Automobile automatic driving system combination test method, electronic equipment and medium

Abstract

The invention relates to the field of automatic driving, in particular to an automobile automatic driving system combined test method, electronic equipment and a medium. The automobile automatic driving system combination test method comprises the following steps: determining a matching threshold value corresponding to each functional scene according to the complexity and risk degree of each scene in a scene library; each scene comprises each function scene and each test scene; according to the comprehensive index of each function scene, the comprehensive index of the pre-selected test scene corresponding to each function scene, and the matching threshold value corresponding to each function scene, determining a primarily selected test scene; determining similarity according to each function scene and the primarily selected test scene; determining the effectiveness of the primarily selected test scene according to the comprehensive index and the similarity of the primarily selected test scene and the coverage rate of the primarily selected test scene; and determining the test scene of each function scene according to the validity of the primarily selected test scene. The economical efficiency, comprehensiveness and accuracy of automatic driving testing can be remarkably improved, and the testing period is shortened.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving, and more specifically, to a combined testing method, electronic equipment and medium for an automobile autonomous driving system. Background Art

[0002] Autonomous driving is not only the research frontier in the field of vehicle engineering, but also the development direction of the future automobile industry. It is also an important technical means to solve problems such as traffic safety, energy waste and environmental pollution. With the development of autonomous driving technology, traffic scenes are becoming more and more complex and the number of traffic participants is increasing. It is difficult for a single test method to ensure the test effect. At present, the most accepted and recognized autonomous driving test method by the international community is the "multi-pillar method" autonomous driving test criteria, which is to test through multiple channels and methods such as simulation testing, field testing and actual road testing. How to organically integrate the three aspects of autonomous driving vehicle testing, closed test sites and open road testing is a research hotspot that the entire automotive industry and academia are currently paying attention to. Compared with simulation testing, the combination of closed road testing and open road testing has gradually become a key method to verify the performance of autonomous driving systems. At present, the research focus has shifted from a single simulation test to a more comprehensive closed and open road combination test. Closed road testing ensures the safety, repeatability and accuracy of the test by simulating extreme conditions and high-risk scenarios in a controlled environment; open road testing is carried out in a real traffic environment, which can comprehensively evaluate the intelligent response and adaptability of autonomous driving systems in complex traffic flows, emergencies and random scenarios. Through the organic combination of the two, the accuracy and safety of closed roads and the reality and diversity of open roads can be effectively utilized to ensure that the performance of the autonomous driving system in different environments is fully verified, providing a scientific basis for its safety and reliability on actual roads. However, the current autonomous driving system has difficulty in selecting test scenarios and insufficient coverage in complex and dynamic road environments, which reduces the accuracy of the test.

[0003] In view of this, this application is hereby filed. Summary of the invention

[0004] The purpose of this application is to provide a combined testing method, electronic equipment and medium for an automobile autonomous driving system, so as to solve the problems existing in the prior art of difficulty in selecting test scenarios and insufficient coverage of autonomous driving systems in complex and dynamic road environments, significantly improve the economy, comprehensiveness and accuracy of autonomous driving tests, and shorten the testing cycle.

[0005] In order to achieve the above objectives, this application adopts the following technical solutions: In a first aspect, the present application provides a combined testing method for an automobile autonomous driving system, comprising: According to the complexity and danger level of each scenario in the scenario library, determine the matching threshold corresponding to each functional scenario; each scenario includes each functional scenario and each test scenario; Determine the preliminary test scenario corresponding to each functional scenario according to the comprehensive index of each functional scenario, the comprehensive index of the pre-selected test scenario corresponding to each functional scenario, and the matching threshold corresponding to each functional scenario; Determine the similarity of each functional scenario based on each functional scenario and the preliminary test scenario corresponding to each functional scenario; Determine the validity of the preliminary test scenarios corresponding to the functional scenarios according to the comprehensive indicators of the preliminary test scenarios corresponding to the functional scenarios, the similarities under the functional scenarios, and the coverage of the preliminary test scenarios corresponding to the functional scenarios; According to the effectiveness of the preliminary test scenarios corresponding to the functional scenarios, the test scenarios for each functional scenario are determined.

[0006] In some technical solutions, the risk level is determined in the following manner: The degree of danger is determined based on the speed difference between the vehicle and the other vehicle, the mass of the vehicle, the mass of the other vehicle, the heading angle of the vehicle, the heading angle of the other vehicle, and the collision time.

[0007] In some technical solutions, determining the matching threshold corresponding to each functional scenario according to the complexity and danger level of each scenario in the scenario library includes: Determine the threshold adjustment factor corresponding to each functional scenario based on the complexity and danger level of each scenario in the scenario library; According to the threshold adjustment factors corresponding to the functional scenarios, the matching thresholds corresponding to the functional scenarios are determined.

[0008] In some technical solutions, determining the matching threshold corresponding to each functional scenario according to the threshold adjustment factor corresponding to each functional scenario includes: When the threshold adjustment factor corresponding to each functional scenario is less than or equal to a preset value, determining the matching threshold corresponding to each functional scenario to be the preset value; When the threshold adjustment factor corresponding to each functional scenario is greater than a preset value, the matching threshold corresponding to each functional scenario is determined to be the threshold adjustment factor.

[0009] In some technical solutions, the comprehensive index is determined in the following manner: Determine the comprehensive indicators of each scenario based on the complexity and danger level of each scenario in the scenario library.

[0010] In some technical solutions, the comprehensive index of each scene is determined according to the complexity and danger of each scene in the scene library, including: Determine the complexity boundary value according to the complexity of each scene in the scene library; Determine the danger level boundary value according to the danger level of each scene in the scene library; The comprehensive index of each scenario is determined based on the complexity and danger level, complexity boundary value, and danger boundary value of each scenario in the scenario library.

[0011] In some technical solutions, determining the similarity under each functional scenario according to each functional scenario and the preliminary test scenario corresponding to each functional scenario includes: The similarity in each functional scenario is determined based on the elements in each functional scenario and the elements in the preliminary test scenario corresponding to each functional scenario.

[0012] In some technical solutions, the coverage of the preliminary test scenarios corresponding to the functional scenarios is determined in the following manner: The coverage of the preliminary test scenarios corresponding to each functional scenario is determined based on the number of autonomous driving functions that can be satisfied by the preliminary test scenarios corresponding to each functional scenario and the number of autonomous driving functions that can be satisfied by this vehicle.

[0013] In a second aspect, the present application provides an electronic device, including: at least one processor, and a memory communicatively coupled to at least one of the processors; The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can perform the above method.

[0014] In a third aspect, the present application provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the above method.

[0015] Compared with the prior art, the beneficial effects of this application are: The combined test method of the automobile automatic driving system provided by the present application includes: determining the matching threshold corresponding to each functional scene according to the complexity and danger of each scene in the scene library; each scene includes each functional scene and each test scene; determining the preliminary test scene corresponding to each functional scene according to the comprehensive index of each functional scene, the comprehensive index of the pre-selected test scene corresponding to each functional scene, and the matching threshold corresponding to each functional scene; determining the similarity under each functional scene according to each functional scene and the preliminary test scene corresponding to each functional scene; determining the effectiveness of the preliminary test scene corresponding to each functional scene according to the comprehensive index of the preliminary test scene corresponding to the functional scene, the similarity under each functional scene, and the coverage of the preliminary test scene corresponding to each functional scene; determining the test scene of each functional scene according to the effectiveness of the preliminary test scene corresponding to each functional scene. This method can cope with the challenges of long test cycle and high test cost of actual road testing of automatic driving, realize automatic matching of functional scenes and test scenes, and screen out reasonable test scenes for each functional scene, so as to guide the testing of automatic driving functions according to the different test characteristics of closed sites and open roads.

[0016] This application is compared horizontally with the traditional allocation method of actual road testing of autonomous driving, and the test cycle and cost are better than the traditional method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 It is a flowchart of a combined test method for an automobile automatic driving system provided in this application; Figure 2 It is a structural schematic diagram of a combined test device for an automobile automatic driving system provided by the present application; Figure 3 It is a structural schematic diagram of the electronic device provided by this application. DETAILED DESCRIPTION

[0019] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0020] As mentioned in the background technology, the existing technology has the problem of difficulty in selecting test scenarios and insufficient coverage in complex and dynamic road environments for autonomous driving systems. In response to this, this application uses the complexity, danger, comprehensive indicators, matching thresholds, coverage, etc. of each scenario to determine the effectiveness of the preliminary test scenarios corresponding to each functional scenario, thereby screening out test scenarios for each functional scenario, thereby improving the economy, comprehensiveness and accuracy of autonomous driving tests and shortening the test cycle. The following is a further detailed description of this application in conjunction with the embodiments.

[0021] Example 1 Figure 1 It is a flow chart of a combined test method for an automobile automatic driving system provided in this embodiment. The method can be executed by a combined test device for an automobile automatic driving system. The device can be composed of software and / or hardware and is generally integrated in an electronic device. The electronic device can be an electronic computer. For ease of understanding, each step in the method of this embodiment is executed by an electronic computer.

[0022] like Figure 1 As shown, this embodiment provides a combined testing method for an automobile automatic driving system, comprising the following steps: S110. Determine a matching threshold corresponding to each functional scenario according to the complexity and risk level of each scenario in the scenario library; the scenarios include each functional scenario and each test scenario.

[0023] The complexity refers to the number of traffic elements (such as vehicles, pedestrians, traffic lights, obstacles, etc.) involved in the scene and their interaction methods. The complexity is divided into two aspects: static and dynamic. The complexity of static scenes introduces the information entropy theory, calculates the sum of the information of discrete information sources, and considers the type and weight of static scene elements; the complexity of dynamic scenes introduces the acceleration gravity model, considering the mutual influence factors such as the distance and speed between the scene participants and the test vehicle.

[0024] Static complexity refers to the complexity of static elements in the scene. The higher the static complexity, the more difficult it is to reproduce the scene. Optionally, the static complexity coefficient calculation formula is: in, is the complexity coefficient of the static scene; The total number of types of grouping labels corresponding to each scene component in the static scene complexity; It is the ratio of the number of nodes of the same type to the total number of nodes in the graph structure constructed according to the complexity of the static scene.

[0025] Optionally, the static scene complexity C1 is calculated as: Among them, m is the number of groups of static scene components; It is the first element of the dynamic scene. The weight corresponding to the group; It is the first element of the static scene. The preset values ​​corresponding to each scene component in the group.

[0026] Dynamic complexity refers to the motion state of traffic participants and the vehicle in the scene. Dynamic complexity also affects the difficulty of realizing dynamic traffic flow in scene reproduction. It is related to the dynamic factors in the scene.

[0027] Optionally, the dynamic complexity calculation formula is: in, is the complexity of dynamic scenes; To test the property values ​​of the vehicle; is the attribute value of the traffic participant; To test the vehicle's reaction time; To test the vehicle's driving speed; To test the distance between vehicles and traffic participants; To test the angle between the vehicle and other traffic participants; is the time when the traffic participant is sensed; is the speed of traffic participants; is the movement distance of traffic participants; is a constant.

[0028] Optionally, the complexity calculation formula is: The evaluation index of the danger level is mainly the degree of damage that may occur to the vehicle in an accident. This indicator directly reflects the potential risks faced by the vehicle in a specific scenario. This danger level is closely related to multiple factors. First, the vehicle mass ratio is one of the important factors affecting the severity of the accident. A large mass difference may cause light vehicles to be more seriously damaged in a collision. Secondly, the smaller the collision time, the higher the danger level. Finally, the angle of the collision is also a key variable. Frontal collisions, side collisions, and rear collisions will cause varying degrees of damage to the vehicle. Taking these factors into consideration, the danger level of the vehicle in a specific scenario can be more accurately assessed, thereby providing a more reliable basis for the decision-making of the autonomous driving system.

[0029] In an optional embodiment, the risk level is determined in the following manner: The degree of danger is determined based on the speed difference between the vehicle and the other vehicle, the mass of the vehicle, the mass of the other vehicle, the heading angle of the vehicle, the heading angle of the other vehicle, and the collision time.

[0030] Optionally, the risk D is calculated using the following formula: in, is the potential collision severity index; T is the collision time; is the heading angle position weight; is the relative speed weight; is the relative quality weight; are the position coordinates of the vehicle and other vehicles respectively; is the speed difference between the own car and the other car; To prevent a tiny constant with a zero denominator; and are the masses of the vehicle and other vehicles respectively; and are the heading angles of the own vehicle and other vehicles respectively; By adjusting the size of the heading angle sensitive parameter, it can adapt to collision scenarios with different accuracy; is the relative distance between the two vehicles; is the safe stopping distance between two vehicles; represents the relative speed of the two vehicles; In this embodiment, the vehicle is a vehicle in the test scene library determined by the method of this embodiment, and the other vehicle is a vehicle that appears in the scene other than the vehicle.

[0031] The complexity and danger level of each scene in the scene library are obtained by quantitative analysis after comprehensively considering the meaning of the physical values ​​covered by the scene, the classification and combination of the scene elements involved, and the scene logic division rules. For example, in the highway scene, the vehicle driving behavior is single, the lanes are clear, the scene logic is simple, and the complexity is low, but due to the high speed, the danger level is high; if there is bad weather (such as fog or icy and snowy roads), the danger level increases significantly. In the city intersection scene, multiple traffic participants such as vehicles, pedestrians, and bicycles intersect, the traffic signals are complex, the priorities are changeable, and the scene complexity is high, but because the speed is generally low, the danger level varies depending on the traffic volume.

[0032] In an optional implementation, determining the matching threshold corresponding to each functional scenario according to the complexity and risk level of each scenario in the scenario library includes: Determine the threshold adjustment factor corresponding to each functional scenario based on the complexity and danger level of each scenario in the scenario library; According to the threshold adjustment factors corresponding to the functional scenarios, the matching thresholds corresponding to the functional scenarios are determined.

[0033] In an optional implementation, determining the matching threshold corresponding to each functional scenario according to the threshold adjustment factor corresponding to each functional scenario includes: When the threshold adjustment factor corresponding to each functional scenario is less than or equal to a preset value, determining the matching threshold corresponding to each functional scenario to be the preset value; When the threshold adjustment factor corresponding to each functional scenario is greater than a preset value, the matching threshold corresponding to each functional scenario is determined to be the threshold adjustment factor.

[0034] Among them, the preset value can be 0.9.

[0035] In order to meet the testing requirements of functional scenarios to the greatest extent, the matching model between functional scenarios and test scenarios is an association mapping model, that is, a single functional scenario needs to be matched with multiple groups of test scenarios, and only matches that meet the threshold requirements can be considered as test scenarios that can represent functional scenarios for testing.

[0036] Furthermore, due to the diversity and uncertainty of scene elements, the scene elements that affect the complexity and danger of the scene are also different. When some of these elements have a greater impact on the autonomous driving function, in order to make the result more accurate, the threshold of the matching model should be adjusted on the original basis, that is, the matching model should adopt dynamic threshold matching according to the two dimensions of the scene. Threshold The calculation formula is: in, is the threshold adjustment factor.

[0037] S120. Determine the preliminary test scenario corresponding to each functional scenario according to the comprehensive index of each functional scenario, the comprehensive index of the pre-selected test scenario corresponding to each functional scenario, and the matching threshold corresponding to each functional scenario.

[0038] Optionally, determining the preliminary test scenario corresponding to each functional scenario according to the comprehensive index of each functional scenario, the comprehensive index of the pre-selected test scenario corresponding to each functional scenario, and the matching threshold corresponding to each functional scenario includes: The preliminary test scenarios corresponding to the functional scenarios are determined according to the ratio of the comprehensive index of each functional scenario to the comprehensive index of the pre-selected test scenarios corresponding to the functional scenarios and the matching thresholds corresponding to the functional scenarios.

[0039] Optionally, when the ratio is greater than the matching threshold corresponding to each functional scenario, the pre-selected test scenario corresponding to the ratio is determined as the preliminary test scenario. For example, for a functional scenario (such as AEB), the pre-selected test scenarios include A1, A2, A3...An, and if the comprehensive index of the functional scenario AEB is Sx, and the comprehensive indexes of the pre-selected test scenarios are S1, S2, S3...Sn, then the ratios are Sx / S1, Sx / S2, Sx / S3...Sx / Sn, respectively. If the matching threshold corresponding to the functional scenario AEB is , if the above ratio is greater than The preliminary test scenarios are Sx / S1, Sx / S2, Sx / S5, Sx / S6, Sx / S7, Sx / S9, and Sx / S10. The preliminary test scenarios are the pre-selected scenarios corresponding to the comprehensive indicators S1, S2, S5, S6, S7, S9, and S10 respectively.

[0040] The relative importance of complexity and danger is mainly weighed according to the needs of the scenario, which include safety, efficiency, comfort, flexibility and stability. If risk control is the focus, then danger is more important; if the scenario focuses more on coping with a changing and complex operating environment, complexity is more critical. Through actual scenario experience and historical data, the weight ratio of complexity and danger can be further optimized to determine the comprehensive indicators of the scenario, which can be used to more accurately evaluate the intelligent performance of the autonomous driving system.

[0041] In an optional implementation, the comprehensive index is determined in the following manner: Determine the comprehensive indicators of each scenario based on the complexity and danger level of each scenario in the scenario library.

[0042] In an optional implementation, determining the comprehensive index of each scene according to the complexity and danger level of each scene in the scene library includes: Determine the complexity boundary value according to the complexity of each scene in the scene library; Determine the danger level boundary value according to the danger level of each scene in the scene library; The comprehensive index of each scenario is determined based on the complexity and danger level, complexity boundary value, and danger boundary value of each scenario in the scenario library.

[0043] The complexity boundary value includes the maximum complexity value and the minimum complexity value. The maximum complexity value is the maximum complexity value of all scenarios, and the minimum complexity value is the minimum complexity value of all scenarios. The risk boundary value includes the maximum risk value and the minimum risk value. The maximum risk value is the maximum risk value of all scenarios, and the minimum risk value is the minimum risk value of all scenarios.

[0044] Optionally, comprehensive indicators for each scenario S i The calculation formula is: in, are the first weight coefficient and the second weight coefficient of each scene respectively. is the minimum complexity, is the maximum complexity, is the minimum risk value, is the maximum risk level, is the complexity of each scene, is the danger level of each scene.

[0045] S130: Determine the similarity of each functional scenario according to each functional scenario and the preliminary test scenario corresponding to each functional scenario.

[0046] In an optional implementation manner, determining the similarity under each functional scenario according to each functional scenario and the preliminary test scenario corresponding to each functional scenario includes: The similarity in each functional scenario is determined based on the elements in each functional scenario and the elements in the preliminary test scenario corresponding to each functional scenario.

[0047] The purpose of performing scene element similarity tests is to ensure that although the comprehensive indicators in different scene combinations are similar, there are differences between the scene elements, so as to avoid potential deviations in the evaluation of the autonomous driving system due to different scene elements.

[0048] Similarity is defined as the degree of similarity between different types of elements in the test scenario and the functional scenario. It is used to refine the comparison between the test scenario and the actual scenario to ensure that they not only match in complexity and dangerousness, but also have a high degree of consistency between key elements.

[0049] Optionally, the similarity calculation formula is: in, They are the number of elements that appear in A but not in B, the number of elements that do not appear in A but appear in B, and the number of elements that appear in both A and B. A and B represent a functional scenario and a preliminary test scenario corresponding to the functional scenario, respectively.

[0050] S140. Determine the validity of the preliminary test scenarios corresponding to the functional scenarios according to the comprehensive indicators of the preliminary test scenarios corresponding to the functional scenarios, the similarities under the functional scenarios, and the coverage of the preliminary test scenarios corresponding to the functional scenarios.

[0051] In an optional implementation manner, the coverage of the preliminary test scenarios corresponding to the functional scenarios is determined in the following manner: The coverage of the preliminary test scenarios corresponding to each functional scenario is determined based on the number of autonomous driving functions that can be satisfied by the preliminary test scenarios corresponding to each functional scenario and the number of autonomous driving functions that can be satisfied by this vehicle.

[0052] Coverage is defined as the extent to which the intelligent driving system can cover various scenarios during its operation. Increased scene complexity and danger will lead to more challenging scenarios, thereby increasing the system's coverage requirements for various scenarios.

[0053] Optionally, coverage The calculation formula is: in, The number of autonomous driving functions that can be satisfied by the preliminary test scenario. The number of autonomous driving functions that the vehicle can provide.

[0054] When evaluating the preliminary test scenarios, the following three principles are used: (1) comprehensive scenario indicators. The closer the indicators are, the better. (2) similarity. The greater the similarity, the better. (3) coverage. The greater the coverage, the better.

[0055] Optionally, the validity calculation formula is: in, is the weight coefficient, the default , the weight coefficient can be adjusted according to the test requirements.

[0056] S150. Determine a test scenario for each functional scenario based on the effectiveness of the preliminarily selected test scenarios corresponding to each functional scenario.

[0057] Optionally, the preliminary selected test scenario with the greatest effectiveness among the preliminary selected test scenarios corresponding to each functional scenario is selected as the test scenario for the functional scenario. This step optimizes the allocation of test resources by screening out the test scenario with the highest effectiveness.

[0058] The above method can cope with the challenges of long test cycle and high test cost in actual road testing of autonomous driving, realize automatic matching of functional scenarios and test scenarios, and screen out reasonable test scenarios for each functional scenario, so as to guide the testing of autonomous driving functions according to the different test characteristics of closed venues and open roads.

[0059] This application is compared horizontally with the traditional allocation method of actual road testing of autonomous driving, and the test cycle and cost are better than the traditional method.

[0060] Optionally, this embodiment further includes the steps of extracting key scene elements and test scene primitives of typical intelligent driving functions and constructing a function scene library and a test scene library before S110.

[0061] Among them, the key scene elements include road elements, traffic elements, temporary facility elements, dynamic traffic elements, weather and sudden dynamic elements and V2X elements; the test scene primitives include entry scene primitives and following vehicle scene primitives.

[0062] Road elements are elements of the road on which smart cars travel, and are divided into six categories: road type, road surface, road edge, road geometry, road intersection, and road characteristics; traffic elements are elements of the traffic environment in which smart cars travel, and are divided into three categories: speed limit, traffic conditions, and driving area; temporary facility elements are various types of temporary targets encountered within the ODD range of smart cars, and are divided into traffic safety facilities, road users, non-road users, and obstacles; dynamic traffic elements mainly refer to the vehicle behavior model jointly formed by the main smart car and surrounding vehicles or other traffic participants during the driving process of the vehicle, including three types of vehicle behavior related to surrounding vehicles, vehicle behavior related to other traffic participants, and vehicle behavior related to blind spots; weather and sudden dynamic elements mainly refer to the weather, road conditions affected by weather, particulate matter, and lighting conditions that vehicles may encounter within the ODD range, mainly including weather, road conditions affected by weather, particulate matter, and lighting conditions; V2X elements mainly refer to the environmental elements of the vehicle networking function within the ODD range, mainly including user density, data rate, reliability, and communication distance.

[0063] A test scenario primitive is a long test scenario sequence that is decomposed into test scenario sequence fragments in which the state of the scene elements remains unchanged. When all traffic vehicles are present, the traffic scene is complex and the requirements for the functions of the intelligent vehicle under test are high. Such scenario primitives are defined as entry scenario primitives; when there is no lane-changing vehicle and other traffic vehicles are present, such scenario primitives are defined as following vehicle scenario primitives.

[0064] The functional scenario library includes 329 benchmark scenarios (136 on highways and 193 in cities), involving six types of intelligent driving functions, including AVP, HWP, NOA, TJA, ALC, and AEB; the test scenario library is based on following scenarios and entry scenarios, and is derived, constructed, and expanded through a generative neural network model to derive test scenario primitives.

[0065] Example 2 like Figure 2 As shown, this embodiment provides a combined test device for an automobile automatic driving system, comprising: The matching threshold determination module 201 is used to determine the matching threshold corresponding to each functional scenario according to the complexity and risk level of each scenario in the scenario library; the scenarios include each functional scenario and each test scenario; A preliminary test scenario determination module 202 is used to determine the preliminary test scenario corresponding to each functional scenario according to the comprehensive index of each functional scenario, the comprehensive index of the pre-selected test scenario corresponding to each functional scenario, and the matching threshold corresponding to each functional scenario; A similarity determination module 203 is used to determine the similarity of each functional scenario according to each functional scenario and the preliminary test scenario corresponding to each functional scenario; The validity determination module 204 is used to determine the validity of the preliminary test scenarios corresponding to the functional scenarios according to the comprehensive indicators of the preliminary test scenarios corresponding to the functional scenarios, the similarities under the functional scenarios, and the coverage of the preliminary test scenarios corresponding to the functional scenarios; The test scenario determination module 205 is used to determine the test scenario for each functional scenario according to the validity of the preliminary selected test scenarios corresponding to each functional scenario.

[0066] The device is used to execute the above method, and thus at least has functional modules and beneficial effects corresponding to the above method.

[0067] Example 3 like Figure 3 As shown, this embodiment provides an electronic device, including: at least one processor; and a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to perform the above method. At least one processor in the electronic device can perform the above method, and thus has at least the same advantages as the above method.

[0068] Optionally, the electronic device also includes interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on a memory to display graphical information of a GUI (Graphical User Interface) on an external input / output device (such as a display device coupled to an interface). In other embodiments, if necessary, multiple processors can be used together with multiple memories, and / or multiple buses can be used together with multiple memories. Similarly, multiple electronic devices can be connected (for example, as a server array, a group of blade servers, or a multi-processor system), and each device provides some necessary operations. Figure 3 A processor 301 is taken as an example.

[0069] The memory 302, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the automobile automatic driving system combination test method in the embodiment of the present application (for example, the matching threshold determination module, the preliminary test scenario determination module, the similarity determination module, the validity determination module, and the test scenario determination module in the automobile automatic driving system combination test device). The processor 301 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 302, that is, realizing the above-mentioned automobile automatic driving system combination test method.

[0070] The memory 302 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 302 may further include a memory remotely arranged relative to the processor 301, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0071] The electronic device may further include: an input device 303 and an output device 304. The processor 301, the memory 302, the input device 303 and the output device 304 may be connected via a bus or other means. Figure 3 The example of connecting through bus is taken in the following.

[0072] The input device 303 can receive input digital or character information, and the output device 304 can include a display device, an auxiliary lighting device (e.g., LED), and a tactile feedback device (e.g., a vibration motor), etc. The display device can include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some embodiments, the display device can be a touch screen.

[0073] Example 4 This embodiment provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the above method. The computer instructions on the computer-readable storage medium are used to enable a computer to execute the above method, and thus have at least the same advantages as the above method.

[0074] The medium in this application may adopt any combination of one or more computer-readable media. The medium may be a computer-readable signal medium or a computer-readable storage medium. The medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of the medium (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, the medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device, or device.

[0075] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0076] The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.

[0077] Computer program code for performing the operation of the present application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., using an Internet service provider to connect through the Internet).

[0078] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution disclosed in this application can be achieved, and this document is not limited here.

[0079] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A combined testing method for an automobile automatic driving system, characterized in that: include: According to the complexity and danger level of each scenario in the scenario library, determine the matching threshold corresponding to each functional scenario; each scenario includes each functional scenario and each test scenario; Determine the preliminary test scenario corresponding to each functional scenario according to the comprehensive index of each functional scenario, the comprehensive index of the pre-selected test scenario corresponding to each functional scenario, and the matching threshold corresponding to each functional scenario; Determine the similarity of each functional scenario based on each functional scenario and the preliminary test scenario corresponding to each functional scenario; Determine the validity of the preliminary test scenarios corresponding to the functional scenarios according to the comprehensive indicators of the preliminary test scenarios corresponding to the functional scenarios, the similarities under the functional scenarios, and the coverage of the preliminary test scenarios corresponding to the functional scenarios; According to the effectiveness of the preliminary test scenarios corresponding to the functional scenarios, the test scenarios for each functional scenario are determined.

2. The combined testing method for the automobile automatic driving system according to claim 1, characterized in that: The risk level is determined in the following manner: The degree of danger is determined based on the speed difference between the vehicle and the other vehicle, the mass of the vehicle, the mass of the other vehicle, the heading angle of the vehicle, the heading angle of the other vehicle, and the collision time.

3. The combined testing method for the automobile automatic driving system according to claim 1, characterized in that: Determining the matching threshold corresponding to each functional scenario according to the complexity and danger level of each scenario in the scenario library includes: Determine the threshold adjustment factor corresponding to each functional scenario based on the complexity and danger level of each scenario in the scenario library; According to the threshold adjustment factors corresponding to the functional scenarios, the matching thresholds corresponding to the functional scenarios are determined.

4. The combined testing method for the automobile automatic driving system according to claim 3 is characterized in that: The determining, according to the threshold adjustment factor corresponding to each functional scenario, a matching threshold corresponding to each functional scenario includes: When the threshold adjustment factor corresponding to each functional scenario is less than or equal to a preset value, determining the matching threshold corresponding to each functional scenario to be the preset value; When the threshold adjustment factor corresponding to each functional scenario is greater than a preset value, the matching threshold corresponding to each functional scenario is determined to be the threshold adjustment factor.

5. The combined testing method for the automobile automatic driving system according to claim 1, characterized in that: The comprehensive index is determined in the following way: Determine the comprehensive indicators of each scenario based on the complexity and danger level of each scenario in the scenario library.

6. The combined testing method for the automobile automatic driving system according to claim 5, characterized in that: The comprehensive index of each scenario is determined according to the complexity and danger of each scenario in the scenario library, including: Determine the complexity boundary value according to the complexity of each scene in the scene library; Determine the danger level boundary value according to the danger level of each scene in the scene library; The comprehensive index of each scenario is determined based on the complexity and danger level, complexity boundary value, and danger boundary value of each scenario in the scenario library.

7. The combined testing method for the automobile automatic driving system according to claim 1, characterized in that: Determining the similarity of each functional scenario according to each functional scenario and the preliminary test scenario corresponding to each functional scenario includes: The similarity in each functional scenario is determined based on the elements in each functional scenario and the elements in the preliminary test scenario corresponding to each functional scenario.

8. The combined testing method for the automobile automatic driving system according to claim 1, characterized in that: The coverage of the preliminary test scenarios corresponding to the functional scenarios is determined in the following manner: The coverage of the preliminary test scenarios corresponding to each functional scenario is determined based on the number of autonomous driving functions that can be satisfied by the preliminary test scenarios corresponding to each functional scenario and the number of autonomous driving functions that can be satisfied by this vehicle.

9. An electronic device, characterized in that: include: at least one processor, and a memory communicatively coupled to at least one of the processors; The memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors so that the at least one processor can execute the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 8.

Images