A method, apparatus, device, and medium for adjusting a test time window.

Abstract

This invention discloses a method, apparatus, device, and medium for adjusting a test time window. The method includes: responding to a received test request; acquiring a decision component under test (DUT), a prior decision component, and an initial test scenario to obtain a test data basis; running the DUT and prior decision components in the initial test scenario to obtain DUT decision information and prior decision information corresponding to each decision component within the initial test time window; further comparing the DUT decision information and prior decision information to calculate the decision difference between them in each frame within the initial test time window; and finally adjusting the initial test time window according to the decision difference to generate a target test time window. This provides a more accurate length of the test time window corresponding to the autonomous driving scenario, thereby improving the accuracy of decision simulation.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving testing technology, and in particular to a method, apparatus, device, and medium for adjusting a test time window. Background Technology

[0002] Intelligentization is one of the key trends in the automotive industry today. With the continuous development of autonomous driving technology, the level of intelligence in autonomous vehicles is gradually improving. However, because autonomous driving systems need to handle highly complex and varied scenarios, large-scale road testing is often required during algorithm iteration.

[0003] In practice, actual road testing is costly and the probability of accidents during such testing is extremely low. However, for autonomous driving systems, it is necessary to consider the potential dangers under different circumstances. Therefore, test time windows are typically set based on basic data to generate autonomous driving test scenarios, on which different versions of the decision-making system are tested.

[0004] However, the start and end times of the current test scenarios are relatively fixed and will not be changed after generation. As the autonomous driving decision-making version changes, the previously generated autonomous driving scenarios may no longer be able to accurately measure the problems of the corresponding version, leading to a decrease in the accuracy of the simulation results. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for adjusting test time windows, solving the problem that the start and end times of current test scenarios are relatively fixed and cannot be changed after generation. Furthermore, as autonomous driving decision-making versions change, the previously generated autonomous driving scenarios may no longer accurately measure the corresponding versions, leading to a decrease in the accuracy of simulation results.

[0006] The first aspect of this invention provides a method for adjusting a test time window, comprising:

[0007] In response to the received test request, the system obtains the decision component to be tested, the prior decision component, and the initial test scenario; the initial test scenario is provided with an initial test time window.

[0008] The test decision component and the prior decision component are run in the initial test scenario to generate test decision information and prior decision information;

[0009] By comparing the decision information to be tested with the prior decision information, the decision difference of the decision information to be tested within the initial test time window is determined;

[0010] The initial test time window is adjusted based on the decision differences to generate the target test time window.

[0011] Optionally, the decision information to be tested includes multiple trajectory points to be tested, and the prior decision information includes multiple prior trajectory points; the step of comparing the decision information to be tested and the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window includes:

[0012] The Euclidean distance between the trajectory point to be tested and the prior trajectory point is calculated one by one in each frame of the initial test time window to obtain multiple Euclidean distance values;

[0013] Calculate the average value among all the Euclidean distance values ​​to determine the decision difference of the decision information to be tested within the initial test time window.

[0014] Optionally, the decision information to be tested includes speed control information to be tested, and the prior decision information includes prior speed control information; the step of comparing the decision information to be tested and the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window includes:

[0015] Extract the first velocity vector and the first acceleration vector of each frame within the initial test time window of the speed control information to be tested;

[0016] Extract the second velocity vector and the second acceleration vector of each frame within the initial test time window from the prior velocity control information;

[0017] Calculate the velocity difference between the first velocity vector and the second velocity vector frame by frame, or calculate the acceleration difference between the first acceleration vector and the second acceleration vector frame by frame;

[0018] The average value of the velocity difference or the average value of the acceleration difference is determined as the decision difference of the decision information to be tested within the initial test time window.

[0019] Optionally, the step of adjusting the initial test time window based on the decision difference to generate the target test time window includes:

[0020] Select target trajectory points from a plurality of trajectory points to be tested, wherein the decision difference is greater than or equal to a preset difference threshold and the spacing is less than a preset distance threshold;

[0021] Determine whether the number of the target trajectory points is greater than or equal to a preset number threshold;

[0022] If so, then select the target time frame number corresponding to the target trajectory point from the initial test time window;

[0023] The initial test time window is extracted according to the target time frame number, and a target test time window with a size greater than or equal to the first preset window size is generated.

[0024] Optionally, the step of adjusting the initial test time window based on the decision difference to generate the target test time window includes:

[0025] Select target trajectory points from a plurality of trajectory points to be tested, wherein the decision difference is greater than or equal to a preset difference threshold and the spacing is less than a preset distance threshold;

[0026] Determine whether the number of the target trajectory points is greater than or equal to a preset number threshold;

[0027] If so, then select the target time frame number corresponding to the target trajectory point from the initial test time window;

[0028] The initial test time window is extracted according to the target time frame number to obtain multiple intermediate test time windows;

[0029] The intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0030] Optionally, the step of merging the intermediate test time windows according to the second preset window size to obtain multiple target test time windows includes:

[0031] Calculate the ratio of the number of target trajectory points to the number of trajectory points to be tested within the intermediate test time window;

[0032] If the proportion of the trajectory points is greater than or equal to a preset proportion threshold, then the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0033] Optionally, the step of merging the intermediate test time windows according to the second preset window size to obtain multiple target test time windows includes:

[0034] Calculate the average difference value of the target trajectory points within the intermediate test time window;

[0035] If the average difference value is greater than or equal to the preset average difference threshold, the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0036] Optionally, the method further includes:

[0037] A buffer time interval is added on both sides of the target test time window to obtain an updated test time window;

[0038] The initial test scenario is captured according to the updated test time window to obtain a new initial test scenario;

[0039] Jump to execute the step of running the decision component under test and the prior decision component in the initial test scenario to generate decision information under test and prior decision information.

[0040] A second aspect of the present invention provides a device for adjusting a test time window, comprising:

[0041] The data acquisition module is used to respond to received test requests and acquire the decision component to be tested, the prior decision component, and the initial test scenario; the initial test scenario is provided with an initial test time window;

[0042] The decision execution module is used to run the decision component under test and the prior decision component in the initial test scenario, and generate the decision information under test and the prior decision information.

[0043] The decision difference calculation module is used to compare the decision information to be tested with the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window.

[0044] The time window adjustment module is used to adjust the initial test time window based on the decision differences and generate the target test time window.

[0045] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the test time window adjustment method as described in any one of the first aspects of the present invention.

[0046] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the method for adjusting the test time window as described in any of the first aspects of the present invention.

[0047] As can be seen from the above technical solutions, the present invention has the following advantages:

[0048] This invention, in response to a received test request, acquires the decision component under test, the prior decision component, and an initial test scenario to obtain the test data foundation. The decision component under test and the prior decision component are then run within the initial test scenario to obtain the decision information under test and the prior decision information corresponding to each decision component within the initial test time window. Furthermore, the decision information under test and the prior decision information are compared to calculate the decision difference between them in each frame within the initial test time window. Finally, the initial test time window is adjusted according to the decision difference to generate a target test time window. This provides a more accurate estimate of the length of the test time window corresponding to the autonomous driving scenario, thereby improving the accuracy of decision simulation. Attached Figure Description

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

[0050] Figure 1 A flowchart illustrating the steps of a method for adjusting a test time window according to Embodiment 1 of the present invention;

[0051] Figure 2 A flowchart illustrating the steps of a method for adjusting a test time window according to Embodiment 2 of the present invention;

[0052] Figure 3 This is a structural block diagram of a test time window adjustment device provided in Embodiment 3 of the present invention. Detailed Implementation

[0053] This invention provides a method, apparatus, device, and medium for adjusting test time windows to address the current problem that the start and end times of test scenarios are relatively fixed and remain unchanged after generation. However, as autonomous driving decision-making versions change, the previously generated autonomous driving scenarios may no longer accurately measure the corresponding version's issues, leading to reduced accuracy in simulation results.

[0054] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0055] Please see Figure 1 , Figure 1 This is a flowchart illustrating the steps of a test time window adjustment method provided in Embodiment 1 of the present invention.

[0056] The present invention provides a method for adjusting a test time window, comprising:

[0057] Step 101: Respond to the received test request and obtain the decision component to be tested, the prior decision component, and the initial test scenario; the initial test scenario has an initial test time window;

[0058] A test request refers to the request information inputted to verify the decision-making of autonomous vehicles in various driving scenarios.

[0059] The decision component under test refers to the current version of the decision component, which is a module component that performs road decisions in response to various data recorded by the autonomous vehicle during driving, such as information on surrounding obstacles, traffic lights, or its own state. The prior decision component refers to the earlier version of the decision component.

[0060] The initial test scenario refers to the test scenario constructed based on the acquired road test data, which is used to simulate the decisions made by the decision-making component under test and the prior decision-making component in the test scenario.

[0061] The initial test time window refers to the time window divided in the initial test scenario based on the start and end points when conducting autonomous driving decision-making tests on the decision-making component under test or the prior decision-making component.

[0062] In this embodiment of the application, by responding to the received test request, the decision component to be tested, the prior decision component, and the initial test scenario corresponding to the test request are obtained to provide the data foundation for subsequent decision testing.

[0063] The initial test scenario includes an initial test time window. By acquiring road test data of the prior decision components during the actual driving process of autonomous vehicles, the initial test scenario is constructed and the initial test time windows under various scenarios are divided for the simulation test of the decision components to be tested before they are put into operation.

[0064] Step 102: Run the decision component to be tested and the prior decision component in the initial test scenario to generate the decision information to be tested and the prior decision information;

[0065] After obtaining the initial test scenario, the decision component to be tested, and the prior decision components, the decision component to be tested can be simulated and run within the initial test scenario to determine the decisions made by the decision component to be tested under various conditions within the initial test time window, thereby obtaining the decision information to be tested. The prior decision components are also run within the initial test scenario to generate corresponding prior decision information.

[0066] Step 103: Compare the decision information to be tested with the prior decision information to determine the decision differences of the decision information to be tested within the initial test time window;

[0067] In this embodiment, after obtaining the decision information to be tested and the prior decision information, the decision information to be tested and the prior decision information are compared frame by frame for each frame within the initial test time window to determine the decision difference between the decision information to be tested and the prior decision information within the initial test time window.

[0068] It should be noted that decision difference refers to the difference in trajectory point distance, speed, acceleration, or orientation between the decision information to be tested and the prior decision information, and is used to measure the impact of the decision information to be tested on the autonomous vehicle.

[0069] Step 104: Adjust the initial test time window based on the decision differences to generate the target test time window.

[0070] Once the decision difference is calculated, the number of frames corresponding to the test trajectory points in the test decision information can be adjusted according to the decision difference, thereby readjusting the length of the initial test time window and realizing the readjustment of the test time window to more accurately simulate the decision of subsequent autonomous vehicles.

[0071] In this embodiment, by responding to a received test request, the test decision component, prior decision component, and initial test scenario are obtained to form the test data basis. The test decision component and prior decision component are run in the initial test scenario to obtain the test decision information and prior decision information corresponding to each decision component within the initial test time window. The test decision information and prior decision information are further compared to calculate the decision difference between them in each frame within the initial test time window. Finally, the initial test time window is adjusted according to the decision difference to generate the target test time window. This provides a more accurate length of the test time window corresponding to the autonomous driving scenario, thereby improving the accuracy of decision simulation.

[0072] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps of a test time window adjustment method provided in Embodiment 2 of the present invention.

[0073] The present invention provides a method for adjusting a test time window, comprising:

[0074] Step 201: Respond to the received test request and obtain the decision component to be tested, the prior decision component, and the initial test scenario; the initial test scenario has an initial test time window;

[0075] Step 202: Run the decision component to be tested and the prior decision component in the initial test scenario to generate the decision information to be tested and the prior decision information;

[0076] In the embodiments of this application, the specific implementation process of steps 201-202 is similar to that of steps 101-102, and will not be repeated here.

[0077] Step 203: Compare the decision information to be tested with the prior decision information to determine the decision differences of the decision information to be tested within the initial test time window;

[0078] Optionally, the decision information to be tested includes multiple trajectory points to be tested, and the prior decision information includes multiple prior trajectory points; step 203 may include the following sub-steps:

[0079] Calculate the Euclidean distance between the trajectory point to be tested and the previous trajectory point in each frame of the initial test time window to obtain multiple Euclidean distance values;

[0080] Calculate the average value among all Euclidean distance values ​​to determine the decision difference of the decision information to be tested within the initial test time window.

[0081] The trajectory points to be tested refer to the decision sequence generated by the decision component under test running the simulation in the initial test scenario, such as the positions reached by the vehicle at various time points on the driving route in the initial test scenario.

[0082] In this embodiment, multiple Euclidean distance values ​​can be obtained by calculating the Euclidean distance between the trajectory point to be tested and the previous trajectory point in each frame within the initial test time window; the average value among all Euclidean distance values ​​can then be calculated to determine the decision difference of the decision information to be tested within the initial test time window.

[0083] It should be noted that for the decision information to be tested and the prior decision information, the vehicle speed in the initial test scenario can be further obtained. If the vehicle speed is greater than or equal to a certain speed threshold, the decision difference at the first preset driving distance can be considered. If the vehicle speed is less than the speed threshold, the decision difference at the second preset driving distance can be considered. The first preset driving distance threshold can be less than the second preset driving distance.

[0084] In another example of this application, the decision information to be tested includes the speed control information to be tested, and the prior decision information includes the prior speed control information; step 203 may include the following sub-steps:

[0085] Extract the first velocity vector and the first acceleration vector of each frame within the initial test time window from the velocity control information to be tested.

[0086] Extract the second velocity vector and second acceleration vector of each frame within the initial test time window from the prior velocity control information;

[0087] Calculate the velocity difference between the first velocity vector and the second velocity vector frame by frame, or calculate the acceleration difference between the first acceleration vector and the second acceleration vector frame by frame;

[0088] The average value of the velocity difference or the average value of the acceleration difference is determined as the decision difference of the decision information to be tested within the initial test time window.

[0089] The decision information to be measured includes the speed control information to be measured, which may include, but is not limited to, a first velocity vector, a first acceleration vector, or a first vehicle orientation. Prior decision information includes prior speed control information, including but not limited to a second velocity vector, a second acceleration vector, or a second vehicle orientation.

[0090] In this embodiment, the first velocity vector and the first acceleration vector of the speed control information to be tested are extracted for each frame within the initial test time window. Simultaneously, the second velocity vector and the second acceleration vector of the prior speed control information are extracted for each frame within the initial test time window. The velocity difference between the first velocity vector and the second velocity vector are calculated frame-by-frame, or the acceleration difference between the first acceleration vector and the second acceleration vector are calculated frame-by-frame. The average of either value is determined as the decision difference of the decision information to be tested within the initial test time window.

[0091] It should be noted that the speed difference and acceleration difference include not only the numerical difference, but also the angular difference in the vehicle's direction.

[0092] Step 204: Adjust the initial test time window based on the decision differences to generate the target test time window.

[0093] Optionally, step 204 may include the following sub-steps S11-S14:

[0094] S11, Select target trajectory points from multiple trajectory points to be tested. The decision difference is greater than or equal to the preset difference threshold, and the distance is less than the preset distance threshold.

[0095] S12, determine whether the number of target trajectory points is greater than or equal to a preset number threshold;

[0096] S13, if so, select the target time frame corresponding to the target trajectory point from the initial test time window;

[0097] S14, extract the initial test time window according to the target time frame number, and generate a target test time window that is greater than or equal to the first preset window size.

[0098] In one example of this application, trajectory points with a decision difference greater than or equal to a preset difference threshold and a distance between them less than a preset distance threshold are selected from multiple trajectory points to be tested as target trajectory points. It is then determined whether the number of target trajectory points is greater than or equal to a preset number threshold. If so, a target time frame corresponding to the target trajectory point is selected from the initial test time window, and the initial test time window is truncated according to this target time frame number, thereby generating a target test time window larger than or equal to a first preset window size.

[0099] For example, if the target time frames corresponding to the target trajectory points are 0.3s, 1.0s, and 1.5s respectively, then the target test time window generated from the initial test time window is [0.3s, 1.5s], and the first preset window size can be set to 0.3s.

[0100] In an optional embodiment of this application, step 204 may include the following sub-steps S21-S25:

[0101] S21, Select target trajectory points from multiple trajectory points to be tested. The decision difference is greater than or equal to the preset difference threshold, and the distance is less than the preset distance threshold.

[0102] S22, determine whether the number of target trajectory points is greater than or equal to a preset number threshold;

[0103] S23, if so, then select the target time frame number corresponding to the target trajectory point from the initial test time window;

[0104] S24, extract the initial test time window according to the target time frame number to obtain multiple intermediate test time windows;

[0105] In another embodiment of this application, the specific implementation process of S21-S24 is similar to that of steps S11-S14, and will not be repeated here.

[0106] S25, merge the intermediate test time windows according to the second preset window size to obtain multiple target test time windows.

[0107] Furthermore, S25 may include the following sub-steps:

[0108] Calculate the ratio of target trajectory points to the number of trajectory points to be tested within the intermediate test time window;

[0109] If the proportion of trajectory points is greater than or equal to the preset proportion threshold, then the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0110] In this embodiment of the application, after obtaining multiple intermediate test time windows, the ratio of the number of target trajectory points to the number of trajectory points to be tested within the intermediate test time window can be further calculated. If the ratio of trajectory points is greater than or equal to a preset ratio threshold, it indicates that the decision of the trajectory points to be tested between the intermediate test time windows may still be continuously changing. At this time, the intermediate test time windows can be merged according to the second preset window size to obtain multiple target test time windows.

[0111] Intermediate test time windows can overlap.

[0112] If the proportion of trajectory points is less than the proportion threshold, no merging will be performed, and the original intermediate test time window will be determined as the target test time window.

[0113] In an optional embodiment of this application, S25 may include the following sub-steps:

[0114] Calculate the average difference between the target trajectory points within the intermediate test time window;

[0115] If the average difference value is greater than or equal to the preset average difference threshold, then the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0116] In this embodiment, the average difference value of the target trajectory points within the intermediate test time window is calculated, for example, the average difference value of the decision differences of each target trajectory point, or the average difference value of acceleration and velocity. If the average difference value is greater than or equal to a preset average difference threshold, the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0117] If the average difference value is less than the average difference threshold, no merging is performed, and the original intermediate test time window is determined as the target test time window.

[0118] Step 205: Add buffer time intervals on both sides of the target test time window to obtain the updated test time window;

[0119] To ensure the accuracy of the test simulation, buffer time intervals can be added on both sides of the target test time window. This allows the vehicle simulation to have a certain buffer time before encountering significant decision differences, and also provides a buffer time from behavioral decision to its impact, thereby obtaining an updated test time window.

[0120] Step 206: Capture the initial test scenario according to the updated test time window to obtain the new initial test scenario;

[0121] The initial test scenario is captured again according to the updated test time window to construct a new initial test scenario.

[0122] Step 207: Jump to the step of running the decision component under test and the prior decision component in the initial test scenario to generate the decision information under test and the prior decision information.

[0123] After obtaining the new initial test scenario, proceed to step 202 to perform the vehicle operation simulation test again to determine whether the vehicle's decision-making differences are consistent. If they are inconsistent, continue the simulation until the simulation test results are consistent, and use this as a basis for improvement to provide to technical personnel for further improvement of the decision-making component under test.

[0124] Generation of road test scenarios: Road test simulations use more realistic test data, but because the probability of accidents occurring in real road tests is very small, finding potential test windows in real road tests is not easy. This algorithm can more effectively and in real-time identify test windows for testing.

[0125] For secondary verification of existing simulation test set results: For a single simulation result, versions A and B will generate two different simulation results. Due to the uncertainty of the original simulation test window, the comparison results between the two simulation results (taking the safety score as an example) are inaccurate. Therefore, the accuracy of the simulation results can be verified by finding new test windows for each of the two simulation results.

[0126] Regularly refresh the simulation test set test window: For major version updates, there may be more changes in decision-making, and the test window also needs to be changed. Therefore, the scenario test window can be refreshed regularly.

[0127] In this embodiment, by responding to a received test request, the test decision component, prior decision component, and initial test scenario are obtained to form the test data basis. The test decision component and prior decision component are run in the initial test scenario to obtain the test decision information and prior decision information corresponding to each decision component within the initial test time window. The test decision information and prior decision information are further compared to calculate the decision difference between them in each frame within the initial test time window. Finally, the initial test time window is adjusted according to the decision difference to generate the target test time window. This provides a more accurate length of the test time window corresponding to the autonomous driving scenario, thereby improving the accuracy of decision simulation.

[0128] Please see Figure 3 , Figure 3 This is a structural block diagram of a test time window adjustment device provided in Embodiment 3 of the present invention.

[0129] This invention provides a device for adjusting a test time window, comprising:

[0130] The data acquisition module 301 is used to respond to the received test request and acquire the decision component to be tested, the prior decision component, and the initial test scenario; the initial test scenario has an initial test time window;

[0131] The decision execution module 302 is used to run the decision component under test and the prior decision component in the initial test scenario, and generate the decision information under test and the prior decision information.

[0132] The decision difference calculation module 303 is used to compare the decision information to be tested with the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window.

[0133] The time window adjustment module 304 is used to adjust the initial test time window based on the decision differences and generate the target test time window.

[0134] Optionally, the decision information to be tested includes multiple trajectory points to be tested, and the prior decision information includes multiple prior trajectory points; the decision difference calculation module 303 includes:

[0135] The Euclidean distance calculation submodule is used to calculate the Euclidean distance between the test trajectory point and the previous trajectory point in each frame of the initial test time window, and obtain multiple Euclidean distance values.

[0136] The first decision difference calculation submodule is used to calculate the average value among all Euclidean distance values ​​and determine the decision difference of the decision information to be tested within the initial test time window.

[0137] Optionally, the decision information to be tested includes the speed control information to be tested, and the prior decision information includes the prior speed control information; the decision difference calculation module 303 includes:

[0138] The first vector extraction submodule is used to extract the first velocity vector and the first acceleration vector of the velocity control information to be tested in each frame within the initial test time window.

[0139] The second vector extraction submodule is used to extract the second velocity vector and the second acceleration vector of each frame within the initial test time window based on the prior velocity control information.

[0140] The difference calculation submodule is used to calculate the velocity difference between the first velocity vector and the second velocity vector frame by frame, or to calculate the acceleration difference between the first acceleration vector and the second acceleration vector frame by frame.

[0141] The second decision difference calculation submodule is used to determine the average value of the velocity difference or the average value of the acceleration difference as the decision difference of the decision information to be tested within the initial test time window.

[0142] Optionally, it includes:

[0143] The first target trajectory point selection submodule is used to select target trajectory points from multiple trajectory points to be tested whose decision difference is greater than or equal to a preset difference threshold and whose spacing is less than a preset distance threshold.

[0144] The first quantity judgment submodule is used to determine whether the number of target trajectory points is greater than or equal to a preset quantity threshold.

[0145] The first target time frame selection submodule is used to select the target time frame corresponding to the target trajectory point from the initial test time window if the condition is met.

[0146] The time window extraction submodule is used to extract the initial test time window according to the target time frame number and generate a target test time window that is greater than or equal to the first preset window size.

[0147] Optionally, the time window adjustment module 304 includes:

[0148] The second target trajectory point selection submodule is used to select target trajectory points from multiple trajectory points to be tested whose decision difference is greater than or equal to a preset difference threshold and whose spacing is less than a preset distance threshold.

[0149] The second quantity judgment submodule is used to determine whether the number of target trajectory points is greater than or equal to a preset quantity threshold.

[0150] The second target time frame selection submodule is used to select the target time frame corresponding to the target trajectory point from the initial test time window if the condition is met.

[0151] The intermediate test time window extraction submodule is used to extract the initial test time window according to the target time frame number, and obtain multiple intermediate test time windows;

[0152] The window merging submodule is used to merge various intermediate test time windows according to the second preset window size to obtain multiple target test time windows.

[0153] Optionally, the window merging submodule is specifically used for:

[0154] Calculate the ratio of target trajectory points to the number of trajectory points to be tested within the intermediate test time window;

[0155] If the proportion of trajectory points is greater than or equal to the preset proportion threshold, then the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0156] Optionally, the window merging submodule is specifically used for:

[0157] Calculate the average difference between the target trajectory points within the intermediate test time window;

[0158] If the average difference value is greater than or equal to the preset average difference threshold, then the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

[0159] Optionally, the device further includes:

[0160] The time window update module is used to add buffer time intervals on both sides of the target test time window to obtain an updated test time window;

[0161] The scene capture module is used to capture the initial test scene according to the updated test time window to obtain a new initial test scene;

[0162] The jump execution module is used to jump to the execution of the steps in the initial test scenario to run the decision component under test and the prior decision component, and to generate the decision information under test and the prior decision information.

[0163] This invention provides an electronic device, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the test time window adjustment method as described in any embodiment of this invention.

[0164] This invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the test time window adjustment method as described in any embodiment of this invention.

[0165] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0166] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0167] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0168] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0169] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0170] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for adjusting a test time window, characterized in that, include: In response to the received test request, obtain the decision component to be tested, the prior decision components, and the initial test scenario; The initial test scenario includes an initial test time window; The test decision component and the prior decision component are run in the initial test scenario to generate test decision information and prior decision information; By comparing the decision information to be tested with the prior decision information, the decision difference of the decision information to be tested within the initial test time window is determined; The initial test time window is adjusted based on the decision differences to generate the target test time window; The decision information to be tested includes multiple trajectory points to be tested. The step of adjusting the initial test time window based on the decision differences to generate a target test time window includes: Select target trajectory points from a plurality of trajectory points to be tested, wherein the decision difference is greater than or equal to a preset difference threshold and the spacing is less than a preset distance threshold; Determine whether the number of the target trajectory points is greater than or equal to a preset number threshold; If so, then select the target time frame number corresponding to the target trajectory point from the initial test time window; The initial test time window is extracted according to the target time frame number, and a target test time window with a size greater than or equal to the first preset window size is generated.

2. The method according to claim 1, characterized in that, The prior decision information includes multiple prior trajectory points; the step of comparing the decision information to be tested and the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window includes: The Euclidean distance between the trajectory point to be tested and the prior trajectory point is calculated one by one in each frame of the initial test time window to obtain multiple Euclidean distance values; Calculate the average value among all the Euclidean distance values ​​to determine the decision difference of the decision information to be tested within the initial test time window.

3. The method according to claim 1, characterized in that, The decision information to be tested includes the speed control information to be tested, and the prior decision information includes the prior speed control information; the step of comparing the decision information to be tested and the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window includes: Extract the first velocity vector and the first acceleration vector of each frame within the initial test time window of the speed control information to be tested; Extract the second velocity vector and the second acceleration vector of each frame within the initial test time window from the prior velocity control information; Calculate the velocity difference between the first velocity vector and the second velocity vector frame by frame, or calculate the acceleration difference between the first acceleration vector and the second acceleration vector frame by frame; The average value of the velocity difference or the average value of the acceleration difference is determined as the decision difference of the decision information to be tested within the initial test time window.

4. The method according to claim 2, characterized in that, The step of adjusting the initial test time window based on the decision difference to generate the target test time window includes: Select target trajectory points from a plurality of trajectory points to be tested, wherein the decision difference is greater than or equal to a preset difference threshold and the spacing is less than a preset distance threshold; Determine whether the number of the target trajectory points is greater than or equal to a preset number threshold; If so, then select the target time frame number corresponding to the target trajectory point from the initial test time window; The initial test time window is extracted according to the target time frame number to obtain multiple intermediate test time windows; The intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

5. The method according to claim 4, characterized in that, The step of merging the intermediate test time windows according to the second preset window size to obtain multiple target test time windows includes: Calculate the ratio of the number of target trajectory points to the number of trajectory points to be tested within the intermediate test time window; If the proportion of the trajectory points is greater than or equal to a preset proportion threshold, then the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

6. The method according to claim 4, characterized in that, The step of merging the intermediate test time windows according to the second preset window size to obtain multiple target test time windows includes: Calculate the average difference value of the target trajectory points within the intermediate test time window; If the average difference value is greater than or equal to the preset average difference threshold, the intermediate test time windows are merged according to the second preset window size to obtain multiple target test time windows.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: A buffer time interval is added on both sides of the target test time window to obtain an updated test time window; The initial test scenario is captured according to the updated test time window to obtain a new initial test scenario; Jump to execute the step of running the decision component under test and the prior decision component in the initial test scenario to generate decision information under test and prior decision information.

8. A device for adjusting a test time window, characterized in that, include: The data acquisition module is used to respond to received test requests and acquire the decision component to be tested, the prior decision components, and the initial test scenario; The initial test scenario includes an initial test time window; The decision execution module is used to run the decision component under test and the prior decision component in the initial test scenario, and generate the decision information under test and the prior decision information. The decision difference calculation module is used to compare the decision information to be tested with the prior decision information to determine the decision difference of the decision information to be tested within the initial test time window. The time window adjustment module is used to adjust the initial test time window according to the decision difference and generate a target test time window; The decision information to be tested includes multiple trajectory points to be tested, and the time window adjustment module includes: The first target trajectory point selection submodule is used to select target trajectory points from multiple trajectory points to be tested whose decision difference is greater than or equal to a preset difference threshold and whose spacing is less than a preset distance threshold. The first quantity determination submodule is used to determine whether the number of the target trajectory points is greater than or equal to a preset quantity threshold. The first target time frame selection submodule is used to select the target time frame corresponding to the target trajectory point from the initial test time window if the condition is met. The time window truncation submodule is used to truncate the initial test time window according to the target time frame number and generate a target test time window that is greater than or equal to the first preset window size.

9. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the test time window adjustment method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the method for adjusting the test time window as described in any one of claims 1-7.

Images