Methods and related equipment for generating extreme test cases for automated parking systems

Abstract

This application belongs to the field of intelligent driving and relates to a method and related equipment for generating extreme test cases for automatic parking in parking lots. It solves the technical problems in the existing technology of generating extreme test cases for automatic parking in parking lots, such as parameter combination explosion leading to waste of computing resources and low efficiency, insufficient accuracy of risk quantification making it difficult to guarantee the extremeness of test cases, lack of reasonable timing logic leading to low reference value of test results, and unclear screening criteria making it impossible to accurately obtain key test cases. It can efficiently generate test cases that are extreme, have reasonable timing, and can effectively trigger the system's extreme response, thereby improving the effectiveness and safety verification level of extreme scenario testing of automatic parking systems.

Description

Technical Field

[0001] This application relates to the field of intelligent driving, and in particular to a method, system, computer equipment, and storage medium for generating extreme test cases for automatic parking in a parking lot. Background Technology

[0002] With the rapid development of intelligent driving technology, automated parking systems have become one of the core functions of vehicle intelligence, and their safety and reliability are directly related to the safety of users' lives and property. To ensure the stable operation of automated parking systems in complex scenarios, it is necessary to verify their extreme performance through a large number of extreme test cases. Extreme test cases typically refer to scenarios that contain high-risk elements (such as pedestrians suddenly entering at close range, narrow parking spaces, and adverse environmental interference) and have reasonable timing logic, which can effectively expose potential defects in the system.

[0003] However, existing extreme test case generation technologies suffer from the following key problems: wasted computing resources, low efficiency, and unreasonable test case logic, resulting in low reference value of the results. These problems ultimately affect the effectiveness and safety verification of extreme scenario testing of automatic parking systems, and fail to fully expose potential system defects to ensure their reliable operation in complex real-world scenarios.

[0004] In view of the above, this application is hereby submitted. Summary of the Invention

[0005] The purpose of this application is to propose a method, system, computer device, and storage medium for generating extreme test cases for automatic parking systems, in order to solve the current technical problem of not being able to efficiently generate test cases that are extreme, have reasonable timing, and can effectively trigger the system's extreme response, thereby improving the effectiveness and safety verification level of extreme scenario testing for automatic parking systems.

[0006] To address the aforementioned technical problems, this application provides a method for generating extreme test cases for automated parking systems, employing the following technical solution:

[0007] A method for generating extreme test cases for automated parking systems in a parking lot includes the following steps:

[0008] Acquire static environmental data, dynamic participant data, and environmental interference data of the parking lot corresponding to the test vehicle;

[0009] The static environment data, the dynamic participant data, and the environmental disturbance data are parameterized to obtain static environment parameters, dynamic participant parameters, and environmental disturbance parameters.

[0010] By combining the static environment parameters, the dynamic participant parameters, and the environmental interference parameters, multiple scenario parameter combinations are obtained;

[0011] Risk quantification is performed on each of the scenario parameter combinations to obtain the risk level of each scenario parameter combination;

[0012] Based on a preset parameter range, at least two scene parameter combinations are selected from multiple scene parameter combinations as the initial scene combination;

[0013] Based on the dynamic participant parameters, determine the time constraints for the dynamic participants;

[0014] Based on the risk level and the time constraints, the matching degree of each initial scenario combination and extreme test case target requirement is determined, and the initial scenario combination with the highest matching degree is used as the extreme test case of the test vehicle.

[0015] Furthermore, the combination of the static environment parameters, the dynamic participant parameters, and the environmental disturbance parameters yields multiple scene parameter combinations, including:

[0016] Based on the static environment parameters, the dynamic participant parameters, and the environmental interference parameters, filter out the associated parameter groups that have physical relationships;

[0017] Within the range that satisfies the preset physical constraints, the associated parameter group is combined to generate an initial combination;

[0018] In the initial combination, combinations that do not meet the spatial compatibility conditions are removed to obtain multiple combinations of the scene parameters.

[0019] Furthermore, the step of quantifying the risk of each scenario parameter combination to obtain the risk level of each scenario parameter combination includes:

[0020] For each combination of scenario parameters, extract the core attribute parameters of the dynamic participants and assign basic weights to each core attribute parameter;

[0021] Based on the environmental disturbance parameters, each of the basic weights is corrected to obtain the initial risk weight;

[0022] Obtain historical fault data, and adjust each of the initial risk weights based on the historical fault data to obtain the final risk weight value;

[0023] The final risk weight value is compared with the preset risk level to determine the risk level corresponding to each combination of scenario parameters.

[0024] Furthermore, the dynamic participant parameters include initial position and velocity-related parameters, relative distance-related parameters, and type identifier-related parameters;

[0025] The step of setting time constraints for dynamic participants based on the dynamic participant parameters includes:

[0026] Based on the initial position and the speed-related parameters, a timing constraint is set for the dynamic participant to appear within a preset time after the test vehicle starts parking mode.

[0027] Based on the relative distance related parameters, the parking phase is divided into parking space search, parking, and correction phases, and phase constraints are set for dynamic participants to enter the preset range of the test vehicle during the parking phase.

[0028] Based on the type identifier-related parameters, constraints are set on the order in which multiple types of dynamic participants enter the preset range of the test vehicle.

[0029] Furthermore, based on the risk level and the time constraints, the matching degree of each initial scenario combination and the target requirements of the extreme test cases is determined, and the initial scenario combination with the highest matching degree is used as the extreme test case of the test vehicle, including:

[0030] The risk level is quantified to obtain a risk value, and the degree to which the initial scenario combination satisfies the time constraint is quantified to obtain a time value.

[0031] According to the preset weight ratio corresponding to the risk level and the preset weight ratio corresponding to the degree of satisfaction of the time constraint, the risk value and time value corresponding to the initial scenario combination are weighted and summed to obtain the matching degree of the initial scenario combination.

[0032] Sort all the initial scenario combinations by their matching degree, and select the initial scenario combination with the highest matching degree as the extreme test case.

[0033] Furthermore, the static environmental parameters include parking space geometric parameters, fixed obstacle parameters, and parking lot topology parameters;

[0034] The environmental interference parameters include illumination parameters, meteorological parameters, and sensor interference parameters.

[0035] A parking lot automatic parking extreme test case generation system, used to execute the parking lot automatic parking extreme test case generation method, includes:

[0036] The data acquisition module is used to acquire static environmental data, dynamic participant data, and environmental interference data of the parking lot corresponding to the test vehicle;

[0037] The parameterization processing module is used to define the static environment data, the dynamic participant data, and the environmental interference data as parameters to obtain static environment parameters, dynamic participant parameters, and environmental interference parameters.

[0038] The parameter combination module is used to combine the static environment parameters, the dynamic participant parameters, and the environmental interference parameters to obtain multiple scenario parameter combinations;

[0039] The risk level calculation module is used to quantify the risk of each combination of scenario parameters to obtain the risk level of each combination of scenario parameters.

[0040] The initial combination filtering module is used to select at least two scene parameter combinations as initial scene combinations from multiple scene parameter combinations according to a preset parameter range.

[0041] The time constraint setting module is used to determine the time constraint conditions for the dynamic participants based on the dynamic participant parameters.

[0042] The matching degree evaluation and test case generation module is used to determine the matching degree of each initial scenario combination and extreme test case target requirement based on the risk level and the time constraint, and to use the initial scenario combination with the highest matching degree as the extreme test case of the test vehicle.

[0043] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution:

[0044] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the method for generating extreme test cases for automatic parking in a parking lot as described above.

[0045] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below:

[0046] A computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of the method for generating extreme test cases for automatic parking in a parking lot as described above.

[0047] Compared with the prior art, the embodiments of this application have the following main advantages:

[0048] This invention acquires and parameterizes data on the static environment, dynamic participants, and environmental interference of a parking lot. It then quantifies the risks associated with parameter combinations, selects initial scenario combinations based on preset parameter ranges, sets time constraints based on dynamic participant parameters, and finally determines the matching degree through risk level and time constraints, selecting the initial scenario combination with the highest matching degree as an extreme test case. This effectively solves the problem of wasted computational resources and low efficiency caused by parameter combination explosion, improves the accuracy of risk quantification to ensure the extreme nature of test cases, ensures the rationality of scenario timing logic through time constraints, and clarifies screening criteria based on matching degree to accurately obtain extreme test cases that can trigger the system's extreme response. This improves the effectiveness and safety verification level of extreme scenario testing for automatic parking systems. Attached Figure Description

[0049] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0051] Figure 2 This is a flowchart of an embodiment of the method for generating extreme test cases for automatic parking in a parking lot according to this application;

[0052] Figure 3 This is a schematic diagram of a structure of an embodiment of the automatic parking test case generation system for parking lots according to this application;

[0053] Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0055] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0056] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0057] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0058] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0059] Terminal devices 101, 102, and 103 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptops, and desktop computers, etc.

[0060] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0061] It should be noted that the method for generating extreme test cases for automatic parking in parking lots provided in this application is generally executed by a server, and correspondingly, the system for generating extreme test cases for automatic parking in parking lots is generally set up on a server.

[0062] It should be understood that Figure 1The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0063] Continue to refer to Figure 2 The diagram illustrates a flowchart of an embodiment of the method for generating extreme test cases for automated parking systems according to this application. The method for generating extreme test cases for automated parking systems includes the following steps:

[0064] Step S1: Obtain static environmental data, dynamic participant data, and environmental interference data of the parking lot corresponding to the test vehicle.

[0065] In this embodiment, the parking lot automatic parking extreme test case generation method runs on electronic devices (e.g., Figure 1 The server shown can send or receive data via wired or wireless connections. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future known wireless connection methods.

[0066] In this embodiment, static environmental data refers to physical environmental information in the parking lot that does not change over time, including but not limited to the dimensions of parking spaces (such as length and width), the positions of fixed obstacles (such as the coordinates of columns and walls), the width of parking lot passages, and the turning radius. For example, the static environmental data of an underground parking lot in a shopping mall may include "parking space width 2.2m, length 5.5m, column coordinates (10m, 5m), passage width 4m".

[0067] Dynamic participant data refers to information related to objects moving in the parking lot over time, including the real-time location, speed, direction of movement, and type (e.g., pedestrians, cars) of pedestrians, non-motorized vehicles (such as bicycles), and motorized vehicles (such as other parked vehicles); for example, "Pedestrian A's initial position (8m, 6m), speed 1.5m / s, direction towards the test vehicle."

[0068] Environmental interference data refers to information on environmental factors that affect the perception and decision-making of the parking system, including light intensity (such as backlight, low light), weather conditions (such as rain, fog), and sensor obstruction (such as cameras being blocked by dirt); for example, "light intensity 8000 lux (backlight scene), visibility 50m (light fog)".

[0069] Step S2: Define the static environment data, the dynamic participant data, and the environmental interference data in a parameterized manner to obtain static environment parameters, dynamic participant parameters, and environmental interference parameters.

[0070] In this embodiment, parameterization refers to transforming the above-mentioned raw data into a quantifiable and computable parameter form, so that the data has a standardized value range and format.

[0071] Static environmental parameters are the parameterized results of static environmental data, such as parking space parameters (width W=2.2m±0.2m, length L=5.5m±0.5m) and fixed obstacle parameters (type is column, coordinates (x=10m, y=5m), size 30cm×30cm).

[0072] Dynamic participant parameters are the parameterized results of dynamic participant data, such as pedestrian parameters (type identifier "pedestrian", initial position (x=8m, y=6m), speed v=1.5m / s±0.3m / s) and motor vehicle parameters (type identifier "car", initial speed 0m / s (stationary state)).

[0073] Environmental interference parameters are parameterized results of environmental interference data, such as illumination parameters (intensity I = 8000 lux ± 1000 lux, angle between the light source direction and the test vehicle 120° (backlight)) and meteorological parameters (fog level 2 (light), visibility 50m ± 5m).

[0074] Step S3: Combine the static environment parameters, the dynamic participant parameters, and the environmental interference parameters to obtain multiple scene parameter combinations.

[0075] In this embodiment, the scene parameter combination refers to selecting specific parameter values ​​from static environment parameters, dynamic participant parameters, and environmental interference parameters to form a complete scene description; for example, "static environment parameters (parking space width 2.0m, column coordinates (10m, 5m)) + dynamic participant parameters (pedestrian speed 1.8m / s, initial position (8m, 6m)) + environmental interference parameters (illuminance 8000 lux, fog level 2)" constitute a scene parameter combination.

[0076] Specifically, the combination process needs to ensure the physical rationality between parameters (such as the dynamic participant's position not overlapping with fixed obstacles), and finally generate multiple scene parameter combinations that conform to real-world logic (such as 100 different parameter combinations).

[0077] Step S4. Quantify the risk of each scenario parameter combination to obtain the risk level of each scenario parameter combination;

[0078] In this embodiment, risk quantification refers to assessing the degree of danger of a scenario through preset rules. Specifically, it involves assigning basic risk values ​​to the core attributes of dynamic participants (such as type, speed, and distance from the test vehicle), and then correcting them by combining environmental interference (such as backlighting, which increases perceived risk) to finally obtain a quantified risk value.

[0079] Risk levels are categorized by risk values ​​(e.g., low, medium, high). For example, a scenario with a risk value of 85 (out of 100) corresponds to a "high risk level," while another scenario with a risk value of 30 corresponds to a "low risk level."

[0080] Step S5. Select at least two scene parameter combinations from the multiple scene parameter combinations as the initial scene combination according to the preset parameter range;

[0081] In this embodiment, the preset parameter range refers to the extreme scenario boundaries set for each parameter (based on physical limits and historical test data), such as "parking space width ≤ 2.0m (narrow parking space), distance between pedestrian and test vehicle ≤ 2m (close distance), light intensity ≥ 7000 lux (strong light)".

[0082] The initial scene combination is selected from all scene parameter combinations that meet the above extreme parameter range. For example, the combination of "parking space width 1.9m, pedestrian distance 1.5m, backlight" is selected as the initial scene combination.

[0083] Step S6. Based on the dynamic participant parameters, determine the time constraints for the dynamic participants;

[0084] In this embodiment, the time constraint refers to the behavioral rules of the dynamic participant in the time dimension, which are set based on parameters such as the dynamic participant's position and speed, and include:

[0085] Timing constraints (e.g., pedestrians must appear within 2-3 seconds after the test vehicle starts and parks);

[0086] Phase constraints (e.g., pedestrians must enter the danger zone when the test vehicle enters the parking phase (not the search phase));

[0087] Sequence constraints (e.g., non-motorized vehicles must enter the danger zone after pedestrians).

[0088] Step S7. Based on the risk level and the time constraint, determine the matching degree of each initial scenario combination and extreme test case target requirement, and take the initial scenario combination with the highest matching degree as the extreme test case of the test vehicle.

[0089] In this embodiment, the target requirements of extreme test cases refer to the set of core requirements that extreme test cases must meet, including "extremeness" (high risk, which can reach the system performance boundary), "time sequence rationality" (dynamic participant behavior conforms to the parking process logic and can effectively trigger the system response), and "response observability" (the scenario can trigger a quantifiable response from the system, which is convenient for performance evaluation). The matching degree refers to the comprehensive quantitative evaluation value of the initial scenario combination in the two core dimensions of "extremeness" (corresponding to the risk level) and "time sequence rationality" (corresponding to the degree of time constraint satisfaction). It is calculated by weighted summation through quantitative indicators, where the risk level corresponds to 60% (extremeness is the core) and the degree of time constraint satisfaction corresponds to 40% (time sequence rationality ensures test effectiveness). That is to say, it is calculated by risk level (e.g., high risk corresponds to 80 points) and time constraint satisfaction (e.g., fully satisfying the constraint corresponds to 100 points) according to preset weights (e.g., risk accounts for 60% and time sequence accounts for 40%), for example, "80×60%+100×40%=88 points". Extreme test cases refer to the initial scenario combination with the highest matching degree. For example, a combination with a matching degree of 92 points is significantly higher than other combinations and is selected as an extreme test case (such as the scenario of "narrow parking space + backlight + sudden appearance of pedestrian at close range + meeting the timing constraints").

[0090] This invention acquires and parameterizes data on the static environment, dynamic participants, and environmental interference of a parking lot. It then quantifies the risks associated with parameter combinations, selects initial scenario combinations based on preset parameter ranges, sets time constraints based on dynamic participant parameters, and finally determines the matching degree through risk level and time constraints, selecting the initial scenario combination with the highest matching degree as an extreme test case. This effectively solves the problem of wasted computational resources and low efficiency caused by parameter combination explosion, improves the accuracy of risk quantification to ensure the extreme nature of test cases, ensures the rationality of scenario timing logic through time constraints, and clarifies screening criteria based on matching degree to accurately obtain extreme test cases that can trigger the system's extreme response. This improves the effectiveness and safety verification level of extreme scenario testing for automatic parking systems.

[0091] In some optional implementations of this embodiment, the step of combining the above-mentioned static environment parameters, the dynamic participant parameters, and the environmental interference parameters to obtain multiple scene parameter combinations specifically includes:

[0092] Step S31. Based on the static environmental parameters, the dynamic participant parameters, and the environmental interference parameters, filter the associated parameter groups that have physical correlations;

[0093] In this embodiment, physical association refers to the dependency relationship between different types of parameters formed by physical rules or scene logic (i.e., parameters influence each other and cannot exist independently), for example:

[0094] The "parking space width" parameter in the static environment parameters is related to the "vehicle width" parameter in the dynamic participant parameters (the parking space width must be greater than the vehicle width for parking to be permitted).

[0095] The "motion trajectory" parameter in the dynamic participant parameters is related to the "channel turning radius" parameter in the static environment parameters (the trajectory must adapt to the channel turning radius, otherwise a collision will occur).

[0096] The environmental disturbance parameter "light intensity" is related to the dynamic participant parameter "visibility" (low light reduces the probability of dynamic participants being perceived).

[0097] A related parameter group refers to a set of parameters that are physically related. These groups need to be logically grouped according to the scenario, for example:

[0098] First related group: {Parking space width (static), vehicle width (dynamic), aisle width (static)} (all related to space capacity);

[0099] The second association group: {light intensity (environment), dynamic participant type (dynamic), sensor occlusion level (environment)} (all related to perception difficulty).

[0100] The screening method is to automatically identify and group parameters by using a preset association rule library (such as "parking space parameters must match vehicle size" and "environmental interference must be associated with the sensing object"), and exclude unrelated parameter combinations (such as "parking space tilt angle" and "rain and fog concentration" have no direct physical relationship and are not included in the same group).

[0101] Step S32. Within the range that satisfies the preset physical constraints, the associated parameter group is matched with values ​​to generate an initial combination;

[0102] In this embodiment, the preset physical constraints refer to the parameter value restrictions set based on real physical laws or the rationality of the scenario, to ensure that the combination does not violate objective laws, for example:

[0103] Spatial constraints: Parking space width ≥ vehicle width + 0.5m (with spare space on both sides for error correction), the initial position of dynamic participants must be within the physical boundary of the parking lot (e.g., x coordinate ≤ 50m, y coordinate ≤ 30m, based on the actual size of the parking lot);

[0104] Movement constraints: Pedestrian speed ≤ 5m / s (human running limit), motor vehicle speed ≤ 10km / h in parking lot (safety regulations).

[0105] Environmental constraints: Light intensity ≥ 0 lux (physical lower limit), rain and fog concentration ≤ 90% (if these limits are exceeded, the scene has no practical testing significance).

[0106] Value combination refers to selecting specific values ​​for each parameter in the associated parameter group and combining them within the constraints. For example, for the first associated group {parking space width, vehicle width, aisle width}:

[0107] Parking space width options: 2.0m, 2.2m, 2.4m (all satisfying ≥ vehicle width + 0.5m);

[0108] Available width values ​​for motor vehicles: 1.5m, 1.7m (both ≤ parking space width - 0.5m);

[0109] Optional lane width values: 4.0m, 4.5m (must be ≥ 2 times the width of the vehicle + 0.5m to ensure safe passing of oncoming vehicles);

[0110] Examples of initial combinations generated: {2.0m, 1.5m, 4.0m}, {2.2m, 1.7m, 4.5m}, etc.

[0111] The generation logic involves combining values ​​within the constraints using either exhaustive search or random sampling to ensure coverage of different extremes (such as combining the minimum parking space width with the maximum vehicle width to form an extreme space scenario).

[0112] Step S33. Eliminate combinations that do not meet the spatial compatibility conditions from the initial combinations to obtain multiple combinations of the scene parameters.

[0113] In this embodiment, spatial compatibility conditions refer to the fact that the parameter combinations do not conflict in three-dimensional space (such as no overlap, no obstacle crossing, etc.), ensuring that the scene is physically feasible, for example:

[0114] No overlap constraint: The initial position of the dynamic participant does not overlap with the coordinates of the fixed obstacle (e.g., if the coordinates of the column are (10m, 5m), then the dynamic participant's x-coordinate ≠ 10m or y-coordinate ≠ 5m, with an allowable error of ±0.3m).

[0115] Path feasibility constraints: The movement trajectory of dynamic participants must not cross walls or pillars (e.g., a straight trajectory from (8m, 6m) to (12m, 6m) must avoid the pillar at (10m, 5m).

[0116] Parking space matching constraints: Vehicle length ≤ parking space length - 0.5m (to prevent the front / rear of the vehicle from exceeding the parking space line).

[0117] Elimination method: Verify each initial combination using a spatial collision detection algorithm (such as coordinate-based distance calculation), for example:

[0118] In a certain initial combination, the "initial position of the motor vehicle (10m, 5m)" overlaps with the "coordinates of the column (10m, 5m)", which is determined to be incompatible and is therefore discarded.

[0119] In a certain combination, "parking space length 5.0m" and "motor vehicle length 5.2m" do not meet the adaptation constraints and are therefore removed.

[0120] Final output: The remaining initial combination after elimination is the "scene parameter combination", such as retaining combinations that meet spatial compatibility, such as {2.0m, 1.5m, 4.0m, illumination 8000 lux, pedestrian speed 1.8m / s}.

[0121] This application effectively solves the combinatorial explosion problem in the parameter combination process by screening related parameter groups, matching values ​​under physical constraints, and eliminating spatially incompatible combinations (reducing irrelevant parameter combinations and reducing computational load), while ensuring that the generated scene parameter combinations conform to physical laws and spatial logic (such as avoiding overlap between dynamic participants and fixed obstacles, and matching parking spaces with vehicle sizes).

[0122] In some optional implementations of this embodiment, the above-described risk quantification of each scenario parameter combination to obtain the risk level of each scenario parameter combination includes:

[0123] Step S41. Extract the core attribute parameters of the dynamic participants for each combination of scene parameters, and assign basic weights to each core attribute parameter;

[0124] In this embodiment, the core attribute parameters of dynamic participants refer to the key features of dynamic elements that directly affect parking risk, specifically including:

[0125] Type parameters (such as pedestrians, non-motorized vehicles, motorized vehicles, traffic cones, etc., with the risk level decreasing in that order);

[0126] Motion state parameters (such as instantaneous velocity, acceleration, and straight-line distance from the test vehicle);

[0127] Special behavioral parameters (such as sudden change of direction, sudden movement after being stationary, etc.).

[0128] The basic weighting rule is as follows: a fixed score is preset based on the risk level of the attribute, and the total score is obtained by accumulating the scores. For example:

[0129] Type weighting: pedestrians 10 points, non-motorized vehicles 8 points, motorized vehicles 5 points, traffic cones 3 points (pedestrians have the highest risk weight due to their high flexibility);

[0130] Distance weighting: 2 points are added for every 0.5m decrease in straight-line distance from the test vehicle (a total of 10 points are added when the distance is 0.5m, and the closer the distance, the higher the risk).

[0131] Speed ​​weighting: 1 point is added for every 1 m / s increase in speed (an additional 5 points are added for speeds > 3 m / s, with increasing risk at high speeds).

[0132] Special behavior weighting: sudden change of direction adds 8 points, sudden movement after standing still adds 6 points (unexpected behavior increases risk).

[0133] Example of calculating the total basic risk weight: In a certain scenario parameter combination, the dynamic participants are "pedestrian (10 points) + distance 1m (4 points) + speed 2m / s (2 points) + sudden change of direction (8 points)", then the total basic risk weight = 10 + 4 + 2 + 8 = 24 points.

[0134] Step S42. Based on the environmental disturbance parameters, adjust each of the basic weights to obtain the initial risk weights;

[0135] Environmental interference parameters refer to environmental factors that affect the perception capabilities of a parking system. These include parameters such as illumination, meteorological parameters, and sensor interference parameters. Specifically, they include: illumination intensity (such as backlight and low light), meteorological conditions (such as rain and fog), and sensor occlusion level (such as the area of ​​the camera that is obstructed).

[0136] The default correction rule uses coefficient multiplication for correction, for example:

[0137] When the light intensity is <500 lux (low light) or >8000 lux (backlight), the base weight is multiplied by 1.2 (increasing the difficulty of visual perception).

[0138] When the weather conditions are rain / fog (visibility < 100m), the base weight is multiplied by 1.3 (sensor accuracy decreases).

[0139] When the sensor occlusion level is ≥2 (occlusion area >30%), the base weight is multiplied by 1.5 (the risk of sensing failure increases).

[0140] When multiple interferences are superimposed, the correction coefficients are multiplied (e.g., backlight + rain / fog: 1.2 × 1.3 = 1.56).

[0141] An example of initial risk weight calculation is as follows: The total basic risk weight is 24 points. If "backlight (1.2) + light fog (1.3)" is added, the initial risk weight = 24 × 1.2 × 1.3 = 37.44 points.

[0142] Step S43. Obtain historical fault data, and adjust each of the initial risk weights based on the historical fault data to obtain the final risk weight value;

[0143] In this embodiment, historical system fault data refers to fault cases recorded during past tests or actual operation of the parking system, for example:

[0144] "The pedestrian recognition latency rate reaches 30% in backlit scenes";

[0145] "When the sensor is obstructed, the distance judgment error of the vehicle is greater than 0.5m";

[0146] "When a non-motorized vehicle enters a narrow parking space (<2.2m), the braking response delay is >0.3s."

[0147] The adjustment rule is as follows: For scenario parameter combinations that match historical failure cases, additional adjustments are made based on the initial risk weights (such as adding a fixed percentage or a fixed score), for example:

[0148] If the scene parameter combination includes "backlight + pedestrian", and matches the historical fault "backlight pedestrian recognition delay", then an additional 20% will be added to the initial risk weight;

[0149] If the data includes "sensor obstruction + motor vehicle", and is matched with "distance judgment error when obstructed", an additional 15% will be added.

[0150] Example of final risk weight calculation: The initial risk weight is 37.44 points. Due to the matching of the "backlight + pedestrian" fault case, an additional 20% adjustment is made. Therefore, the final risk weight value = 37.44 × (1 + 20%) = 44.93 points.

[0151] Step S44. Compare the final risk weight value with the preset risk level to determine the risk level corresponding to each combination of scenario parameters.

[0152] In this embodiment, the preset risk level threshold is based on the risk range divided according to testing requirements, and is typically divided into three levels:

[0153] Low risk level: Final risk weight value < 30 points (the scenario puts little pressure on the system and has a low probability of triggering failure).

[0154] Medium risk level: 30 points ≤ final risk weight value < 60 points (the scenario presents certain challenges, but the system can usually handle them).

[0155] High risk level: ≥60 points (the scenario has extreme characteristics, and the system is prone to triggering extreme response or failure).

[0156] An example of risk level determination is as follows: The final risk weight value is 44.93 points, which is in the range of 30-60 points. Therefore, the risk level of this scenario parameter combination is "medium risk". If the final weight of another combination is 65 points, it is determined to be "high risk".

[0157] This embodiment achieves precise risk level classification through a four-level quantitative process: basic weight accumulation, environmental interference correction, historical fault adjustment, and threshold comparison and grading. This process covers not only the inherent dangerous characteristics of dynamic participants but also the impact of environmental interference and historical system weaknesses, providing quantifiable risk basis for subsequent initial scenario combination screening and matching degree evaluation, making the generation of extreme test cases more targeted.

[0158] In some optional implementations of this embodiment, the dynamic participant parameters mentioned above include initial position and velocity-related parameters, relative distance-related parameters, and type identifier-related parameters;

[0159] Among them, the initial position and velocity-related parameters include the three-dimensional initial coordinates of the dynamic participant (e.g., x=8m, y=5m, θ=0°, where θ is the orientation angle), instantaneous velocity (e.g., v=1.2m / s), and acceleration (e.g., a=0.5m / s²), which are used to determine its motion trajectory and the timing of its appearance;

[0160] Relative distance-related parameters include the real-time straight-line distance between the dynamic participant and the test vehicle (e.g., d=3m), relative orientation (e.g., 45° to the left front of the test vehicle), and distance change rate (e.g., shortening by 0.8m per second), which are used to correlate the parking phase with the danger zone.

[0161] The type identifier parameters include the physical type of the dynamic participant (e.g., pedestrian, electric bicycle, car, traffic cone) and behavior priority (e.g., pedestrian > non-motorized vehicle > motorized vehicle), which are used to distinguish the interaction order.

[0162] The step of setting time constraints for dynamic participants based on the dynamic participant parameters includes:

[0163] S61. Based on the initial position and speed-related parameters, set a timing constraint that the dynamic participant must appear within a preset time after the test vehicle starts parking mode;

[0164] In this embodiment, the timing constraint refers to the time window for dynamic participants to enter the test scenario. It is necessary to calculate the reachable time based on their initial position and speed to ensure that they appear at key nodes in the parking process. Based on this, the logic for setting the timing constraint is as follows:

[0165] 1. Calculate the time required for a dynamic participant to reach a preset area around the test vehicle (e.g., a radius of 5m centered on the test vehicle): Time t = initial straight-line distance d ÷ velocity v (simplified calculation ignoring acceleration).

[0166] 2. The moment when the test vehicle starts parking mode is recorded as t0. Dynamic participants are required to enter the preset area within the time window from t0+Δt1 to t0+Δt2 (Δt1 and Δt2 are preset thresholds).

[0167] An example of this embodiment is as follows:

[0168] The dynamic participants are "pedestrians, initial positions (x=15m, y=5m), speeds v=1.5m / s", and the test vehicle's initial position (x=10m, y=5m) and initial straight-line distance d=5m.

[0169] The arrival time is calculated as t = 5m ÷ 1.5m / s ≈ 3.3s;

[0170] Set timing constraints: After the test vehicle starts parking (t0), the pedestrian must enter the 5m range within t0+2s to t0+4s (covering the calculation time to ensure the reasonableness of the timing).

[0171] S62. Based on the relative distance related parameters, the parking stage is divided into parking space search, parking, and correction stages, and stage constraints are set for dynamic participants to enter the preset range of the test vehicle during the parking stage.

[0172] In this embodiment, the parking phase is divided as follows: Based on the typical process of an automatic parking system, the entire process is divided into:

[0173] Parking space search phase (t0 to t0+10s): Test vehicle drives at low speed to identify available parking spaces;

[0174] Parking phase (t0+10s to t0+18s): The vehicle begins to turn and move into the parking space, which is the most complex phase of the operation.

[0175] Correction phase (t0+18s to t0+25s): Fine-tune the vehicle position to ensure it is parked straight.

[0176] Phase constraints refer to the requirement that dynamic participants must enter the danger zone at a specific phase (defined based on relative distance parameters) to amplify the extreme nature of the test scenario. The logic is as follows:

[0177] 1. Based on relative distance-related parameters, set a threshold for the danger range (e.g., ≤2m from the test vehicle).

[0178] 2. Dynamic participants must enter the area during the parking phase (rather than the search or correction phase), as the system is more prone to failure due to sudden disturbances during the parking phase due to frequent vehicle turning.

[0179] The following is an example of this embodiment:

[0180] The dynamic participant was an electric bicycle with a relative distance change rate of 0.8 m / s, and the test vehicle entered the parking phase at t0+10s.

[0181] Setting phase constraints: The electric bicycle must shorten the distance from the initial 5m to ≤2m (i.e., enter the danger zone) within t0+10s to t0+18s, and must continue for at least 3s within this phase.

[0182] S63. Based on the type identifier-related parameters, set the order constraints for multiple types of dynamic participants to enter the preset range of the test vehicle.

[0183] In this embodiment, the order constraint refers to setting the order of appearance or interaction according to the type priority when multiple dynamic participants of different types exist simultaneously, simulating the logic of multi-element interference in a real scenario. The logic is set as follows:

[0184] 1. Sort the relevant parameters by type according to their priority (e.g., pedestrians > electric bicycles > cars > traffic cones).

[0185] 2. Constrain high-priority participants to enter the scene or trigger interactions first, and low-priority participants to enter later, to avoid scene logic confusion caused by disordered appearance.

[0186] The following is an example of this embodiment:

[0187] The dynamic participants of various types include "pedestrian A, electric bicycle B, and small car C", with the type priority as follows: pedestrian A > electric bicycle B > small car C;

[0188] Set sequence constraints: Pedestrian A enters the danger zone first at t0+3s, electric bicycle B enters at t0+5s (2s after pedestrian A appears), and car C enters at t0+8s (3s after electric bicycle B appears), and the movement trajectories of the three must have potential intersections (such as all moving towards the right front of the test vehicle).

[0189] This embodiment deeply correlates dynamic participant parameters with the time dimension of the parking process through a three-level temporal logic: timing constraints (when it occurs), stage constraints (when it enters the danger zone), and sequence constraints (how multiple elements interact). Initial position and speed parameters ensure the physical feasibility of the timing, relative distance parameters strengthen the matching of stages and danger zones, and type identification parameters guarantee the realistic rationality of multi-element interactions. Together, these three constitute a complete time constraint system, providing a clear evaluation standard for quantifying the degree of time constraint compliance in subsequent steps.

[0190] In some optional implementations of this embodiment, the above-described method of determining the matching degree of each initial scenario combination and extreme test case target requirement based on the risk level and the time constraint, and using the initial scenario combination with the highest matching degree as the extreme test case of the test vehicle, includes:

[0191] S71. Quantify the risk level to obtain a risk value, and quantify the degree to which the initial scenario combination satisfies the time constraint to obtain a time value;

[0192] In this embodiment, the risk quantification rule is as follows: based on the final risk weight value, the risk level is converted into a numerical value of 0-100 (the higher the score, the higher the risk).

[0193] If the final risk weight value is ≥60 points (high risk level), the corresponding risk value is 80-100 points (linearly mapped according to the weight value, such as 60 points → 80 points, 100 points → 100 points).

[0194] If 30 points ≤ final risk weight value < 60 points (medium risk level), the corresponding risk value = 40-79 points (e.g., 45 points → 60 points).

[0195] If the final risk weight value is less than 30 points (low risk level), the corresponding risk value is 0-39 points (e.g., 20 points → 20 points).

[0196] The following is an example of this embodiment: The final risk weight value of a certain initial scenario combination is 75 points (high risk), then the risk value = 80 + (75-60) × (20 / 40) = 87.5 points (rounded to 88 points).

[0197] Time-based numerical quantification rules: Based on the degree to which time constraints are met, the compliance score is converted into a numerical value of 0-100 (the higher the score, the better the compliance with temporal logic):

[0198] All constraints are fully met (timing, stage, and sequence all meet the preset requirements): 100 points;

[0199] Partially satisfied (e.g., timing deviation ±1s, stage advance / delay by 1 stage, sequence reversal of 1 item): points are deducted according to the deviation ratio (e.g., 20 points are deducted for timing deviation of 1s, and 30 points are deducted for stage deviation).

[0200] Completely unsatisfactory (e.g., not appearing at the preset time, not entering the specified stage): 0 points.

[0201] The following is an example of this embodiment: A certain initial scenario combination satisfies the "timing constraint (deviation 0.5s)" and "stage constraint", but the low-priority participant appears first in the "sequence constraint" (deviation 1 item), then the time value = 100 - 10 (timing deviation) - 20 (sequence deviation) = 70 minutes.

[0202] S72. According to the preset weight ratio corresponding to the risk level and the preset weight ratio corresponding to the degree of satisfaction of the time constraint, perform a weighted summation on the risk value and time value corresponding to the initial scenario combination to obtain the matching degree of the initial scenario combination;

[0203] In this embodiment, the preset weight ratio is set based on the impact of risk level and time constraint satisfaction on extreme scenarios, and is typically as follows:

[0204] Preset weighting percentage for risk level: 60% (high-risk scenarios should be prioritized in extreme cases);

[0205] The preset weight percentage corresponding to the degree of satisfaction of time constraints: 40% (the timing must be reasonable to avoid invalid scenarios).

[0206] The matching degree is calculated as follows: Matching degree = Risk value × 60% + Time value × 40%.

[0207] For example: If the risk score is 88 and the time score is 70, then the matching degree = 88 × 60% + 70 × 40% = 52.8 + 28 = 80.8 points (rounded to 81 points).

[0208] S73. Sort the matching degree of all the initial scenario combinations and select the initial scenario combination with the highest matching degree as the extreme test case.

[0209] In this embodiment, the matching degree of all initial scene combinations is sorted in descending order (from high to low), and the combination with the highest score is selected first.

[0210] The screening logic is as follows: if multiple combinations have the same matching degree (e.g., all of them are 90 points), then the risk values ​​are compared (higher risk is given priority); if the risk values ​​are also the same, the time values ​​are compared (more reasonable time sequence is given priority).

[0211] Here is an example: Suppose there are 3 initial scenario combinations with matching scores of 81, 75 and 92 respectively. Select the combination with a score of 92 as the extreme test case according to the ranking results (this combination may be "high-risk value 95 + time value 90", that is, high risk and completely reasonable timing).

[0212] This embodiment achieves accurate generation of extreme test cases through a process of dual-dimensional quantification (risk + timing), weighted comprehensive evaluation, and optimal sorting and selection. The quantification process integrates risk quantification results with time constraints, ensuring data consistency. The weighting balances the core requirements of high risk and timing rationality, conforming to the design concept of dual-dimensional evaluation. The sorting rules guarantee the uniqueness and validity of the selection results, ensuring that the final generated extreme test cases both trigger the system's extreme response and possess logical rationality in real-world scenarios.

[0213] In some optional implementations of this embodiment, the static environmental parameters mentioned above include parking space geometric parameters, fixed obstacle parameters, and parking lot topology parameters;

[0214] In this embodiment, parking space geometric parameters refer to the physical dimensions and shape characteristics of the parking space, used to define the extremes of the target parking area (such as narrow parking spaces), including:

[0215] Length L: ranges from 5.0 to 6.0m (e.g., 5.0m for small parking spaces, 5.5m for standard parking spaces);

[0216] Width W: Value range 1.8-2.8m (in extreme scenarios, it can be as low as 1.8m to simulate narrow parking spaces).

[0217] Inclination angle θ: The angle between the parking space and the passage, with values ​​of 0° (perpendicular parking space), 45° (angled parking space), and 90° (parallel parking space).

[0218] Pavement wear level M: 0-5 (0 is complete, 5 is blurry and unclear, affecting parking space recognition).

[0219] Fixed obstacle parameters refer to the characteristics of immovable objects within a parking lot, used to determine the feasibility (spatial compatibility) of dynamic participant trajectories, including:

[0220] Type T1: Columns, walls, floor locks, fire hydrants, etc.;

[0221] Three-dimensional coordinates (x, y, θ): absolute coordinates with the parking lot entrance as the origin, where θ is the orientation of the obstacle (e.g., column coordinates (10m, 5m, 0°)).

[0222] Dimensions (length × width × height): For example, column 0.5m × 0.5m × 2.5m, wall thickness 0.2m.

[0223] Parking lot topology parameters refer to the overall layout characteristics of the parking lot, which affect the driving path of the test vehicle, including:

[0224] Lane width C: Main lane 4-6m, auxiliary lane 3-4m (narrow lanes increase the risk of vehicles passing each other).

[0225] Turning radius R: 5-8m (small radius turning simulates control difficulty);

[0226] Entrance / exit location: coordinates and width (e.g., entrance (0m, 0m), width 3m).

[0227] The dynamic participant parameters include type identification parameters, initial position and velocity parameters, and relative distance parameters;

[0228] In this embodiment, the type identifier-related parameters are used to distinguish the physical attributes and behavioral priorities of dynamic participants, including:

[0229] Physical type T2: Pedestrians, non-motorized vehicles (bicycles, electric scooters), motorized vehicles (cars, SUVs), traffic cones, etc.

[0230] Behavioral Pattern B: Normal movement (uniform linear motion), sudden change of direction (such as lateral crossing), sudden movement after coming to a standstill;

[0231] Interaction priority: pedestrians > non-motorized vehicles > motorized vehicles > traffic cones (participants with higher priority trigger constraints first).

[0232] Initial position and velocity-related parameters are used to calculate the timing of the dynamic participant's appearance (timing constraints), including:

[0233] Initial 3D coordinates (x0, y0, θ0): e.g., the initial position of a pedestrian (8m, 6m, 0°), where θ0 is the initial direction of motion;

[0234] Instantaneous speed v: pedestrians 0-5m / s (up to 5m / s in extreme scenarios, simulating running), motor vehicles 0-10km / h (parking lot speed limit);

[0235] Acceleration a: -2 to 2 m / s² (negative values ​​indicate deceleration, such as sudden braking; positive values ​​indicate acceleration, such as sudden acceleration).

[0236] Relative distance parameters are used to correlate parking phases with hazardous areas (phase constraints), including:

[0237] Initial straight-line distance d0: Initial distance to the test vehicle, 0-10m (d0 < 2m is a high-risk initial distance).

[0238] Relative orientation D: Directly in front of, to the side (left / right), and behind the test vehicle (e.g., a pedestrian is located 45° to the left front of the test vehicle).

[0239] Distance change rate Δd / Δt: calculated based on the velocity vectors of both (e.g., moving closer by 0.8m per second, rapidly reducing the distance).

[0240] The environmental interference parameters include illumination parameters, meteorological parameters, and sensor interference parameters.

[0241] In this embodiment, the illumination parameters include:

[0242] Illumination Intensity I: 0-10000 lux (0-500 lux is weak light, 5000-10000 lux is strong light, and above 8000 lux simulates backlighting);

[0243] Light source orientation: the angle between the light source and the test vehicle (e.g., 180° is direct backlighting, 90° is sidelighting).

[0244] Meteorological parameters include:

[0245] Weather type Wt: Sunny, Rain (light / moderate / heavy rain), Fog (visibility 20-100m), Snow;

[0246] Visibility V: The visible distance in rainy or foggy weather (e.g., V=20m in heavy fog, which significantly reduces the range of perception).

[0247] Sensor interference parameters include:

[0248] Obstructed areas: Areas where camera lenses, radar sensors, etc. are obstructed (e.g., the upper left 10% area of ​​the front camera is obscured by dirt).

[0249] Occlusion level S: 0-3 (Level 0: no occlusion; Level 3: complete occlusion).

[0250] Noise intensity N: The fluctuation range of sensor data (e.g., radar ranging error ±0.5m, which increases the difficulty of distance judgment).

[0251] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0252] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0253] Further reference Figure 3 As a response to the above Figure 2 The implementation of the method shown in this application provides an embodiment of an automatic parking lot extreme test case generation system. This device embodiment is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0254] like Figure 3 As shown, the parking lot automatic parking extreme test case generation system 300 described in this embodiment includes: a data acquisition module 301, a parameterization processing module 302, a parameter combination module 303, a risk level calculation module 304, an initial combination screening module 305, a time constraint setting module 306, and a matching degree evaluation and test case generation module 307. Wherein:

[0255] The data acquisition module 301 is used to acquire static environmental data, dynamic participant data, and environmental disturbance data of the parking lot;

[0256] The parameterization processing module 302 is used to define the static environment data, the dynamic participant data and the environmental interference data in a parameterized manner to obtain static environment parameters, dynamic participant parameters and environmental interference parameters.

[0257] The parameter combination module 303 is used to combine the static environment parameters, the dynamic participant parameters, and the environmental interference parameters to obtain multiple scene parameter combinations;

[0258] The risk level calculation module 304 is used to quantify the risk of each scenario parameter combination to obtain the risk level of each scenario parameter combination.

[0259] The initial combination filtering module 305 is used to select an initial scene combination from multiple scene parameter combinations according to a preset parameter range;

[0260] The time constraint setting module 306 is used to set time constraint conditions for dynamic participants based on the dynamic participant parameters.

[0261] The matching degree evaluation and test case generation module 307 is used to determine the matching degree of each of the initial scenario combinations based on the risk level and the time constraint, and to take the initial scenario combination with the highest matching degree as the extreme test case.

[0262] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0263] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0264] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0265] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for generating extreme test cases for automatic parking in parking lots. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0266] In some embodiments, the processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the automatic parking test case generation method.

[0267] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0268] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the above-described method for generating extreme test cases for automatic parking in parking lots.

[0269] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0270] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A method for generating test cases for automated parking lot extreme testing, the method comprising: receiving a plurality of parking lot parameters; generating a plurality of test cases based on the plurality of parking lot parameters; and outputting the plurality of test cases. The method comprises the following steps: obtaining static environment data, dynamic participant data and environmental interference data of a parking lot corresponding to a test vehicle, wherein the dynamic participant data comprises real-time positions of moving objects in the parking lot over time, moving speeds of the moving objects in the parking lot over time, moving directions of the moving objects in the parking lot over time, and types of the moving objects in the parking lot over time, and the moving objects in the parking lot over time comprise pedestrians, non-motor vehicles and motor vehicles; parameterizing the static environment data, the dynamic participant data and the environmental interference data to obtain static environment parameters, dynamic participant parameters and environmental interference parameters, wherein the parameterization refers to converting original data into quantifiable and calculable parameter forms, so that the data have standardized value ranges and formats; combining the static environment parameters, the dynamic participant parameters and the environmental interference parameters to obtain a plurality of scene parameter combinations, wherein the scene parameter combination refers to selecting specific parameter values from the static environment parameters, the dynamic participant parameters and the environmental interference parameters to form a complete scene description; risk quantification is performed on each scene parameter combination to obtain a risk level of each scene parameter combination; at least two scene parameter combinations are selected from the plurality of scene parameter combinations as initial scene combinations according to a preset parameter range; a time constraint condition of a dynamic participant is determined based on the dynamic participant parameters; a matching degree of each initial scene combination and an extreme test case target requirement is determined based on the risk level and the time constraint condition, and an initial scene combination with the highest matching degree is selected as an extreme test case of the test vehicle; wherein the risk quantification of each scene parameter combination to obtain a risk level of each scene parameter combination comprises: extracting core attribute parameters of a dynamic participant for each scene parameter combination, and assigning a basic weight to each core attribute parameter; correcting each basic weight based on the environmental interference parameters to obtain an initial risk weight; obtaining historical failure data, adjusting each initial risk weight based on the historical failure data to obtain a final risk weight value; comparing the final risk weight value with a preset risk level to determine a risk level corresponding to each scene parameter combination.

2. The automated parking lot extreme test case generation method of claim 1, wherein, The combination of the static environment parameters, the dynamic participant parameters and the environmental interference parameters to obtain a plurality of scene parameter combinations comprises: screening associated parameter groups having physical association based on the static environment parameters, the dynamic participant parameters and the environmental interference parameters; value matching of the associated parameter groups is performed within a range satisfying a preset physical constraint condition to generate an initial combination; combinations not satisfying a spatial compatibility condition are eliminated from the initial combination to obtain a plurality of scene parameter combinations.

3. The test case generation method of claim 1, wherein, The dynamic participant parameters comprise initial position and speed related parameters, relative distance related parameters and type identification related parameters. The time constraint condition of the dynamic participant is set based on the dynamic participant parameter, including: An opportunity constraint is set that the dynamic participant needs to appear within a preset time after the test vehicle starts the parking mode based on the initial position and the speed-related parameter; Based on the relative distance-related parameter, the parking stage is divided into a parking space search, parking-in, and correction stage, and a stage constraint is set that the dynamic participant enters the preset range of the test vehicle in the parking-in stage; Based on the type identification-related parameter, a sequence constraint is set that multiple types of dynamic participants enter the preset range of the test vehicle.

4. The test case generation method of claim 1, wherein, The matching degree of each initial scene combination and the extreme test case target requirement is determined based on the risk level and the time constraint condition, and the initial scene combination with the highest matching degree is taken as the extreme test case of the test vehicle, including: The risk value is obtained by quantifying the risk level, and the time value is obtained by quantifying the satisfaction degree of the initial scene combination to the time constraint condition; The risk value and the time value corresponding to the initial scene combination are weighted and summed according to the preset weight proportion corresponding to the risk level and the preset weight proportion corresponding to the satisfaction degree of the time constraint condition, to obtain the matching degree of the initial scene combination; The matching degrees of all the initial scene combinations are sorted, and the initial scene combination with the highest matching degree is selected as the extreme test case.

5. The test case generation method of claim 1, wherein, The static environment parameter includes a parking space geometric parameter, a fixed obstacle parameter, and a parking lot topology parameter; The environmental interference parameter includes an illumination parameter, a meteorological parameter, and a sensor interference parameter.

6. A parking lot automatic parking extreme test case generation system for performing the parking lot automatic parking extreme test case generation method according to any one of claims 1 to 5, characterized by Including: A data acquisition module is configured to acquire static environment data, dynamic participant data, and environmental interference data of a parking lot corresponding to a test vehicle, wherein the dynamic participant data includes real-time positions of time-varying objects in the parking lot, movement speeds of the time-varying objects in the parking lot, movement directions of the time-varying objects in the parking lot, and types of the time-varying objects in the parking lot, and the time-varying objects in the parking lot include pedestrians, non-motor vehicles, and motor vehicles; A parameterization processing module is configured to perform parameterization definition on the static environment data, the dynamic participant data, and the environmental interference data to obtain static environment parameters, dynamic participant parameters, and environmental interference parameters, wherein the parameterization definition refers to converting original data into quantifiable and calculable parameter forms, so that the data has a standardized value range and format; A parameter combination module is configured to combine the static environment parameters, the dynamic participant parameters, and the environmental interference parameters to obtain multiple scene parameter combinations, wherein the scene parameter combination refers to selecting specific parameter values from the static environment parameters, the dynamic participant parameters, and the environmental interference parameters to form a complete scene description; A risk level calculation module is configured to quantize the risk of each scene parameter combination to obtain a risk level of each scene parameter combination. An initial combination screening module is configured to select at least two scene parameter combinations from a plurality of scene parameter combinations as initial scene combinations according to a preset parameter range; A time constraint setting module is configured to determine a time constraint condition of a dynamic participant based on the dynamic participant parameter; A matching degree evaluation and use case generation module is configured to determine a matching degree of each of the initial scene combinations and an extreme test use case target requirement based on the risk level and the time constraint condition, and select the initial scene combination with the highest matching degree as the extreme test use case of the test vehicle. The risk quantification of each of the scene parameter combinations to obtain the risk level of each of the scene parameter combinations comprises: Extracting a core attribute parameter of a dynamic participant for each of the scene parameter combinations, and assigning a basic weight to each of the core attribute parameters; Correcting each of the basic weights based on the environmental interference parameter to obtain an initial risk weight; Obtaining historical failure data, and adjusting each of the initial risk weights based on the historical failure data to obtain a final risk weight value; Comparing the final risk weight value with a preset risk level to determine the risk level corresponding to each of the scene parameter combinations.

7. A computer device, comprising: A memory and a processor are included, the memory stores computer readable instructions, and the processor implements the steps of the parking lot automatic parking extreme test use case generation method according to any one of claims 1 to 5 when executing the computer readable instructions.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by the processor to implement the steps of the parking lot automatic parking extreme test use case generation method according to any one of claims 1 to 5.

Images