomous Emergency Braking System (AEBS) is an advanced driver assistance system that, based on environmental perception sensors (such as millimeter-wave radar or cameras), senses the risk of potential collisions with vehicles, pedestrians, or other traffic participants ahead, and automatically triggers the actuator through the system to apply braking to avoid collisions or reduce the severity of collisions. The operation of the Autonomous Emergency Braking System mainly involves two functions: forward collision warning and autonomous emergency braking. Among them, the Front Collision Warning (FCW) function, based on environmental perception sensors (such as millimeter-wave radar or cameras), senses the risk of potential collisions with vehicles, pedestrians, or other traffic participants ahead, and when there is a collision risk, sends a warning signal to the human-machine interaction module, causing the human-machine interaction module of the vehicle to send a human-machine interaction alarm signal to the driver (such as a beep from the audio or an alarm signal displayed on the display screen, etc.). The Autonomous Emergency Braking (AEB) is based on environmental perception sensors (such as millimeter-wave radar or cameras) to sense the risk of potential collisions with vehicles, pedestrians, or other traffic participants ahead. When the collision risk increases and the driver does not take effective actions to avoid the risk, the vehicle actively decelerates to avoid collisions or reduce the severity of collisions.

[0070] S104. Based on the test data, adjust the algorithm recognition threshold and / or braking strategy threshold for the vehicle under test to trigger the emergency braking function.

[0071] By analyzing the above test data, it is possible to calibrate the perception boundary range of the sensor recognition area of the Autonomous Emergency Braking System, the time points for the Autonomous Emergency Braking System to trigger human-machine interaction alarms, forward collision warnings, and autonomous emergency braking, etc., and provide data references and optimization directions for the braking strategy.

[0072] Determine the optimization adjustment direction of algorithm recognition and braking control for the automatic emergency braking function of the vehicle to be tested through test data. Exemplarily, when the boundary data for triggering and not triggering the automatic emergency braking system is too large, in step S104, the algorithm recognition threshold can be optimized according to actual needs. For example, when identifying the scenario of a pedestrian crossing the road, if the vehicle to be tested triggers the automatic emergency braking warning when it is far away from the pedestrian crossing the road, the algorithm recognition range of the automatic emergency braking system can be narrowed. For example, when identifying the scenario of a vehicle moving ahead, if the distance between the vehicle to be tested and the target vehicle is too large before and after triggering the automatic emergency braking system, the TTC of the target recognition algorithm of the automatic emergency braking system can be reduced to avoid prematurely triggering the automatic emergency braking system and affecting the user experience. Exemplarily, when the boundary data for not triggering and mis-triggering the automatic emergency braking system is too large, in step S104, the algorithm recognition threshold can be optimized according to actual needs. For example, in a scenario with a guardrail, if a pedestrian does not cross the fence to the side of the vehicle to be tested (the fence is 0.5 meters away from the vehicle), the automatic emergency braking system does not need to be triggered. However, during the test of the vehicle to be tested, the automatic emergency braking system was not triggered at 0.9 meters, but was mis-triggered at 0.8 meters. Then, when the vehicle to be tested recognizes an obstacle scenario with a fence, the algorithm recognition range of the automatic emergency braking system can be narrowed.

[0073] Based on the above boundary performance test method for automatic emergency braking, the present application obtains the test data of the vehicle to be tested corresponding to each test parameter by passing multiple test parameters and implementing the boundary test cases corresponding to each test parameter. In this way, the relevant test data for triggering the automatic emergency braking system can be determined according to multiple test parameters, and then the optimization adjustment direction of the algorithm recognition threshold and braking strategy can be determined based on the test data.

[0074] Based on the above embodiments, the present application can set corresponding test cases based on different boundary test scenarios to optimize and adjust the algorithm recognition threshold or braking strategy. The following will be introduced separately. It should be noted that the implementation methods given in the following introduction are only exemplary explanations and do not represent all implementation methods of the embodiments of the present application.

[0075] See Figure 3 The schematic diagram of the boundary performance test of automatic emergency braking for a pedestrian crossing the road shown in Figure 1 In the embodiment shown in

[0076] First, use the pedestrian - adult target distal crossing 50% (CPFA50%) collision scenario specified by the C-NCAP2024 regulation as the boundary test scenario.

[0077] Optionally, it is also possible to use the above CPFA 50% collision scenario and add conditions of bad weather (such as rain, snow, fog, haze, etc.) as the boundary test scenario, or use special road conditions such as curved roads or ramp conditions as the boundary test scenario.

[0078] Secondly, the vehicle under test and the pedestrian move forward according to the test cases of the CPFA 50% collision scenario specified in the C-NCAP 2024 regulations, and the pedestrian stops when moving to the target distance from the vehicle's driving central axis. It is detected whether the vehicle under test triggers the automatic emergency braking system, and the target distance parameter is one of multiple distance values.

[0079] The test cases of the CPFA 50% collision scenario specified in the above C-NCAP 2024 regulations put forward quantitative requirements for the test scenario of the automatic emergency braking system, such as the running speed, direction, driving distance of the vehicle under test, and the running speed, direction, triggering position of the target vehicle or pedestrian.

[0080] Exemplarily, this boundary test case may specifically include that the target pedestrian starts to accelerate from 6 m away from the vehicle center, accelerates to 8 km / h after 1.5 m, and stops at a distance of Y meters from the vehicle's driving central axis, and tests the critical lateral distance at which the automatic emergency braking system of the vehicle under test is triggered in the pedestrian crossing scenario.

[0081] Optionally, execute the test cases specified in the C-NCAP 2024 regulations in the pedestrian - adult target far - end crossing 50% - 100% (CPFA 50% - 100%) collision scenario specified in the C-NCAP 2024 regulations to obtain a test score, so as to evaluate and obtain the algorithm recognition threshold and braking strategy that can meet the C-NCAP 2024 regulations and ensure the performance and safety of the AEB system.

[0082] Then, based on the order of a plurality of decreasing first distances, starting from the first first distance value, execute the above - mentioned boundary test case once for each first distance, and detect whether the vehicle under test triggers the automatic emergency braking system corresponding to this distance value until the forward collision warning in the automatic emergency braking system is not triggered at the first target distance, and the forward collision warning in the automatic emergency braking system is triggered at the second target distance adjacent to the first target distance. Both the first target distance and the second target distance are used as the first boundary data, and the first target distance and the second target distance are two adjacent first distances among the plurality of decreasing first distances.

[0083] Furthermore, based on the order of multiple second distances that decrease successively, starting from the first second distance value, the above-mentioned boundary test cases are executed once for each second distance to detect whether the vehicle under test triggers the automatic emergency braking system corresponding to the second distance, until it is obtained that the vehicle under test does not trigger the automatic emergency braking in the automatic emergency braking system corresponding to the third target distance, and the fourth target distance adjacent to the third target distance triggers the automatic emergency braking in the automatic emergency braking system. Both the third target distance and the fourth target distance are used as the second boundary data, and the third target distance and the fourth target distance are two adjacent second distances among the multiple second distances that decrease successively.

[0084] The above-mentioned boundary data for testing forward collision warning or automatic emergency braking, the multiple distance values include multiple first distances that decrease successively and multiple second distances that decrease successively. Among them, the first distance is set to test the distance of the forward collision warning when the vehicle under test triggers the automatic emergency braking system, and the second distance is set to test the distance of the automatic emergency braking when the vehicle under test triggers the automatic emergency braking system. Thus, referring to Figure 3 , based on the pedestrian - adult target distal crossing 50% (CPFA50%) collision scenario specified in the C-NCAP2024 regulation, the present application sets the intrusion degrees of the pedestrian crossing the road differently to detect the reaction of the automatic emergency braking system of the vehicle under test. Through the multiple first distances set, the range value (the first target distance, the second target distance) of the critical lateral distance Y1 for releasing the brakes after point braking (triggering the forward collision warning) and not triggering the forward collision warning of the automatic emergency braking system is determined; through setting multiple second distances, the range value (the third target distance, the fourth target distance) of the critical lateral distance Y2 for stopping the vehicle (triggering the automatic emergency braking of the automatic emergency braking system) / releasing the brakes after point braking (triggering the forward collision warning) is determined. For example, Figure 3 , the automatic emergency braking system is not triggered in the green area, in the yellow area, the forward collision warning of the automatic emergency braking system is triggered for point braking but not stopping, and the automatic emergency braking of the automatic emergency braking system is triggered to stop in the red area.

[0085] Finally, based on the above first boundary data (the first target distance and the second target distance) and second boundary data (the third target distance and the fourth target distance), the algorithm recognition threshold for the vehicle under test to trigger the emergency braking function can be adjusted.

[0086] Referring to Figure 4 As shown in a schematic diagram of a radar sensing area, based on the first boundary data and the second boundary data, adjusting the algorithm recognition threshold for the vehicle under test to trigger the emergency braking function may include the following adjustments:

[0087] Adjustment 1: If the first boundary data exceeds the first width threshold, both sides of the algorithm clipping area for algorithm recognition are translated and narrowed towards the central axis of the vehicle's travel. If the first boundary data does not exceed the first width threshold, both sides are translated and widened away from the central axis.

[0088] From a plurality of decreasing first distances, by detecting the first target distance at which the vehicle under test does not trigger a forward collision warning and the second target distance at which it triggers a forward collision warning in the CPFA50% collision scenario, it can be determined that the critical lateral distance at which the vehicle under test actually triggers a forward collision warning is between the first target distance and the second target distance. Therefore, when the first boundary data (the first target distance and the second target distance) exceeds the first width threshold, that is, when the critical lateral distance at which the forward collision warning is actually triggered is too large, both sides of the algorithm clipping area recognized by the algorithm can be translated and narrowed inward ( Figure 4 AB and BD are translated inward in the figure), and vice versa, they are translated outward ( Figure 4 AB and BD are translated outward in the figure) and widened.

[0089] Adjustment 2: If the second boundary data exceeds the second width threshold, both sides of the collision warning area for algorithm recognition are translated and narrowed towards the central axis of the vehicle's travel. If the second boundary data does not exceed the second width threshold, both sides are translated and widened away from the central axis.

[0090] Based on the above first target distance and second target distance, similarly, by determining the third target distance and the fourth target distance from a plurality of decreasing second distances as the second boundary data, when the critical lateral distance at which the automatic emergency braking is actually triggered is too large, both sides of the collision warning area recognized by the algorithm can be translated and narrowed towards the central axis of the vehicle's travel ( Figure 4 EH and GF are translated inward in the figure). Conversely, they are translated outward ( Figure 4 EH and GF are translated outward in the figure) and widened.

[0091] Adjustment 3: If the second boundary data is lower than the third width threshold, both sides of the collision warning area are translated and widened away from the central axis of the vehicle's travel, and the TTC is increased. The third width threshold is the distance at which there is a risk of the vehicle scraping a pedestrian. The first width threshold is greater than the second width threshold, and the second width threshold is greater than the third width threshold.

[0092] When the second boundary data is too small and too close to the vehicle, resulting in a risk of scraping the dummy target, both sides of the collision warning area are translated and widened away from the central axis of the vehicle's travel ( Figure 4Among them, when EH and GF are translated outward), while slightly increasing the TTC of the algorithm, the automatic emergency braking system can be triggered in advance to perform automatic emergency braking during the physical examination, so as to reduce the risk of rubbing pedestrians.

[0093] Adjustment Four: If all the second boundary data exceed the fourth width threshold, both sides of the collision warning area are translated and narrowed towards the central axis of the vehicle's travel, and the TTC is reduced. The fourth width threshold is the distance at which the vehicle accidentally triggers the automatic emergency braking system.

[0094] When it is easy for the vehicle to frequently and accidentally trigger the automatic emergency braking system, both sides of the collision warning area can be appropriately translated and narrowed towards the central axis of the vehicle's travel ( Figure 4 Among them, EH and GF are translated inward), while slightly increasing the TTC of the algorithm.

[0095] Exemplarily, refer to Figure 2 As shown in the schematic diagram of the boundary test for accidental triggering of a guardrail obstacle, in the current scenario where there is an additional obstacle (guardrail), pedestrians do not need to trigger the automatic emergency braking system when they do not cross the guardrail to the side of the vehicle to be tested (the fence is 0.5 meters away from the vehicle). However, when the vehicle to be tested is tested, it does not trigger the automatic emergency braking system at 0.9 meters, but accidentally triggers the automatic emergency braking system at 0.8 meters. Then, when the vehicle to be tested recognizes the obstacle scenario with a fence, the recognition range of the algorithm of the automatic emergency braking system can be narrowed.

[0096] In this way, the first boundary data and the second boundary data obtained from the quantitative test can provide data references for optimizing the algorithm clipping area and the collision warning area range of the radar recognition area in the recognition algorithm of the automatic emergency braking AEB function. In addition, test obstacles such as guardrails can be added based on the current scenario, and the test critical value for the accidentally triggered scenario of AEB in the boundary test can be analyzed, which can provide data references for setting the anti-accidental trigger threshold of the automatic emergency trigger system. Therefore, by analyzing the test data of the boundary test, quantitative and referenceable suggestions can be provided for setting the threshold of the algorithm recognition, improving the performance and safety of the AEB function.

[0097] It should be noted that all pedestrians mentioned in this application for testing are dummy models for experimental testing.

[0098] It should be noted that the above radar sensing area can be the area effectively sensed by millimeter-wave radar / camera, and the radar sensing ability can identify all moving obstacles within the area, such as Figure 4 the fan-shaped area in Figure 4An area of rectangle ABCD. For obstacles outside rectangle ABCD (such as the pedestrian crossing at point a in the figure), the algorithm does not consider it as a valid / obstacle that needs attention; for obstacles inside rectangle ABCD (such as the pedestrian crossing at point b in the figure), the algorithm marks its distance and relative speed, and calculates the TTC of target b. The above collision warning area can be, for example, Figure 4 An area of rectangle EFGH. For targets within the collision warning area (such as the pedestrian crossing at point c in the figure), the algorithm marks its distance and relative speed, calculates the TTC of target c. If there are multiple targets within the area (such as c1, c2, c3), the algorithm of the automatic emergency braking system compares and determines the target with the minimum TTC. When its TTC < the set TTC value for forward collision warning or automatic emergency braking, the system performs the corresponding forward collision warning or automatic emergency braking.

[0099] See Figure 5 The schematic diagram of the boundary performance test of automatic emergency braking for a vehicle going straight and a moving target vehicle in front as shown in Figure 1 Based on the embodiment shown in

[0100] First, a boundary test scenario based on the test scenario of the vehicle going straight and a moving target vehicle in front (CCRm) of the vehicle automatic emergency braking system.

[0101] The vehicle automatic emergency braking system (Autonomous Emergency Braking Car to Car, abbreviated as AEBC2C) is the automatic emergency braking for vehicle targets in the regulatory test project. It automatically performs emergency braking by detecting the dynamics of the vehicle in front in real time. It aims to avoid or reduce traffic accidents that may occur when the driver fails to notice the vehicle in front or on the front side.

[0102] Optionally, it can also be the test scenario of the vehicle going straight and a moving target vehicle in front (CCRm) of the vehicle automatic emergency braking system with the addition of conditions of bad weather (such as rain, snow, fog, haze, etc.) as the boundary test scenario, or under special road conditions such as curved roads or slopes as the boundary test scenario.

[0103] Second, the test vehicle and the target vehicle travel based on the use cases and test parameters of the corresponding vehicle evaluation procedure, and the test ends when any one of the target conditions is met; the target conditions include that the test vehicle triggers the automatic emergency braking system and the speed V of the test vehicle VUT is less than the speed V of the target vehicle GVTWhen the vehicle under test and the target vehicle come into contact and the time to collision (TTC) of the vehicle under test is less than the target time threshold (e.g., 1.5 seconds), the automatic emergency braking system is not triggered.

[0104] The above test parameters can include multiple groups. Each group of test parameters includes the first speed at which the target vehicle travels and the second speed at which the vehicle under test travels. The second speed in multiple groups of test parameters with the same first speed increases sequentially, and the second speed is greater than the first speed. Exemplarily, the first speed of the target vehicle GVT can be 20 kilometers per hour, and the corresponding second speeds of the vehicle under test VUT can be 30, 35, 40, 45, 50, 55, 60 kilometers per hour in sequence. It can be understood that when the first speed is 20 kilometers per hour and the corresponding second speed is 30 kilometers per hour, it can be used as a group of test parameters; when the first speed is 20 kilometers per hour and the corresponding second speed is 35 kilometers per hour, it can be used as another group of test parameters, and so on. Of course, multiple first speeds can also be set. The first speed of the target vehicle GVT can also be 30 kilometers per hour, and the corresponding second speeds of the vehicle under test VUT can be 40, 45, 50, 55, 60, 65, 70 kilometers per hour in sequence, and the second speed is greater than the first speed.

[0105] It can be understood that when the TTC of the above vehicle under test is less than the target time threshold (e.g., 1.5 seconds), the automatic emergency braking system is not triggered, which means that if the VUT does not issue an alarm or take braking or steering or other means to abort the test when the TTC < 1.5s, it is judged that the AEB is not triggered / works normally, and manual intervention takes over, and manual braking or steering is used to prevent the VUT and GVT from colliding at a high relative speed.

[0106] The above target vehicle (Global Vehicle Target, GVT) is the vehicle target specified in the vehicle evaluation procedure. The vehicle under test is an experimental vehicle (Vehicle Under Test, VUT) equipped with relevant driving assistance function systems and will be tested according to the corresponding vehicle evaluation procedure.

[0107] Then, in the boundary test scenario, the boundary test cases are implemented corresponding to each test parameter, and the test data of the vehicle under test corresponding to the test parameter are obtained. The specific steps can be:

[0108] Step A1: Drive the target vehicle at the first speed (e.g., 20 kilometers per hour) in the above boundary test scenario, and then drive the vehicle under test at the second speed (e.g., 30 kilometers per hour). The test ends when any one of the target conditions is met, and the test data of a group of test parameters corresponding to the first speed and the second speed are obtained.

[0109] Step A2: Increase the second speed by 5 km / h to obtain a new second speed (e.g., 35 km / h), and continue to execute the above Step A1 until the second speed reaches the target speed (e.g., 60 km / h).

[0110] Of course, the above first speed can be other values such as 30 km / h, and then repeat the above Steps A1 - A2 to obtain the test data corresponding to multiple groups of test parameters.

[0111] The above test data may include the distance between the vehicle under test and the target vehicle when the automatic emergency braking system is triggered, the TTC when the automatic emergency braking system is triggered, and the complete braking curve when the automatic emergency braking system is triggered.

[0112] The distance between the vehicle under test and the target vehicle when the automatic emergency braking system of the above vehicle under test is triggered may refer to the distance between the vehicle under test and the target vehicle at the moment when the automatic emergency braking system sends an alarm signal to the driver through the vehicle's human - machine interaction module. Set multiple groups of test parameters, set different second speeds corresponding to the same first speed to obtain the critical longitudinal distance at which the vehicle under test triggers the automatic emergency braking system. Of course, the corresponding human - machine interaction TTC, forward collision warning TTC, and automatic emergency braking TTC during the process of triggering the automatic emergency braking system can also be further obtained. The braking curve during the entire process of triggering the automatic positive braking system can also be obtained. The deceleration - time curve can be obtained through inertial measurement. The speed - time curve can also be obtained through the GPS of the vehicle under test to obtain the vehicle speed change situation until the end of the test.

[0113] Finally, based on the above test data, adjust the algorithm recognition threshold and / or braking strategy threshold for the vehicle under test to trigger the emergency braking function. The specific adjustment may include:

[0114] Algorithm adjustment: If the distance between the vehicle under test and the target vehicle when the automatic emergency braking system is triggered exceeds the first distance threshold, then reduce the TTC in the algorithm recognition threshold. If the distance between the vehicle under test and the target vehicle when the automatic emergency braking system is triggered does not exceed the first distance threshold, then increase the TTC in the algorithm recognition threshold.

[0115] It can be understood that if the distance to trigger the automatic emergency braking system is too long, slightly reduce the vehicle - to - vehicle TTC threshold in the algorithm recognition threshold to improve the user experience, and vice versa.

[0116] If the vehicle speed change amount of the vehicle under test corresponding to each group of test parameters is greater than or equal to the vehicle speed difference between the vehicle under test and the target vehicle, then reduce the TTC in the algorithm recognition threshold. If, when the test ends, there is a vehicle speed change amount of the vehicle under test that is less than the vehicle speed difference between the vehicle under test and the target vehicle, then increase the TTC in the algorithm recognition threshold.

[0117] It is understandable that when the speed change amounts of the vehicles under test corresponding to each group of test parameters can all reach the speed difference between the VUT and the GVT (without collision occurring finally), the vehicle-to-vehicle TTC threshold is slightly decreased. If, at the end of the test, the speed change amounts of the vehicles under test corresponding to each group of test parameters are less than the speed difference between the two vehicles (collision occurs), the vehicle-to-vehicle TTC threshold is slightly increased so that the vehicle under test can trigger the alarm of the automatic emergency braking system earlier, improving safety.

[0118] Adjustment of the braking strategy: If the relative speed between the vehicle under test and the target vehicle does not exceed the speed threshold, the vehicle under test brakes with the first deceleration. When the TTC continues to decrease, the first deceleration is increased to the second deceleration for braking; if the relative speed between the vehicle under test and the target vehicle exceeds the speed threshold, the third deceleration is used for braking, and the third deceleration is greater than or equal to the second deceleration; if the distance between the vehicle under test and the target vehicle exceeds the first distance threshold when the vehicle under test triggers the automatic emergency braking system, and the TTC when triggering the automatic emergency braking system is greater than the preset value, the speed threshold is increased. On the contrary, if the distance between the vehicle under test and the target vehicle does not exceed the first distance threshold when the vehicle under test triggers the automatic emergency braking system, and the TTC when triggering the automatic emergency braking system is not greater than the preset value, the speed threshold is increased, then the speed threshold is decreased.

[0119] In the adjustment of the braking strategy, different braking strategies are adopted for different vehicle speeds and different small relative speeds. When the relative speed between the vehicle under test and the target vehicle is small (not exceeding the speed threshold), pre-pressure can be built. First, brake with a smaller first deceleration. When the TTC further decreases, a larger second deceleration is adopted for segmented braking strategy to avoid suddenly adopting a larger deceleration and improve the user experience. When the relative speed between the vehicle under test and the target vehicle is large (exceeding the speed threshold), an emergency braking strategy of using the maximum deceleration (the third deceleration) for emergency braking is adopted to avoid the risk of collision due to a large relative speed and ensure safety. In addition, when the distance between the vehicle under test and the target vehicle is too long when the vehicle under test triggers the automatic emergency braking system, and the TTC when triggering the automatic emergency braking system is greater than the preset value, the speed threshold can be increased, and vice versa.

[0120] In this way, by optimizing the overall braking curve of the automatic emergency braking system, better braking performance of the brake actuator under extreme braking conditions is determined, providing a data reference for the braking decision-making of the automatic emergency braking AEB function; and by analyzing the collision time TTC under the critical triggering conditions in the AEB boundary test and comparing the time points when the FCW is triggered during the conventional test, the time points of the forward collision warning and the automatic emergency braking execution are optimized, providing a data reference for the setting of the collision warning time.

[0121] The above are some specific implementation manners of a method for testing the boundary performance of automatic emergency braking provided by the embodiments of the present application. Based on this, the present application also provides a corresponding device. The device provided by the embodiments of the present application will be introduced from the perspective of functional modularization below.

[0122] See Figure 6 the structural schematic diagram of a device for testing the boundary performance of automatic emergency braking shown in, a device for testing the boundary performance of automatic emergency braking includes:

[0123] A first processing unit 601, configured to set a boundary test scenario based on a standard scenario in a vehicle evaluation regulation;

[0124] A second processing unit 602, configured to set boundary test cases based on use cases of a vehicle evaluation regulation, where the boundary test cases include test parameters for testing a target object sensed by a vehicle under test to trigger an automatic emergency braking system;

[0125] A test unit 603, configured to execute the boundary test cases corresponding to each test parameter in the boundary test scenario, and obtain test data of the vehicle under test corresponding to the test parameter;

[0126] An optimization unit 604, configured to adjust an algorithm recognition threshold and / or a braking strategy threshold for the vehicle under test to trigger an emergency braking function based on the test data.

[0127] For the above device, the first processing unit 601 configures a boundary test scenario, the second processing unit 602 configures boundary test cases including multiple test parameters, and the test unit 603 executes the boundary test cases based on each test parameter to determine test data of the vehicle under test for triggering an automatic emergency braking system. Furthermore, the optimization unit 604 optimally adjusts the algorithm recognition threshold and / or the braking strategy threshold for the vehicle under test to trigger an emergency braking function according to the test data for the vehicle under test to trigger an automatic emergency braking system. In this way, by using multiple test parameters and executing the boundary test cases corresponding to each test parameter, the test data of the vehicle under test corresponding to the test parameter is obtained, and with the change of the test data corresponding to the change of multiple test parameters, the optimization adjustment direction of the algorithm recognition threshold and the braking strategy can be better determined.

[0128] In a possible implementation manner, the first processing unit is specifically configured to configure weather conditions or road conditions that interfere with the recognition of the vehicle under test based on a standard scenario in a vehicle evaluation regulation as a boundary test scenario.

[0129] In a possible implementation, the test data includes one or more of the boundary data for triggering and not triggering the automatic emergency braking system, the boundary data for not triggering and false triggering the automatic emergency braking system, the overall braking curve of the triggered automatic emergency braking system, and the TTC of the triggered automatic emergency braking system; the triggered automatic emergency braking system includes the automatic emergency braking of the triggered automatic emergency braking system, the collision warning of the triggered automatic emergency braking system, and the human-machine interaction alarm of the triggered automatic emergency braking system.

[0130] In a possible implementation, the target is a pedestrian crossing the road, and the test parameters are multiple distance values; the second processing unit is specifically configured to make the vehicle under test and the pedestrian travel according to the corresponding use case regulations in the vehicle evaluation procedure, and the pedestrian stops when moving to a target distance from the vehicle driving central axis, and detect whether the vehicle under test triggers the automatic emergency braking system, and the target distance parameter is one of the multiple distance values.

[0131] Optionally, the multiple distance values include multiple first distances decreasing in sequence and multiple second distances decreasing in sequence; the test unit is specifically configured to obtain the forward collision warning in the automatic emergency braking system not triggered by the vehicle under test corresponding to the first target distance, and the forward collision warning in the automatic emergency braking system triggered by the vehicle under test corresponding to the second target distance adjacent to the first target distance; use both the first target distance and the second target distance as the first boundary data, where the first target distance and the second target distance are two adjacent first distances among the multiple first distances decreasing in sequence; obtain that the vehicle under test does not trigger the automatic emergency braking in the automatic emergency braking system corresponding to the third target distance, and the vehicle under test triggers the automatic emergency braking in the automatic emergency braking system corresponding to the fourth target distance adjacent to the third target distance; use both the third target distance and the fourth target distance as the second boundary data, where the third target distance and the fourth target distance are two adjacent second distances among the multiple second distances decreasing in sequence.

[0132] Optionally, the optimization unit is specifically configured to, if the first boundary data exceeds the first width threshold, translate and narrow both sides of the algorithm clipping area for algorithm recognition towards the central axis of the vehicle's travel; if the first boundary data does not exceed the first width threshold, translate and widen both sides away from the central axis; if the second boundary data exceeds the second width threshold, translate and narrow both sides of the collision warning area for algorithm recognition towards the central axis of the vehicle's travel; if the second boundary data does not exceed the second width threshold, translate and widen both sides away from the central axis; if the second boundary data is lower than the third width threshold, translate and widen both sides of the collision warning area away from the central axis of the vehicle's travel and increase the TTC, where the third width threshold is the distance at which the vehicle has a risk of scraping a pedestrian, the first width threshold is greater than the second width threshold, and the second width threshold is greater than the third width threshold; if the second boundary data exceeds the fourth width threshold, translate and narrow both sides of the collision warning area towards the central axis of the vehicle's travel and decrease the TTC, where the fourth width threshold is the distance at which the vehicle erroneously triggers the automatic emergency braking system.

[0133] In a possible implementation manner, the target object is a target vehicle traveling in front of the vehicle to be tested; the test parameters include multiple groups, and each group of test parameters includes a first speed at which the target vehicle travels and a second speed at which the vehicle to be tested travels. The second speed in multiple groups with the same first speed increases sequentially, and the second speed is greater than the first speed; the second processing unit is specifically configured to make the vehicle to be tested and the target vehicle travel based on the use cases and test parameters of the corresponding vehicle evaluation procedure, and end the test when any one of the target conditions is met; the target conditions include that the vehicle to be tested triggers the automatic emergency braking system and the speed of the vehicle to be tested is less than the speed of the target vehicle, the vehicle to be tested and the target vehicle come into contact, and the vehicle to be tested does not trigger the automatic emergency braking system when the TTC of the vehicle to be tested is less than the target time threshold.

[0134] Optionally, the test unit is specifically configured to obtain the test data of the vehicle to be tested corresponding to each group of test parameters, where the test data includes the distance between the vehicle to be tested and the target vehicle when the vehicle to be tested triggers the automatic emergency braking system, the TTC when the automatic emergency braking system is triggered, and the complete braking curve when the automatic emergency braking system is triggered.

[0135] Optionally, the optimization unit is specifically configured to reduce the TTC in the algorithm recognition threshold if the distance between the vehicle under test and the target vehicle exceeds the first distance threshold when the vehicle under test triggers the automatic emergency braking system; increase the TTC in the algorithm recognition threshold if the distance between the vehicle under test and the target vehicle does not exceed the first distance threshold when the vehicle under test triggers the automatic emergency braking system; reduce the TTC in the algorithm recognition threshold if the vehicle speed change of the vehicle under test corresponding to each group of test parameters is greater than or equal to the vehicle speed difference between the vehicle under test and the target vehicle, and increase the TTC in the algorithm recognition threshold if there is a vehicle speed change of the vehicle under test less than the vehicle speed difference between the vehicle under test and the target vehicle when the test ends; if the relative speed between the vehicle under test and the target vehicle does not exceed the speed threshold, the vehicle under test brakes with the first deceleration, and when the TTC continues to decrease, increase the first deceleration to the second deceleration for braking; if the relative speed between the vehicle under test and the target vehicle exceeds the speed threshold, brake with the third deceleration, and the third deceleration is greater than or equal to the second deceleration; increase the speed threshold if the distance between the vehicle under test and the target vehicle exceeds the first distance threshold when the vehicle under test triggers the automatic emergency braking system and the TTC when triggering the automatic emergency braking system is greater than the preset value, and vice versa, reduce the speed threshold.

[0136] The embodiment of the present application also provides corresponding devices and computer storage media for implementing the solutions provided by the embodiments of the present application.

[0137] Among them, the device includes a memory and a processor. The memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes an automatic emergency braking boundary performance test method according to any embodiment of the present application.

[0138] The computer storage medium stores codes, and when the codes are run, the device running the codes implements an automatic emergency braking boundary performance test method according to any embodiment of the present application.

[0139] In the embodiments of the present application, the "first", "second" (if any) in the names such as "first" and "second" are only used as name identifiers and do not represent the first and second in order.

[0140] As can be seen from the description of the above embodiments, those skilled in the art can clearly understand that all or part of the steps in the above method embodiments can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as read-only memory (ROM) / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or some parts of the embodiments of the present application.

[0141] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. One can select some or all of the modules according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0142] The above is only an exemplary embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. A method for testing the boundary performance of automatic emergency braking, characterized in that: include: Set boundary test scenarios based on standard scenarios in vehicle evaluation procedures; Based on the use cases of the vehicle evaluation regulations, boundary test cases are set, wherein the boundary test cases include test parameters for testing the automatic emergency braking system triggered by the target object sensed by the vehicle to be tested; Implementing the boundary test case corresponding to each test parameter in the boundary test scenario, and obtaining test data of the vehicle to be tested corresponding to the test parameter; Based on the test data, an algorithm recognition threshold and / or a braking strategy threshold for triggering an emergency braking function of the vehicle to be tested are adjusted.

2. The method according to claim 1, characterized in that Setting boundary test scenarios based on standard scenarios in vehicle evaluation procedures also includes: Based on the standard scenarios in the vehicle evaluation procedures, configure weather conditions or road conditions that interfere with the identification of the vehicle to be tested as boundary test scenarios.

3. The method according to claim 1, characterized in that The test data includes boundary data of triggering and not triggering the automatic emergency braking system, boundary data of not triggering and falsely triggering the automatic emergency braking system, one or more of the overall braking curve of triggering the automatic emergency braking system and TTC of triggering the automatic emergency braking system; the triggering of the automatic emergency braking system includes triggering automatic emergency braking of the automatic emergency braking system, triggering collision warning of the automatic emergency braking system and triggering human-computer interaction alarm of the automatic emergency braking system.

4. The method according to claim 1, characterized in that: The target object is a pedestrian crossing the road, and the test parameter is a plurality of distance values; The use cases based on the vehicle evaluation procedures set boundary test cases, including: The vehicle and pedestrian to be tested move in accordance with the corresponding use case regulations in the vehicle evaluation procedure, and the pedestrian stops at a target distance from the vehicle's centerline, to detect whether the vehicle to be tested triggers the automatic emergency braking system, where the target distance parameter is one of multiple distance values.

5. The method according to claim 4, characterized in that The plurality of distance values ​​include a plurality of first distances that decrease in sequence and a plurality of second distances that decrease in sequence; The step of obtaining the test data of the vehicle to be tested corresponding to the test parameters includes: Acquire that a first target distance corresponding to the vehicle under test does not trigger a forward collision warning in the automatic emergency braking system, and a second target distance adjacent to the first target distance triggers a forward collision warning in the automatic emergency braking system; The first target distance and the second target distance are both used as first boundary data, wherein the first target distance and the second target distance are two adjacent first distances among the plurality of first distances decreasing in sequence; Obtaining a third target distance corresponding to the vehicle to be tested that does not trigger the automatic emergency braking in the automatic emergency braking system, and a fourth target distance adjacent to the third target distance that triggers the automatic emergency braking in the automatic emergency braking system; using the third target distance and the fourth target distance as second boundary data, the third target distance and the fourth target distance being two adjacent second distances among the multiple second distances that decrease in sequence.

6. The method according to claim 5, characterized in that The adjusting, based on the test data, one or more of the algorithm recognition threshold and the braking strategy threshold of the emergency braking function triggered by the vehicle under test comprises: If the first boundary data exceeds the first width threshold, both sides of the algorithm clipping area used for algorithm identification are translated and narrowed toward the central axis of the vehicle; if the first boundary data does not exceed the first width threshold, both sides are translated and widened away from the central axis; If the second boundary data exceeds the second width threshold, both sides of the collision warning area used for algorithm identification are translated and narrowed toward the central axis of the vehicle; if the second boundary data does not exceed the second width threshold, both sides are translated and widened away from the central axis; If the second boundary data is lower than the third width threshold, both sides of the collision warning area are translated and widened in the direction away from the central axis of the vehicle, and the TTC is increased. The third width threshold is the distance at which the vehicle has a risk of scratching a pedestrian. The first width threshold is greater than the second width threshold, and the second width threshold is greater than the third width threshold. If the second boundary data exceeds the fourth width threshold, both side edges of the collision warning area are translated and narrowed toward the central axis of the vehicle, and the TTC is reduced. The fourth width threshold is the distance at which the vehicle mistakenly triggers the automatic emergency braking system.

7. The method according to claim 1, characterized in that The target object is a target vehicle traveling in front of the vehicle to be tested; The test parameters include multiple groups, each group of test parameters includes a first speed of the target vehicle and a second speed of the vehicle to be tested, and the second speeds of the multiple groups of test parameters using the same first speed increase in sequence, and the second speed is greater than the first speed; The use cases based on the vehicle evaluation procedures set boundary test cases, including: The vehicle under test and the target vehicle move based on the use cases and test parameters of the corresponding vehicle evaluation procedures, and the test ends when any target condition is met; the target conditions include that the vehicle under test triggers the automatic emergency braking system and the speed of the vehicle under test is less than the speed of the target vehicle, the vehicle under test and the target vehicle come into contact, and the automatic emergency braking system is not triggered when the TTC of the vehicle under test is less than the target time threshold.

8. The method according to claim 7, characterized in that The step of obtaining the test data of the vehicle to be tested corresponding to the test parameters includes: The test data of the vehicle to be tested corresponding to each set of test parameters is obtained, wherein the test data includes the distance between the vehicle to be tested and the target vehicle when the automatic emergency braking system is triggered, the TTC of triggering the automatic emergency braking system, and the complete braking curve of triggering the automatic emergency braking system.

9. The method according to claim 8, characterized in that The adjusting, based on the test data, the algorithm recognition threshold and / or the braking strategy threshold of the emergency braking function triggered by the vehicle to be tested comprises: If the distance between the vehicle to be tested and the target vehicle exceeds the first distance threshold when the automatic emergency braking system is triggered, the TTC in the algorithm recognition threshold is reduced; if the distance between the vehicle to be tested and the target vehicle does not exceed the first distance threshold when the automatic emergency braking system is triggered, the TTC in the algorithm recognition threshold is increased; if the speed change of the vehicle to be tested corresponding to each group of test parameters is greater than or equal to the speed difference between the vehicle to be tested and the target vehicle, the TTC in the algorithm recognition threshold is reduced; if the test is completed and there is a speed change of the vehicle to be tested that is less than the speed difference between the vehicle to be tested and the target vehicle, the TTC in the algorithm recognition threshold is increased; If the relative speed between the vehicle under test and the target vehicle does not exceed the speed threshold, the vehicle under test adopts the first deceleration for braking. When the TTC continues to decrease, the first deceleration is increased to the second deceleration for braking. If the relative speed between the vehicle under test and the target vehicle exceeds the speed threshold, the third deceleration is adopted for braking, and the third deceleration is greater than or equal to the second deceleration. If the distance between the vehicle under test and the target vehicle exceeds the first distance threshold when the automatic emergency braking system is triggered and the TTC when the automatic emergency braking system is triggered is greater than the preset value, the speed threshold is increased; otherwise, the speed threshold is reduced.

10. A boundary performance test device for automatic emergency braking, characterized in that: include: A first processing unit, configured to set a boundary test scenario based on a standard scenario in a vehicle evaluation procedure; A second processing unit is used to set a boundary test case based on a case of a vehicle evaluation procedure, wherein the boundary test case includes a test parameter for testing that a target object is sensed by the vehicle to be tested and triggers an automatic emergency braking system; A testing unit, configured to implement the boundary test case in the boundary test scenario corresponding to each test parameter, and obtain test data of the vehicle to be tested corresponding to the test parameter; The optimization unit is used to adjust the algorithm recognition threshold and / or braking strategy threshold of the emergency braking function triggered by the vehicle to be tested based on the test data.

Images