Performance test evaluation method and system for automatic emergency braking system

Abstract

The invention belongs to the technical field of vehicle performance testing, and discloses a performance testing and evaluating method and system for an automatic emergency braking system. All accident description information corresponding to participant types is determined in a preset traffic accident database; all accident description information is divided into scene information sets of various accident scenes to fully consider all types of accident scenes, so that a performance test result has universality and reliability; secondly, a corresponding scene correction coefficient can be calculated according to the scene information set corresponding to each accident scene, and a performance test result of the test vehicle can be obtained according to the scene correction coefficient corresponding to each accident scene and the scoring result, so that the relevance between the performance test result and the traffic accident can be enhanced, and the reliability of the test vehicle can be improved. And thus, the performance in the real traffic accident can be effectively reflected.

Description

Technical Field

[0001] The present application belongs to the technical field of vehicle performance testing, and in particular, relates to a performance testing and evaluation method and system for an automatic emergency braking system. Background Art

[0002] At present, the active and passive safety technologies of vehicles are being used more and more widely, but the actual number of traffic accidents has remained high, especially the number of accidents involving vulnerable road users (Vulnerable Road Users, VRU) has shown an upward trend. Therefore, safety research on vulnerable road users is particularly important.

[0003] As an important active and passive safety technology in modern cars, the vehicle's Automatic Emergency Braking (AEB) system is designed to avoid or reduce the risk of collision with obstacles ahead (such as other vehicles, pedestrians, bicycles, etc.) by automatically slowing down or stopping the car. However, when the existing automatic emergency braking system is tested and evaluated for performance on vulnerable road users, it mainly focuses on typical scenarios with a high probability of traffic accidents, and adopts unified evaluation criteria or rules, which can easily lead to the performance test results of the automatic emergency braking system not being universal, and thus failing to reflect the performance in real traffic accidents. Summary of the invention

[0004] In order to solve the above-mentioned technical defects that the existing automatic emergency braking system, when conducting performance test and evaluation for vulnerable road users, mainly focuses on typical scenarios with a high probability of traffic accidents and adopts unified evaluation standards or rules, which easily leads to the performance test results of the automatic emergency braking system not being universal and thus failing to reflect the performance in real traffic accidents, a performance test and evaluation method and system for an automatic emergency braking system are proposed, and the technical solution is as follows: In a first aspect, an embodiment of the present application provides a performance test and evaluation method for an automatic emergency braking system, comprising: Determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database for at least one accident description information corresponding to each type of participant; wherein the preset traffic accident database includes at least two traffic accident tags, accident description information corresponding to each traffic accident tag, and at least one type of accident participant; Divide all accident description information into scene information sets of at least two accident scenes, and calculate corresponding scene correction coefficients according to the scene information sets corresponding to each accident scene; The scoring results of the test vehicle in each accident scenario are obtained, and the performance test results of the test vehicle are obtained based on the scenario correction coefficient corresponding to each accident scenario and the scoring results.

[0005] In an optional solution of the first aspect, all accident description information is divided into scene information sets of at least two accident scenes, including: Identify the scene field label from each accident description information, and input every two scene field labels into the language processing model to predict the association result of the corresponding two scene field labels; wherein the language processing model is trained by at least two groups of scene field label samples and the association result samples corresponding to each group of scene field label samples; At least two scene field labels corresponding to at least two accident scenes are obtained according to all the association results, and based on all the scene field labels corresponding to each accident scene, a corresponding scene information set is obtained.

[0006] In another optional solution of the first aspect, based on all scene field labels corresponding to each accident scene, a corresponding scene information set is obtained, including: Based on each scene field label corresponding to each accident scene, the accident description information containing the corresponding scene field label is filtered out from all accident description information; The accident name information and the part damage description information are identified from each filtered accident description information, and all the accident name information and all the part damage description information corresponding to each accident scene are integrated into a corresponding scene information set.

[0007] In yet another optional solution of the first aspect, a corresponding scene correction coefficient is calculated according to a scene information set corresponding to each accident scene, including: According to all the accident name information corresponding to each accident scenario, the corresponding number of accidents is counted, and the sum of all accident numbers is calculated; The ratio between the number of accidents corresponding to each accident scenario and the summed result is taken as the corresponding scenario probability score, and the corresponding scenario correction coefficient is obtained based on the scenario probability score corresponding to each accident scenario and the damage description information of all parts.

[0008] In another optional solution of the first aspect, according to the scenario probability score corresponding to each accident scenario and the description information of all the damages of the parts, a corresponding scenario correction coefficient is obtained, including: From the injury description information of each body part corresponding to each accident scenario, identifying at least one body part and an injury description keyword corresponding to each body part; The injury score of each body part is determined based on a preset keyword-injury score correspondence table, and the accident injury score is calculated according to all the injury scores of the injury description information of each body part; wherein the preset keyword-injury score correspondence table includes at least two standard description keywords and the standard injury score corresponding to each standard description keyword; The sum of all accident damage scores corresponding to each accident scenario is calculated, and the ratio of the sum result to the corresponding number of accidents is used as the scenario damage score; The scenario probability score and scenario damage score corresponding to each accident scenario are multiplied and the product is used as the corresponding scenario correction coefficient.

[0009] In another optional solution of the first aspect, a performance test result of the test vehicle is obtained according to the scenario correction coefficient and the scoring result corresponding to each accident scenario, including: The scenario correction coefficient corresponding to each accident scenario and the scoring result are multiplied, and the performance test result of the test vehicle is determined based on the sum of all the product results.

[0010] In yet another optional solution of the first aspect, after obtaining the performance test result of the test vehicle, the method further includes: The preset traffic accident database is updated at preset time intervals, and when at least one new accident description information is detected corresponding to any participant type in the processed preset traffic accident database, the performance test results of the test vehicle are updated according to the processed preset traffic accident database.

[0011] In a second aspect, an embodiment of the present application provides a performance test and evaluation system for an automatic emergency braking system, comprising: A data extraction module, used to determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database for at least one accident description information corresponding to each type of participant; wherein the preset traffic accident database includes at least two traffic accident tags, accident description information corresponding to each traffic accident tag, and at least one type of accident participant; A coefficient calculation module, used to divide all accident description information into scene information sets of at least two accident scenes, and calculate the corresponding scene correction coefficient according to the scene information set corresponding to each accident scene; The test evaluation module is used to obtain the scoring results of the test vehicle in each accident scenario, and obtain the performance test results of the test vehicle based on the scenario correction coefficient corresponding to each accident scenario and the scoring results.

[0012] In a third aspect, an embodiment of the present application further provides a performance test and evaluation system for an automatic emergency braking system, including a processor and a memory; The processor is connected to the memory; A memory for storing executable program codes; The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the performance test and evaluation method for the automatic emergency braking system provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the performance testing and evaluation method for an automatic emergency braking system provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect can be implemented.

[0014] Beneficial effects of this application: When the automatic emergency braking system performs performance testing and evaluation on vulnerable road users, at least two types of participants are determined according to the preset number of accident participants, and at least one accident description information corresponding to each type of participant is queried in the preset traffic accident database; all accident description information is divided into scene information sets of at least two accident scenarios, and the corresponding scene correction coefficient is calculated according to the scene information set corresponding to each accident scenario; the scoring result of the test vehicle in each accident scenario is obtained, and the performance test result of the test vehicle is obtained according to the scene correction coefficient and the scoring result corresponding to each accident scenario. By determining all accident description information corresponding to the participant type in the preset traffic accident database, and dividing all accident description information into scene information sets of multiple accident scenarios, all types of accident scenarios are fully considered, so that the performance test result has universality and reliability; secondly, the corresponding scene correction coefficient can be calculated according to the scene information set corresponding to each accident scenario, and the performance test result of the test vehicle can be obtained according to the scene correction coefficient and the scoring result corresponding to each accident scenario, so as to enhance the correlation between the performance test result and the traffic accident, and effectively reflect the performance in the real traffic accident. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 An overall flow chart of a performance test and evaluation method for an automatic emergency braking system provided in an embodiment of the present application; Figure 2 A schematic diagram of a scene information set of an accident scene provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of a performance test and evaluation system for an automatic emergency braking system provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of another performance testing and evaluation system for an automatic emergency braking system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The technical solutions 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.

[0018] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, although the embodiment may not be clearly recorded in the following text.

[0019] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of the present application. Various processes or components may be appropriately omitted, substituted or added to each example. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.

[0020] See also Figure 1 , Figure 1 An overall flow chart of a performance testing and evaluation method for an automatic emergency braking system provided in an embodiment of the present application is shown.

[0021] like Figure 1 As shown, the performance test and evaluation method for the automatic emergency braking system may at least include the following steps: Step 102: Determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database to obtain at least one accident description information corresponding to each type of participant.

[0022] In the embodiment of the present application, the performance test and evaluation method for the automatic emergency braking system can be applied to, but not limited to, a test terminal, which can establish a connection with the vehicle control system of the test vehicle to control the test vehicle to drive normally according to the specified trajectory in different accident scenarios, and generate the scoring results of the automatic emergency braking system in the vehicle control system for vulnerable road users in each accident scenario. The vehicle control system of the test vehicle can be used to control the test vehicle to drive normally according to the specified trajectory, and can feed back the trigger state parameters (which can include but are not limited to trigger time, trigger relative distance, and trigger speed drop) corresponding to the automatic emergency braking system for vulnerable road users in different accident scenarios to the test terminal, so that the test terminal generates the scoring results corresponding to each accident scenario according to the trigger state parameters and the preset scoring rules. Here, the types of accident scenarios can include but are not limited to intersection accidents, lane collision accidents, lane rear-end accidents, and pedestrian crossing accidents, and the test personnel can simulate the layout in advance according to the scene layout information of each accident scene (for example, but not limited to the lane line size, the number of traffic lights, zebra crossings, weather conditions, and the location of vulnerable road users, etc.) to ensure the authenticity of the accident scene.

[0023] In addition, the test terminal can also establish a connection with the traffic accident platform, obtain all traffic accident information recorded by the traffic accident platform, and generate a preset traffic accident database according to all traffic accident information, and can combine the preset traffic accident database to obtain the scene correction coefficient for vulnerable road users in different accident scenes, and then use the scoring results corresponding to each accident scene mentioned above and the corresponding scene correction coefficient here to finally determine the performance test results of the automatic emergency braking system on the test vehicle for vulnerable road users. Here, the preset traffic accident database can include traffic accident labels extracted from each traffic accident information, accident description information corresponding to each traffic accident label, and one or more types of accident participants. The accident description information can at least include road type, accident information, information about the persons involved, and description information of the injuries, etc. For example, the traffic accident label extracted from a certain traffic accident information can be a vehicle-pedestrian collision (or three keyword labels of vehicle, pedestrian, and collision), and the accident description information corresponding to the traffic accident label can include the intersection, the identity driving information of the driver A, the identity information of the injured pedestrian B, and the facial skin abrasion of the injured pedestrian B. The type of accident participant corresponding to the traffic accident label can be a pedestrian crossing the intersection. Of course, the accident description information mentioned in the embodiment of the present application may also include but is not limited to information such as the time of the accident, the location of the accident, the status of traffic lights, and the size of the road, which will not be elaborated here.

[0024] It should be noted that the traffic accident platform can also regularly update all traffic accident information, and the test terminal can also update the preset traffic accident database in real time based on all updated traffic accident information to ensure the timeliness and reliability of the preset traffic accident database.

[0025] It can be understood that the test terminal can determine all accident description information corresponding to the participant type in the preset traffic accident database, and divide all accident description information into scene information sets of various accident scenarios to fully consider all types of accident scenarios, thereby making the performance test results universal and reliable; secondly, the corresponding scene correction coefficient can be calculated according to the scene information set corresponding to each accident scenario, and the performance test results of the test vehicle can be obtained according to the scene correction coefficient corresponding to each accident scenario and the scoring results, so as to enhance the correlation between the performance test results and the traffic accident, thereby effectively reflecting the performance in the real traffic accident.

[0026] Specifically, when the automatic emergency braking system conducts performance testing and evaluation on vulnerable road users, the test terminal can, but is not limited to, determine the number of vulnerable road users based on the preset number of accident participants set by the tester. For example, when the preset number of accident participants is two, the test terminal can determine that the number of vulnerable road users among the accident participants is one (the other may correspond to the accident vehicle); when the preset number of accident participants is three, the test terminal can determine that the number of vulnerable road users among the accident participants is two (the other may correspond to the accident vehicle), and so on. In the embodiments mentioned below in this application, the number of vulnerable road users is explained as one, and other numbers of vulnerable road users are not elaborated here.

[0027] Further, after determining that the number of vulnerable road users is one, the test terminal may, but is not limited to, use pedestrians, elderly, children, and disabled persons as keywords, and query all participant types corresponding to all keywords in the preset participant type database. For example, but not limited to, taking the keyword pedestrian as an example, all corresponding participant types may include pedestrians crossing the intersection, pedestrians riding bicycles, pedestrians crossing the road, and pedestrians running, etc.; for example, taking the keyword elderly as an example, all corresponding participant types may include elderly crossing the road and elderly on the roadside, etc.; for example, taking the keyword child as an example, all corresponding participant types may include children crossing the road, children walking on zebra crossings, children running, and children riding bicycles, etc.; for example, taking the keyword disabled persons as an example, all corresponding participant types may include people in wheelchairs and blind people, etc., and are not limited to all the participant types mentioned above. Here, the preset participant type database can be obtained by the tester in combination with natural language processing technology, including multiple standard keywords set by the tester, and multiple participant types associated with each standard keyword obtained by using natural language processing technology. Of course, the trained natural language processing model can also be used, which is also not limited to this.

[0028] It is understandable that when it is determined that the number of vulnerable road users is two, the test terminal can also, but is not limited to, use pedestrians and elderly people, pedestrians and children, pedestrians and disabled people, elderly people and children as keywords, and query all participant types corresponding to all keywords in the above-mentioned preset participant type database, which may include, for example, pedestrians on bicycles and elderly people crossing the road, running pedestrians and children on bicycles, pedestrians crossing the road and blind people, elderly people crossing the road and children on bicycles, etc., and is also not limited to this.

[0029] Furthermore, after obtaining all participant types, the test terminal can query a plurality of accident description information corresponding to all participant types in a preset traffic accident database. Each accident description information can correspond to any one of all participant types, and the accident information of each accident description information can include descriptive words associated with the participant type. For example, taking a participant type as a cyclist as an example, the accident information of the corresponding accident description information records two descriptive words, cyclist and pedestrian (or descriptive words associated with pedestrians), and is not limited thereto. Here, the preset traffic accident database can be generated by all traffic accident information provided by the above-mentioned traffic accident platform, including traffic accident labels extracted from each traffic accident information, accident description information corresponding to each traffic accident label, and one or more accident participant types. Each accident description information can at least include road type, accident information, information on persons involved, and description information on injuries.

[0030] Step 104: Divide all accident description information into scene information sets of at least two accident scenes, and calculate corresponding scene correction coefficients according to the scene information sets corresponding to each accident scene.

[0031] Specifically, after determining all the accident description information corresponding to the vulnerable road users, the test terminal can, but is not limited to, use natural language processing technology to identify and process each accident description information, so as to use the descriptive words corresponding to the preset scene field keywords queried in each accident description information as scene field labels, and input every two scene field labels into the language processing model, and obtain the association result between the corresponding two scene field labels by means of model prediction, and when the association result exceeds the preset result threshold, it indicates that the correlation between the corresponding two scene field labels is strong; when the association result does not exceed the preset result threshold, it indicates that the correlation between the corresponding two scene field labels is poor.

[0032] Here, the language processing model can be trained by multiple groups of scene field label samples and the associated result samples corresponding to each group of scene field label samples. Each group of scene field label samples and the corresponding associated result samples can be pre-set by the tester. Of course, in the embodiment of the present application, the similarity between each two scene field labels can also be calculated by a preset similarity algorithm, and when the similarity exceeds the preset similarity threshold, it indicates that the correlation between the corresponding two scene field labels is strong; when the similarity does not exceed the preset similarity threshold, it indicates that the correlation between the corresponding two scene field labels is poor.

[0033] Further, after predicting the association result between every two scene field labels, the test terminal can, but is not limited to, divide and process all scene field labels according to all the association results to obtain multiple sets of field label sets, each set of field label sets includes at least two scene field labels, and the association result between every two scene field labels in each set of field label sets exceeds the preset result threshold, and the scene type of the corresponding accident scene can be generated according to all the scene field labels corresponding to each set of field label sets. Here, when generating the scene type of each accident scene, the test terminal can also input all the scene field labels corresponding to each set of field label sets into a preset deep learning model to accurately summarize the scene type of the corresponding accident scene through model prediction, and is not limited to this.

[0034] Furthermore, after determining the scenario type of each accident scenario and all the corresponding scenario field labels, the test terminal can, but is not limited to, filter out the accident description information containing the corresponding scenario field label from all the accident description information according to each scenario field label corresponding to each accident scenario, and use natural language processing technology to identify the accident name information and the part injury description information from the filtered accident description information, and then integrate the scene type corresponding to each accident scenario, all the accident name information and all the part injury description information into a corresponding scene information set. Of course, the scene type corresponding to all accident scenarios, all the accident name information and all the part injury description information can also be integrated into a scene information set, but is not limited to this.

[0035] Also see here Figure 2 A schematic diagram of scene information set of an accident scene provided by an embodiment of the present application is shown, such as Figure 2 As shown, the scene information set of the accident scene may include two accident scenes, and the corresponding scene types may be respectively represented as accident scene A and accident scene B. Among them, accident scene A corresponds to accident name information A1 and accident name information A2, the accident name information A1 corresponds to part damage description information A1 and scene layout information A1, and the accident name information A2 corresponds to part damage description information A2 and scene layout information A2; accident scene B corresponds to accident name information B1 and accident name information B2, the accident name information B1 corresponds to part damage description information B1 and scene layout information B1, and the accident name information B2 corresponds to part damage description information B2 and scene layout information B2. It can be understood that the scene layout information mentioned in the scene information set of the accident scene can be, but is not limited to, obtained by analyzing the accident description information by the test terminal, and may specifically include information such as lane line size, number of traffic lights, zebra crossings, weather conditions, and the location of vulnerable road users, and is not limited thereto.

[0036] Further, after obtaining the scene information set corresponding to each accident scene, the test terminal may, but is not limited to, count the number of all accident name information corresponding to each accident scene, and use the number of all accident name information as the number of accidents of the corresponding accident scene, then sum the number of accidents of all accident scenes, and use the ratio between the number of accidents corresponding to each accident scene and the sum result as the corresponding scene probability score. Here, the calculation method of the scene probability score of each accident scene may, but is not limited to, refer to the following: In the above formula, Can correspond to the first The scenario probability score of the accident scenario, Can correspond to the first The number of accidents in each accident scenario, It can correspond to the sum of the number of accidents in all accident scenarios, and n can correspond to the number of accidents in each accident scenario.

[0037] Further, after obtaining the scene information set corresponding to each accident scene, the test terminal can also, but is not limited to, use natural language processing technology to identify all body parts and the injury description keywords corresponding to each body part from the injury description information of each part corresponding to each accident scene. For example, three body parts of face, upper limbs and lower limbs can be identified from the injury description information of a certain part, and the injury description keyword corresponding to the face is skin abrasion, the injury description keyword corresponding to the upper limb is tendon rupture, and the injury description keyword corresponding to the lower limb is large-area soft tissue avulsion. Here, the types of all body parts can include, but are not limited to, face, chest, abdomen, spine, upper limbs and lower limbs, and the injury description keywords can include, but are not limited to, skin abrasion, nasal bone fracture, orbital fracture, maxillofacial composite fracture, large-area soft tissue avulsion, soft tissue contusion, tendon rupture, vascular rupture, liver and spleen contusion, vertebral compression fracture, joint dislocation and vascular injury, etc., and are not limited to this.

[0038] Further, after obtaining the injury description keywords corresponding to each body part, the test terminal can query the injury score of each body part in the preset keyword-injury score correspondence table, and calculate the corresponding accident injury value according to all the injury scores corresponding to the injury description information of each part. Here, the preset keyword-injury score correspondence table can be, but is not limited to, obtained by preset analysis by the tester, including multiple standard description keywords set by the tester, and the standard injury score corresponding to each standard description keyword, for example, when the standard description keyword is skin abrasion, the corresponding standard injury score is 1; when the standard description keyword is tendon tear, nasal bone fracture, orbital fracture, liver and spleen contusion or vertebral compression fracture, the corresponding standard injury score is 2; when the standard description keyword is joint dislocation, the corresponding standard injury score is 3; when the standard description keyword is large-area soft tissue avulsion or vascular injury, the corresponding standard injury score is 4, and is not limited thereto.

[0039] It can be understood that when calculating the corresponding accident damage value based on all the part damage scores corresponding to the damage description information of each part, the test terminal can count the number of all the part damage scores corresponding to the damage description information of each part. When it is detected that the number of all the part damage scores exceeds a preset number threshold (for example, it can be set to 3), the three part damage scores with the highest scores can be screened out from all the part damage scores, and the squares of the three part damage scores can be summed up and the sum result can be used as the corresponding accident damage value; when it is detected that the number of all the part damage scores does not exceed the preset number threshold, the squares of all the part damage scores can be directly summed up and the sum result can be used as the corresponding accident damage value.

[0040] Furthermore, after calculating the accident damage score corresponding to the damage description information of each part, the test terminal can sum up all the accident damage scores corresponding to each accident scene, and use the ratio between the summation result and the number of accidents in each accident scene as the corresponding scene damage score. Here, the scene damage score calculation method for each accident scene can be, but is not limited to, as follows: In the above formula, Can correspond to the first The scene damage score of the accident scene i can correspond to The number of all accident damage scores for this accident scenario, Can correspond to the first The sum of all accident injury scores of the accident scenario can be expressed as The damage score of each accident scenario is Can correspond to the first The number of accidents in each accident scenario.

[0041] Furthermore, after respectively obtaining the scenario probability score and the scenario damage score corresponding to each accident scenario, the corresponding scenario correction coefficient can be obtained by the following calculation method: In the above formula, Can correspond to the first The scenario correction factor for the accident scenario is Can correspond to the first The scenario probability score of the accident scenario, Can correspond to the first The scenario damage score of each accident scenario.

[0042] Step 106: Obtain the scoring result of the test vehicle in each accident scenario, and obtain the performance test result of the test vehicle according to the scenario correction coefficient corresponding to each accident scenario and the scoring result.

[0043] Specifically, after determining the scene information set for each accident scenario, the test terminal can, but is not limited to, feed back the scene type in each scene information set to the tester (of course, the above-mentioned scene layout information can also be fed back to the tester, but is not limited to this), so that the tester can arrange the corresponding simulated accident scene; then, after the tester places the test vehicle and the simulated robot corresponding to the vulnerable road user at the designated position in each simulated accident scene, the test terminal can control the test vehicle to travel along the designated trajectory through the vehicle control system, and receive in real time the trigger state parameters corresponding to the vulnerable road users in different accident scenarios by the automatic emergency braking system.

[0044] It can be understood that the trigger state parameters may include but are not limited to trigger time, trigger relative distance, and trigger speed drop, and after receiving the trigger state parameters, the test terminal may also score the trigger state parameters according to preset scoring rules to obtain a scoring result corresponding to each accident scenario.

[0045] Further, after obtaining the scoring result of the test vehicle in each accident scenario, the test terminal may, but is not limited to, perform product calculations on the scenario correction coefficient corresponding to each accident scenario and the scoring result, and may take the sum of all the product results as the scoring result of the test vehicle in different accident scenarios. Here, the scoring result calculation formula of the test vehicle in different accident scenarios may, but is not limited to, refer to the following: In the above formula, T can correspond to the scoring result of the test vehicle under different accident scenarios. It can be corresponded to the product result of each accident scenario, It can correspond to the number of all accident scenes. Can correspond to the first The product of the accident scenarios is Can correspond to the first The scoring results of the accident scenarios are Can correspond to Scenario correction factor for the accident scenario.

[0046] Furthermore, after obtaining the scoring results of the test vehicle under different accident scenarios, the test terminal can also determine the corresponding performance test results based on the scoring range in which the scoring results are located. For example, but not limited to, when the scoring result is in a preset first scoring range, the corresponding performance test result can be determined to be excellent; when the scoring result is in a preset second scoring range, the corresponding performance test result can be determined to be qualified; when the scoring result is in a preset second scoring range, the corresponding performance test result can be determined to be poor, and is not limited to this.

[0047] As an option of the embodiment of the present application, after obtaining the performance test result of the test vehicle, the method further includes: The preset traffic accident database is updated at preset time intervals, and when at least one new accident description information is detected corresponding to any participant type in the processed preset traffic accident database, the performance test results of the test vehicle are updated according to the processed preset traffic accident database.

[0048] Specifically, in order to ensure the real-time and accuracy of the performance test results of the test vehicle, the test terminal can also obtain multiple new traffic accident information from the above-mentioned traffic accident platform at preset time intervals, and update the preset traffic accident database based on the multiple new traffic accident information. It can be understood that when it is detected that any of the above-mentioned participant types corresponds to at least one new accident description information, it indicates that the scene correction coefficient corresponding to each accident scene has changed, and then the performance test results of the test vehicle can be obtained again according to the preset traffic accident database after the update processing, and then the performance test results of the test vehicle can be updated; when it is detected that there is no new accident description information for any of the above-mentioned participant types, it indicates that the scene correction coefficient corresponding to each accident scene has not changed, and there is no need to update the performance test results of the test vehicle.

[0049] In one or more embodiments of the present application, the test terminal not only effectively improves the correlation between the scoring results and traffic accidents, so that the scoring results can take into account the probability of occurrence of different accident scenarios and the differences in personal injuries, but also when the number of traffic accidents gradually increases with the accumulation, the scene correction coefficient can also be modified synchronously, thereby ensuring that the scene correction coefficient used can represent the actual coverage of traffic accidents for vulnerable road users to the greatest extent.

[0050] See next Figure 3 , Figure 3 A schematic structural diagram of a performance testing and evaluation system for an automatic emergency braking system provided in an embodiment of the present application is shown.

[0051] like Figure 3 As shown, the performance test and evaluation system for the automatic emergency braking system may include at least a data extraction module 301, a coefficient calculation module 302 and a test and evaluation module 303, wherein: The data extraction module 301 is used to determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database for at least one accident description information corresponding to each type of participant; wherein the preset traffic accident database includes at least two traffic accident tags, accident description information corresponding to each traffic accident tag, and at least one type of accident participant; The coefficient calculation module 302 is used to divide all accident description information into scene information sets of at least two accident scenes, and calculate the corresponding scene correction coefficient according to the scene information set corresponding to each accident scene; The test evaluation module 303 is used to obtain the scoring result of the test vehicle in each accident scenario, and obtain the performance test result of the test vehicle according to the scenario correction coefficient corresponding to each accident scenario and the scoring result.

[0052] In some possible embodiments, all accident description information is divided into scene information sets of at least two accident scenes, including: Identify the scene field label from each accident description information, and input every two scene field labels into the language processing model to predict the association result of the corresponding two scene field labels; wherein the language processing model is trained by at least two groups of scene field label samples and the association result samples corresponding to each group of scene field label samples; At least two scene field labels corresponding to at least two accident scenes are obtained according to all the association results, and based on all the scene field labels corresponding to each accident scene, a corresponding scene information set is obtained.

[0053] In some possible embodiments, based on all scene field labels corresponding to each accident scene, a corresponding scene information set is obtained, including: Based on each scene field label corresponding to each accident scene, the accident description information containing the corresponding scene field label is filtered out from all accident description information; The accident name information and the part damage description information are identified from each filtered accident description information, and all the accident name information and all the part damage description information corresponding to each accident scene are integrated into a corresponding scene information set.

[0054] In some possible embodiments, the corresponding scene correction coefficient is calculated according to the scene information set corresponding to each accident scene, including: According to all the accident name information corresponding to each accident scenario, the corresponding number of accidents is counted, and the sum of all accident numbers is calculated; The ratio between the number of accidents corresponding to each accident scenario and the summed result is taken as the corresponding scenario probability score, and the corresponding scenario correction coefficient is obtained based on the scenario probability score corresponding to each accident scenario and the damage description information of all parts.

[0055] In some possible embodiments, the corresponding scenario correction coefficient is obtained according to the scenario probability score corresponding to each accident scenario and the description information of all the damage parts, including: From the injury description information of each body part corresponding to each accident scenario, identifying at least one body part and an injury description keyword corresponding to each body part; The injury score of each body part is determined based on a preset keyword-injury score correspondence table, and the accident injury score is calculated according to all the injury scores of the injury description information of each body part; wherein the preset keyword-injury score correspondence table includes at least two standard description keywords and the standard injury score corresponding to each standard description keyword; The sum of all accident damage scores corresponding to each accident scenario is calculated, and the ratio of the sum result to the corresponding number of accidents is used as the scenario damage score; The scenario probability score and scenario damage score corresponding to each accident scenario are multiplied and the product is used as the corresponding scenario correction coefficient.

[0056] In some possible embodiments, the performance test results of the test vehicle are obtained according to the scenario correction coefficient and the scoring result corresponding to each accident scenario, including: The scenario correction coefficient corresponding to each accident scenario and the scoring result are multiplied, and the performance test result of the test vehicle is determined based on the sum of all the product results.

[0057] In some possible embodiments, after obtaining the performance test result of the test vehicle, the method further includes: The preset traffic accident database is updated at preset time intervals, and when at least one new accident description information is detected corresponding to any participant type in the processed preset traffic accident database, the performance test results of the test vehicle are updated according to the processed preset traffic accident database.

[0058] Those skilled in the art can clearly understand that the technical solutions of the embodiments of the present application can be implemented with the help of software and / or hardware. The "unit" and "module" in this specification refer to software and / or hardware that can independently complete or cooperate with other components to complete specific functions, where the hardware can be, for example, a field programmable gate array (FPGA), an integrated circuit (IC), etc.

[0059] See next Figure 4 , Figure 4 A schematic structural diagram of another performance testing and evaluation system for an automatic emergency braking system provided in an embodiment of the present application is shown.

[0060] like Figure 4 As shown, the performance test and evaluation system 400 for the automatic emergency braking system may include at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .

[0061] The communication bus 402 may be used to realize the connection and communication among the above-mentioned components.

[0062] The user interface 403 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.

[0063] The network interface 404 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0064] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the performance test and evaluation system 400 for the automatic emergency braking system, and executes various functions and processes data of the performance test and evaluation system 400 for the automatic emergency braking system by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 401 can integrate one or a combination of CPU, GPU, modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented separately through a chip.

[0065] The memory 405 may include a RAM or a ROM. Optionally, the memory 405 includes a non-transitory computer-readable medium. The memory 405 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may optionally be at least one storage device located away from the aforementioned processor 401. As Figure 4 As shown, the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a performance test and evaluation application for an automatic emergency braking system.

[0066] Specifically, the processor 401 may be used to call the performance test and evaluation application for the automatic emergency braking system stored in the memory 405, and specifically perform the following operations: Determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database for at least one accident description information corresponding to each type of participant; wherein the preset traffic accident database includes at least two traffic accident tags, accident description information corresponding to each traffic accident tag, and at least one type of accident participant; Divide all accident description information into scene information sets of at least two accident scenes, and calculate corresponding scene correction coefficients according to the scene information sets corresponding to each accident scene; The scoring results of the test vehicle in each accident scenario are obtained, and the performance test results of the test vehicle are obtained based on the scenario correction coefficient corresponding to each accident scenario and the scoring results.

[0067] In some possible embodiments, all accident description information is divided into scene information sets of at least two accident scenes, including: Identify the scene field label from each accident description information, and input every two scene field labels into the language processing model to predict the association result of the corresponding two scene field labels; wherein the language processing model is trained by at least two groups of scene field label samples and the association result samples corresponding to each group of scene field label samples; At least two scene field labels corresponding to at least two accident scenes are obtained according to all the association results, and based on all the scene field labels corresponding to each accident scene, a corresponding scene information set is obtained.

[0068] In some possible embodiments, based on all scene field labels corresponding to each accident scene, a corresponding scene information set is obtained, including: Based on each scene field label corresponding to each accident scene, the accident description information containing the corresponding scene field label is filtered out from all accident description information; The accident name information and the part damage description information are identified from each filtered accident description information, and all the accident name information and all the part damage description information corresponding to each accident scene are integrated into a corresponding scene information set.

[0069] In some possible embodiments, the corresponding scene correction coefficient is calculated according to the scene information set corresponding to each accident scene, including: According to all the accident name information corresponding to each accident scenario, the corresponding number of accidents is counted, and the sum of all accident numbers is calculated; The ratio between the number of accidents corresponding to each accident scenario and the summed result is taken as the corresponding scenario probability score, and the corresponding scenario correction coefficient is obtained based on the scenario probability score corresponding to each accident scenario and the damage description information of all parts.

[0070] In some possible embodiments, the corresponding scenario correction coefficient is obtained according to the scenario probability score corresponding to each accident scenario and the description information of all the damage parts, including: From the injury description information of each body part corresponding to each accident scenario, identifying at least one body part and an injury description keyword corresponding to each body part; The injury score of each body part is determined based on a preset keyword-injury score correspondence table, and the accident injury score is calculated according to all the injury scores of the injury description information of each body part; wherein the preset keyword-injury score correspondence table includes at least two standard description keywords and the standard injury score corresponding to each standard description keyword; The sum of all accident damage scores corresponding to each accident scenario is calculated, and the ratio of the sum result to the corresponding number of accidents is used as the scenario damage score; The scenario probability score and scenario damage score corresponding to each accident scenario are multiplied and the product is used as the corresponding scenario correction coefficient.

[0071] In some possible embodiments, the performance test results of the test vehicle are obtained according to the scenario correction coefficient and the scoring result corresponding to each accident scenario, including: The scenario correction coefficient corresponding to each accident scenario and the scoring result are multiplied, and the performance test result of the test vehicle is determined based on the sum of all the product results.

[0072] In some possible embodiments, after obtaining the performance test result of the test vehicle, the method further includes: The preset traffic accident database is updated at preset time intervals, and when at least one new accident description information is detected corresponding to any participant type in the processed preset traffic accident database, the performance test results of the test vehicle are updated according to the processed preset traffic accident database.

[0073] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0074] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0075] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0076] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0077] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0078] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

Claims

1. A performance test and evaluation method for an automatic emergency braking system, characterized in that: include: Determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database for at least one accident description information corresponding to each of the participant types; wherein the preset traffic accident database includes at least two traffic accident tags, accident description information corresponding to each of the traffic accident tags, and at least one type of accident participant; Dividing all the accident description information into scene information sets of at least two accident scenes, and calculating a corresponding scene correction coefficient according to the scene information set corresponding to each accident scene; The scoring result of the test vehicle in each of the accident scenarios is obtained, and the performance test result of the test vehicle is obtained according to the scenario correction coefficient corresponding to each of the accident scenarios and the scoring result.

2. The method according to claim 1, characterized in that The dividing all the accident description information into scene information sets of at least two accident scenes includes: Identify a scene field label from each of the accident description information, and input every two of the scene field labels into a language processing model to predict the association results of the corresponding two scene field labels; wherein the language processing model is trained by at least two groups of scene field label samples and the association result samples corresponding to each group of the scene field label samples; At least two of the scene field labels corresponding to at least two accident scenes are obtained according to all of the association results, and a corresponding scene information set is obtained based on all of the scene field labels corresponding to each of the accident scenes.

3. The method according to claim 2, characterized in that The step of obtaining a corresponding scene information set based on all the scene field labels corresponding to each accident scene includes: Based on each of the scene field labels corresponding to each of the accident scenes, the accident description information containing the corresponding scene field label is filtered out from all the accident description information; The accident name information and the part damage description information are identified from each of the filtered accident description information, and all the accident name information and all the part damage description information corresponding to each of the accident scenes are integrated into a corresponding scene information set.

4. The method according to claim 3, characterized in that The calculating of the corresponding scene correction coefficient according to the scene information set corresponding to each accident scene includes: According to all the accident name information corresponding to each accident scene, the corresponding number of accidents is counted, and the sum of all the accident numbers is calculated; The ratio between the number of accidents corresponding to each accident scenario and the summed result is used as the corresponding scenario probability score, and the corresponding scenario correction coefficient is obtained based on the scenario probability score corresponding to each accident scenario and all the part damage description information.

5. The method according to claim 4, characterized in that The corresponding scenario correction coefficient is obtained according to the scenario probability score corresponding to each accident scenario and all the part damage description information, including: Identifying at least one body part and an injury description keyword corresponding to each body part from the injury description information of each body part corresponding to each accident scene; The part injury score of each of the body parts is determined based on a preset keyword-injury score correspondence table, and the accident injury score is calculated according to all the part injury scores of each part injury description information; wherein the preset keyword-injury score correspondence table includes at least two standard description keywords and the standard injury score corresponding to each of the standard description keywords; Sum up all the accident damage scores corresponding to each accident scenario, and use the ratio of the summation result to the corresponding number of accidents as the scenario damage score; The scenario probability score and the scenario damage score corresponding to each accident scenario are multiplied and the product is used as the corresponding scenario correction coefficient.

6. The method according to claim 1, characterized in that The step of obtaining the performance test result of the test vehicle according to the scenario correction coefficient corresponding to each accident scenario and the scoring result includes: The scenario correction coefficient corresponding to each of the accident scenarios and the scoring result are multiplied, and the performance test result of the test vehicle is determined based on the sum of all the product results.

7. The method according to claim 1, characterized in that After obtaining the performance test result of the test vehicle, the method further includes: The preset traffic accident database is updated at preset time intervals, and when it is detected in the processed preset traffic accident database that any one of the participant types has at least one new accident description information, the performance test results of the test vehicle are updated according to the processed preset traffic accident database.

8. A performance test and evaluation system for an automatic emergency braking system, characterized in that: include: A data extraction module, used to determine at least two types of participants according to a preset number of accident participants, and query a preset traffic accident database for at least one accident description information corresponding to each of the participant types; wherein the preset traffic accident database includes at least two traffic accident tags, accident description information corresponding to each of the traffic accident tags, and at least one type of accident participant; A coefficient calculation module, used to divide all the accident description information into scene information sets of at least two accident scenes, and calculate a corresponding scene correction coefficient according to the scene information set corresponding to each accident scene; The test evaluation module is used to obtain the scoring result of the test vehicle in each of the accident scenarios, and obtain the performance test result of the test vehicle according to the scenario correction coefficient corresponding to each of the accident scenarios and the scoring result.

9. A performance test and evaluation system for an automatic emergency braking system, characterized in that: including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the steps of the method according to any one of claims 1 to 7.

Images