Simulation scenario updating method, electronic device, and readable storage medium

Abstract

The application discloses a simulation scene updating method, an electronic device and a readable storage medium, and relates to the technical field of automobile testing. The method comprises the following steps: acquiring first motion parameters of each obstacle in a simulation scene for a first number of frames before a current frame and second motion parameters corresponding to the current frame; predicting third motion parameters of the current frame of each obstacle based on the first motion parameters of the obstacle; determining a target obstacle that satisfies a target condition based on the difference between the second motion parameters of each obstacle and the third motion parameters of the obstacle; and deleting the target obstacle from the simulation scene to update the simulation scene. The obstacles in the simulation scene can be automatically updated, the artificial cost is reduced, and the accuracy of simulation backfilling is improved.

Description

Technical Field

[0001] This application relates to the field of automotive testing technology, and more specifically, to a simulation scene update method, electronic device, and readable storage medium. Background Technology

[0002] Currently, with the development of electronic information technology, vehicles can be tested through simulated scenarios. However, obstacles in simulated scenarios generally require manual adjustment, which incurs high labor costs. Summary of the Invention

[0003] This application proposes a simulation scene update method, an electronic device, and a readable storage medium.

[0004] In a first aspect, embodiments of this application provide a simulation scene update method, comprising: obtaining first motion parameters of each obstacle in the simulation scene for a first number of frames prior to the current frame and second motion parameters corresponding to the current frame; predicting third motion parameters of the obstacle in the current frame based on the first motion parameters of each obstacle; determining target obstacles whose differences satisfy target conditions based on the differences between the second motion parameters of each obstacle and the third motion parameters of the obstacle; and deleting the target obstacles from the simulation scene to update the simulation scene.

[0005] Optionally, in one possible implementation, the second motion parameter includes the second position information and / or the second heading angle information of the obstacle, and the third motion parameter includes the third position information and / or the third heading angle information of the obstacle predicted based on the first motion parameters of each obstacle. The step of determining a target obstacle whose difference satisfies the target condition based on the difference between the second motion parameters of each obstacle and the third motion parameters of that obstacle includes: obtaining a difference threshold corresponding to the position parameters and / or heading angle parameters of the obstacle; obtaining the difference between the second position information and the third position information of the obstacle, and / or obtaining the difference between the second heading angle information and the third heading angle information of the obstacle; when it is detected that the difference between the second position information and the third position information of the obstacle is greater than the difference threshold corresponding to the position parameters, and / or, the difference between the second heading angle information and the third heading angle information of the obstacle is greater than the difference threshold corresponding to the heading angle parameters, the obstacle is determined to be a target obstacle.

[0006] Optionally, in one possible implementation, the second motion parameter includes a second velocity and / or a second acceleration of the obstacle, and the method further includes: obtaining the maximum value of the velocity parameter and / or acceleration parameter of the obstacle; when the second velocity is detected to be greater than the maximum value of the velocity parameter, and / or the second acceleration is detected to be greater than the maximum value of the acceleration parameter, the obstacle is determined to be a target obstacle.

[0007] Optionally, in one possible implementation, the obstacle includes multiple types of obstacles, each type of obstacle corresponding to its own position parameter difference threshold, heading angle parameter difference threshold, maximum value of velocity parameter, and maximum value of acceleration parameter.

[0008] Optionally, in one possible implementation, after deleting the target obstacle from the simulation scene to update the simulation scene, the method further includes: determining a collision event in which the vehicle collides with the obstacle based on the motion parameters of the vehicle and the motion parameters of the obstacle in the updated simulation scene; acquiring feature information of the collision event, wherein the feature information includes obstacle feature information, vehicle feature information, and collision scene feature information; determining core features of the colliding obstacle based on the feature information of the collision event, wherein the core features include at least one of confidence, velocity, acceleration, and position, and the colliding obstacle includes obstacles that cause collision events due to detection errors; and updating the obstacle filtering rules of the simulation scene based on the algorithm logic corresponding to the core features.

[0009] Optionally, in one possible implementation, the core feature includes confidence level, and the algorithm logic based on the core feature updates the obstacle filtering rules for the simulation scene, including: obtaining the current confidence level threshold, wherein the confidence level threshold is used to characterize the critical threshold for filtering obstacles; determining a first proportion of collision obstacles with a confidence level less than the current confidence level threshold; and increasing the current confidence level threshold if the first proportion is greater than or equal to the first threshold.

[0010] Optionally, in one possible implementation, the core feature includes speed, and the algorithm logic based on the core feature updates the filtering rules for obstacles in the simulation scene, including: obtaining a current speed threshold, wherein the speed threshold is used to characterize the critical speed for filtering obstacles of the corresponding type; determining a second proportion of collision obstacles whose speed is greater than the current speed threshold; and reducing the current speed threshold if the second proportion is greater than or equal to the second threshold.

[0011] Optionally, in one possible implementation, the core feature includes location, and the algorithm logic based on the core feature updates the filtering rules for obstacles in the simulation scene, including: determining a third proportion of collision obstacles whose locations are in a specified location region, wherein the specified location region is used to characterize the location for filtering obstacles that meet the target size; and updating the rule corresponding to the location to filter obstacles in the specified location region when the third proportion is greater than or equal to a third threshold.

[0012] Secondly, embodiments of this application also provide an electronic device, the electronic device comprising: one or more processors; a memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, the one or more application programs being configured to perform the method described in the first aspect.

[0013] Thirdly, embodiments of this application also provide a computer-readable storage medium storing program code that can be invoked by a processor to execute the method described in the first aspect above.

[0014] The simulation scene update method, electronic device, and readable storage medium provided in this application first obtain the first motion parameters of each obstacle in the simulation scene for a first number of frames prior to the current frame, and the second motion parameters corresponding to the current frame. Then, based on the first motion parameters of each obstacle, a third motion parameter of the obstacle in the current frame is predicted. Next, based on the difference between the second motion parameters and the third motion parameters of each obstacle, a target obstacle whose difference satisfies a target condition is identified. The target obstacle is then deleted from the simulation scene to update the simulation scene. In this application, whether to delete the corresponding obstacle from the simulation scene can be determined based on whether the difference between the predicted third motion parameter and the second motion parameter satisfies the target condition. This achieves automatic updating of obstacles in the simulation scene, reduces manual costs, and enables seamless updates. Furthermore, since the deleted target obstacle is an obstacle whose difference satisfies the target condition, and the difference is the difference between the predicted third motion parameter and the second motion parameter, the target obstacle can represent an obstacle whose motion parameters exceed a reasonable range. Therefore, deleting the target obstacle from the simulation scene can improve the accuracy of simulation refeedback.

[0015] Other features and advantages of the embodiments of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the embodiments of this application. The objects and other advantages of the embodiments of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1A flowchart of the simulation scene update method provided in an embodiment of this application is shown; Figure 2 A flowchart of a simulation scene update method according to another embodiment of this application is shown; Figure 3 A flowchart of a simulation scene update method according to another embodiment of this application is shown; Figure 4 A flowchart of a simulation scene update method provided in another embodiment of this application is shown; Figure 5 A flowchart of a simulation scene update method according to another embodiment of this application is shown; Figure 6 A structural block diagram of the electronic device provided in an embodiment of this application is shown; Figure 7 This paper shows a structural block diagram of a computer-readable storage medium provided in an embodiment of this application; Figure 8 A structural block diagram of a computer program product provided in an embodiment of this application is shown. Detailed Implementation

[0018] 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. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. The components of the embodiments of the present application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are within the scope of protection of the present application.

[0019] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] Currently, with the development of electronic information technology, vehicles can be tested through simulated scenarios. However, obstacles in simulated scenarios generally require manual adjustment, which incurs high labor costs.

[0021] Currently, data can be collected from real vehicles and then converted into simulation scene files, allowing for simulation re-engineering. Simulation re-engineering is a technique in automotive development and testing that reintroduces real-world data or model results into a simulation system or controller. Its core principle is to improve simulation accuracy, reduce testing costs, and accelerate system verification through a combination of virtual and real technologies, with widespread applications, particularly in intelligent driving and automotive electronics development. The simulation scene can include obstacles, each with corresponding motion parameters. These parameters allow for manual adjustments to the obstacles in the simulation scene, such as deleting them.

[0022] However, the inventors discovered during their research that manually determining whether obstacles in a simulated scene need to be deleted before doing so is labor-intensive. Therefore, this application provides a simulation scene update method, an electronic device, and a readable storage medium.

[0023] Please see Figure 1 , Figure 1 A flowchart of a simulation scene update method provided in an embodiment of this application is shown. This simulation scene update method can be applied to electronic devices, specifically including steps S110 to S140.

[0024] Step S110: Obtain the first motion parameters of each obstacle in the simulation scene for the first number of frames prior to the current frame, and the second motion parameters corresponding to the current frame.

[0025] It is understandable that a simulation scene is essentially a configuration file or data file, storing the structural information of the scene. For example, a simulation scene may include environmental parameters, entity models, behavioral rules, and simulation parameters, where the instance models may include obstacles and vehicles.

[0026] In this simulation, obstacles can have corresponding motion parameters. For example, motion parameters may include position, velocity, or acceleration. Each obstacle in the simulation scene can correspond to multiple frames, and each frame's obstacles can have corresponding motion parameters. It's understood that for the same obstacle, the motion parameters in adjacent frames should change continuously and smoothly, without significant variations. If significant variations occur, or if the variations do not conform to physical limits, the motion parameters of that obstacle may be incorrect, and in this case, removing the obstacle from the simulation scene should be considered.

[0027] Therefore, in this embodiment, various obstacles in the simulated scene can be detected, thereby obtaining the first motion parameters of each obstacle in the first number of frames preceding the current frame. The first number of frames can be a small number of frames, such as 3 frames, 4 frames, etc.

[0028] In a simulation scenario, there may be multiple frames of images. Different obstacles may appear for the first time in different frames. Therefore, in some implementations, each frame of the simulation scenario can be traversed to determine each obstacle. Then, the obstacles corresponding to the current frame are determined, and then the motion parameters of each obstacle in the first number of frames preceding the current frame are determined as the first motion parameters corresponding to that obstacle. This first motion parameter can also be regarded as the initial motion parameters of the obstacle.

[0029] It should be noted that before the current frame, some obstacles may not have appeared in the first number of frames. In this case, the first motion parameters of these obstacles do not need to be obtained. For example, if the first number of frames is 3, and obstacle A has only appeared in 2 frames before the current frame, then the first motion parameters of obstacle A do not need to be obtained.

[0030] As can be seen from the foregoing introduction, motion parameters can include position, velocity, or acceleration, etc. Therefore, the first motion parameter determined here can also include position, velocity, or acceleration, etc.

[0031] Furthermore, the subsequent motion parameters of the obstacle can be predicted using the first motion parameters. Please refer to the following embodiments for a detailed explanation.

[0032] Additionally, the second motion parameters corresponding to the current frame can be obtained, specifically the second motion parameters of each obstacle in the current frame. For example, the second motion parameters corresponding to the current frame can be obtained from the motion parameters stored in the simulation scene.

[0033] Step S120: Predict the third motion parameters of each obstacle in the current frame based on the first motion parameters of each obstacle.

[0034] After obtaining the first motion parameters of each obstacle, the third motion parameters of that obstacle in the current frame can be predicted based on those first motion parameters. It should be noted that this prediction is based on the obstacle's own first motion parameters; it is a predicted value, not the motion parameters of the measured values ​​directly obtained from the current frame.

[0035] In some implementations, predicting the third motion parameter of the obstacle in the current frame can be done by determining the initial motion parameter based on the first motion parameter of the obstacle, and then predicting the third motion parameter based on the acceleration model combined with the initial motion parameter. For details, please refer to the following embodiments.

[0036] Step S130: Based on the difference between the second motion parameter and the third motion parameter of each obstacle, determine the target obstacle whose difference satisfies the target condition.

[0037] Understandably, the aforementioned predicted third motion parameter is derived from the first motion parameter through physical laws. Subsequently, the third motion parameter can be compared with the second motion parameter. If the third motion parameter differs significantly from the second motion parameter, it indicates that the second motion parameter stored in the simulation scene may be incorrect. In this case, the corresponding obstacle can be removed from the simulation scene.

[0038] It should be noted that the second motion parameter is the actual measured value stored in the simulation scene, while the third motion parameter is the predicted value obtained by predicting the first motion parameter based on the actual measurement.

[0039] Specifically, the difference between the third motion parameter and the second motion parameter of the obstacle can be obtained. The larger the difference, the more likely the second motion parameter stored in the simulation scene is to be incorrect; conversely, the smaller the difference, the more likely the second motion parameter stored in the simulation scene is to be correct. An incorrect second motion parameter indicates that the corresponding obstacle is abnormal.

[0040] In some implementations, target obstacles can be identified by determining whether the difference meets target conditions. Specifically, obstacles corresponding to differences that meet target conditions can be considered target obstacles. It should be noted that meeting target conditions can be used to characterize large differences.

[0041] Optionally, obstacles may correspond to position parameters and / or heading angle parameters, and these position parameters and / or heading angle parameters may correspond to difference thresholds. Therefore, the difference can be compared with the difference threshold to determine whether the difference meets the target conditions. Please refer to subsequent embodiments for a detailed description.

[0042] Step S140: Delete the target obstacle from the simulation scene to update the simulation scene.

[0043] Once the target obstacle is identified, it can be removed from the simulation scene to update the simulation scene.

[0044] For example, the target obstacle and its corresponding motion parameters can be deleted from the simulation scene.

[0045] The simulation scene update method provided in this application first obtains the first motion parameters of each obstacle in the simulation scene for a first number of frames prior to the current frame, and the second motion parameters corresponding to the current frame; then, based on the first motion parameters of each obstacle, it predicts the third motion parameters of the obstacle in the current frame; and then, based on the difference between the second motion parameters of each obstacle and the third motion parameters of the obstacle, it determines the target obstacle whose difference meets the target condition; and deletes the target obstacle from the simulation scene to update the simulation scene. In this application, it is possible to determine whether to delete the corresponding obstacle from the simulation scene based on whether the difference between the predicted third motion parameters and the second motion parameters meets the target condition. Thus, automatic updating of obstacles in the simulation scene is achieved, reducing manual costs and enabling seamless updates. In addition, since the deleted target obstacle is an obstacle whose difference meets the target condition, and the difference is the difference between the predicted third motion parameters and the second motion parameters, the target obstacle can represent an obstacle whose motion parameters exceed a reasonable range. Therefore, deleting the target obstacle from the simulation scene can improve the accuracy of simulation refeedback.

[0046] Please see Figure 2 , Figure 2 A flowchart of a simulation scene update method provided in an embodiment of this application is shown. This simulation scene update method can be applied to electronic devices, specifically including steps S210 to S270.

[0047] Step S210: Obtain the first motion parameters of each obstacle in the simulation scene for the first number of frames before the current frame and the second motion parameters corresponding to the current frame.

[0048] Step S210 has been described in detail in the foregoing embodiments and will not be repeated here.

[0049] Step S220: Predict the third motion parameters of each obstacle in the current frame based on the first motion parameters of each obstacle.

[0050] In some implementations, the acquired first motion parameter can be considered as a sampled value, which can be used to initialize the fourth motion parameter of the obstacle. For example, the fourth motion parameter can be a Kalman filter model parameter. Thus, the Kalman filter model parameter can be initialized using the first motion parameter of the obstacle in the first number of frames prior to the current frame.

[0051] Furthermore, based on the uniform acceleration model and the fourth motion parameters of the obstacle, the subsequent third motion parameters of the obstacle are predicted. For example, the third motion parameters can be calculated by keeping the magnitude and direction of the acceleration constant based on the fourth motion parameters. It is understood that the third motion parameters are predicted, not measured values.

[0052] Step S230: Obtain the difference threshold corresponding to the position parameters and / or heading angle parameters of the obstacle.

[0053] Step S240: Obtain the difference between the second and third position information of the obstacle, and / or obtain the difference between the second and third heading angle information of the obstacle.

[0054] Step S250: When the difference between the second and third position information of the obstacle is greater than the difference threshold corresponding to the position parameter, and / or the difference between the second and third heading angle information of the obstacle is greater than the difference threshold corresponding to the heading angle parameter, the obstacle is determined to be a target obstacle.

[0055] In some implementations, the second motion parameter may include second position information and / or second heading angle information of the obstacle, and the third motion parameter may include third position information and / or third heading angle information of the obstacle predicted based on the first motion parameters of each obstacle.

[0056] Therefore, the difference thresholds corresponding to the obstacle's position parameters and / or heading angle parameters can be obtained first. For example, the difference threshold for the position parameters can be 4m, and the difference threshold for the heading angle parameters can be 25°. Optionally, the difference threshold for the position parameters can also be called the position deviation threshold, and the difference threshold for the heading angle parameters can also be called the heading deviation threshold.

[0057] In some implementations, the difference thresholds corresponding to the position parameters and / or heading angle parameters of each obstacle can be stored as data in the simulation scene, for example, they can be preset by the user.

[0058] Therefore, the difference between the second and third position information of the obstacle can be obtained, and / or the difference between the second and third heading angle information of the obstacle can be obtained. Then, the difference between the second and third position information is compared with the difference threshold corresponding to the position parameter, and / or, the difference between the second and third heading angle information of the obstacle is compared with the difference threshold corresponding to the heading angle parameter.

[0059] For example, if the difference between the second and third position information of an obstacle is 3m, then the difference between the second and third position information of the obstacle is less than the difference threshold corresponding to the position parameter.

[0060] For example, if the difference between the second heading angle information and the third heading angle information of the obstacle is 30°, then the difference between the second heading angle information and the third heading angle information of the obstacle is greater than the difference threshold corresponding to the heading angle parameter.

[0061] This confirms that the obstacle is the target obstacle.

[0062] Additionally, in some embodiments, the second motion parameter may include the second velocity and / or second acceleration of the obstacle.

[0063] Therefore, the simulation scene update method provided in this application embodiment may further include steps S260 and S270.

[0064] Step S260: Obtain the maximum value of the velocity parameters and / or acceleration parameters of the obstacle.

[0065] Step S270: When the second speed is detected to be greater than the maximum value of the speed parameter, and / or the second acceleration is greater than the maximum value of the acceleration parameter, the obstacle is determined to be a target obstacle.

[0066] First, the maximum values ​​of the obstacle's velocity and / or acceleration parameters can be obtained. These maximum values ​​characterize the physical critical values ​​for the obstacle's motion; if these values ​​are exceeded, there is a high probability that the obstacle's motion parameters are incorrect. For example, the maximum value for the velocity parameter could be 25 m / s², and the maximum value for the acceleration parameter could be 5 m / s². 2 Optionally, the maximum value of the velocity parameter can also be called the maximum velocity; the maximum value of the acceleration parameter can also be called the maximum acceleration.

[0067] In some implementations, the maximum values ​​of the velocity and / or acceleration parameters of each obstacle can be stored as data in the simulation scene, for example, in a user-preset manner.

[0068] Therefore, the second velocity can be compared with the maximum value of the velocity parameter, and / or the second acceleration can be compared with the maximum value of the acceleration parameter.

[0069] When the second speed is detected to be greater than the maximum value of the speed parameter, and / or the second acceleration is greater than the maximum value of the acceleration parameter, the obstacle is determined to be a target obstacle.

[0070] For example, if the second speed is 20 m / s, then the second speed is less than the maximum value of the speed parameter, which is 25 m / s.

[0071] For example, if the second acceleration is 6 m / s² 2 At this point, the second acceleration is greater than the maximum value corresponding to the acceleration parameter, which is 5 m / s². 2 .

[0072] This allows us to identify the obstacle as the target obstacle.

[0073] Optionally, the second motion parameter may also include the obstacle's heading angle and velocity direction.

[0074] Therefore, it is also possible to determine whether an obstacle is a target obstacle based on whether its heading angle and velocity direction are consistent. Specifically, if the heading angle and velocity direction of an obstacle are inconsistent, it can be determined that the obstacle is a target obstacle.

[0075] For example, if the heading angle is forward but the speed is backward, the obstacle can be identified as the target obstacle, thereby further improving the accuracy of judging whether the obstacle is abnormal.

[0076] Optionally, in some implementations, to further improve the accuracy of judging whether there are errors in the motion parameters of the obstacle, it can be set that the obstacle is identified as a target obstacle only when a specified condition is detected multiple times consecutively.

[0077] Optionally, the motion parameters corresponding to the next frame after the current frame can be obtained as new second motion parameters. The process then returns to the previous steps, including obtaining the difference between the second and third position information of the obstacle, and / or obtaining the difference between the second and third heading angle information of the obstacle, and subsequent steps, until a specified number of consecutive detections are made to determine that the obstacle is the target obstacle. The specified conditions include: the difference between the second and third position information of the obstacle being greater than the difference threshold corresponding to the position parameters; and / or, the difference between the second and third heading angle information of the obstacle being greater than the difference threshold corresponding to the heading angle parameter; and / or, the second velocity being greater than the maximum value of the velocity parameter; and / or, the second acceleration being greater than the maximum value of the acceleration parameter.

[0078] Specifically, when a specified condition is detected, it can be determined whether the specified condition has been detected a second number of times consecutively. If it is determined that the specified condition has been detected a second number of times consecutively, the obstacle can be identified as the target obstacle. If the specified condition has not been detected a second number of times consecutively, the motion parameters corresponding to the next frame after the current frame can be obtained as the new second motion parameters, and the process can be returned to obtain the difference between the second and third position information of the obstacle, and / or obtain the difference between the second and third heading angle information of the obstacle, and subsequent steps, until the specified condition is detected a second number of times consecutively, and the obstacle is identified as the target obstacle.

[0079] In some implementations, the second number of times can be three. This means that the second motion parameters corresponding to multiple consecutive frames must all be abnormal; that is, the detection of a specified situation is determined based on the second motion parameters corresponding to multiple consecutive frames, thereby improving the accuracy of obstacle anomaly detection.

[0080] Optionally, when returning to the process of obtaining the difference between the second and third position information of the obstacle, and / or obtaining the difference between the second and third heading angle information of the obstacle, and subsequent steps, the fourth motion parameter can also be updated based on the historical motion parameters of the obstacle to obtain a new fourth motion parameter. The historical motion parameters may include the motion parameters corresponding to the obstacle in each frame prior to the current frame.

[0081] Optionally, if the specified condition of the second number of consecutive detections has not yet been achieved, the predicted third motion parameter can be updated using the current second motion parameter to obtain a new third motion parameter, and the new third motion parameter can be used to determine whether the specified condition is met in the future.

[0082] Optionally, obstacles may include multiple types of obstacles, each type of obstacle corresponding to its own position parameter difference threshold, heading angle parameter difference threshold, maximum value of velocity parameter, and maximum value of acceleration parameter.

[0083] For example, various types of obstacles may include four-wheeled vehicles, two-wheeled vehicles, and pedestrians. The maximum values ​​of position parameter difference threshold, heading angle parameter difference threshold, velocity parameter, and acceleration parameter corresponding to each type of obstacle can be found in Table 1 below.

[0084] Table 1. Difference thresholds and maximum values Obstacle types Location parameter difference threshold Maximum value of speed parameter Maximum value of acceleration parameter Heading angle parameter difference threshold Four-wheeled vehicle 4m 25m / s <![CDATA[5m / s 2 ]]> 25° two-wheeled vehicle 1.5m 8m / s <![CDATA[3m / s 2 ]]> 40° pedestrian 2.5m 10m / s <![CDATA[1.5m / s 2 ]]> 25° As can be seen from Table 1 above, four-wheeled vehicles, two-wheeled vehicles, and pedestrians each have their own threshold values ​​for position parameter difference, heading angle parameter difference, maximum value of speed parameter, and maximum value of acceleration parameter.

[0085] Step S280: Delete the target obstacle from the simulation scene to update the simulation scene.

[0086] Step S280 has been described in detail in the foregoing embodiments and will not be repeated here.

[0087] The simulation scene update method provided in this application can obtain the difference threshold corresponding to the position parameters and / or heading angle parameters of an obstacle; then, it obtains the difference between the second and third position information of the obstacle, and / or the difference between the second and third heading angle information of the obstacle; thus, when the difference between the second and third position information of the obstacle is detected to be greater than the difference threshold corresponding to the position parameters, and / or the difference between the second and third heading angle information of the obstacle is greater than the difference threshold corresponding to the heading angle parameters, the obstacle is determined to be a target obstacle. This can improve the accuracy of judging obstacle anomalies, thereby improving the accuracy of subsequent simulation refeedback.

[0088] Please see Figure 3 , Figure 3 A flowchart of a simulation scene update method provided in an embodiment of this application is shown. This simulation scene update method can be applied to electronic devices, specifically including steps S310 to S380.

[0089] Step S310: Obtain the first motion parameters of each obstacle in the simulation scene for the first number of frames before the current frame and the second motion parameters corresponding to the current frame.

[0090] Step S320: Predict the third motion parameters of each obstacle in the current frame based on the first motion parameters of each obstacle.

[0091] Step S330: Based on the difference between the second motion parameter and the third motion parameter of each obstacle, determine the target obstacle whose difference satisfies the target condition.

[0092] Step S340: Delete the target obstacle from the simulation scene to update the simulation scene.

[0093] Steps S310 to S340 have been described in detail in the foregoing embodiments and will not be repeated here.

[0094] Step S350: Based on the motion parameters of the vehicle in the updated simulation scene and the motion parameters of the obstacles in the updated simulation scene, determine the collision event in which the vehicle collides with the obstacle.

[0095] Step S360: Obtain the feature information of the collision event, wherein the feature information includes obstacle feature information, vehicle feature information and collision scene feature information.

[0096] Step S370: Determine the core features of the colliding obstacle based on the feature information of the collision event, wherein the core features include at least one of confidence, velocity, acceleration and position, and the colliding obstacle includes obstacles that cause a collision event due to detection errors of the obstacle.

[0097] Step S380: Update the obstacle filtering rules for the simulation scene based on the algorithm logic corresponding to the core features.

[0098] The simulation scenario includes collision events. By learning the motion parameters of traffic participants in these events, the filtering rules for obstacles in the simulation scenario can be optimized and updated, thereby improving the accuracy of obstacle optimization and the accuracy of simulation feedback. Traffic participants can include vehicles and obstacles.

[0099] First, a collision event in which the vehicle collides with an obstacle can be determined. For example, the collision event can be determined based on the motion parameters of the vehicle and the obstacle in the updated simulation scene.

[0100] The updated simulation scene can be the simulation scene after the aforementioned steps have been completed, in which the target obstacles have been deleted to update the simulation scene.

[0101] For example, the motion parameters of the vehicle may include the position of the vehicle, and the motion parameters of the obstacle may include the position of the obstacle. Thus, it can be determined whether the distance between the vehicle and any obstacle is less than or equal to 0 based on the position of the vehicle and the position of the obstacle. If the distance is less than or equal to 0, it can be determined that a collision event has occurred between the vehicle and the obstacle.

[0102] For example, a collision event can also include the existence of a collision risk, which indicates a high probability that a collision event is likely to occur. Specifically, the motion parameters of the vehicle can include its position and speed, and the motion parameters of the obstacle can include its position and speed. Based on the position and speed of the vehicle and the obstacle, it can be determined whether the distance between the vehicle and the obstacle is less than a safe threshold and the speed is relatively positive. For example, the safe threshold is 0.5m, which can be preset. If the distance to the obstacle is less than the safe threshold and the speed is relatively positive, a collision risk can be determined, and this can also be considered as a detected collision event between the vehicle and the obstacle.

[0103] Furthermore, feature information of the collision event can be obtained, wherein the feature information includes obstacle feature information, vehicle feature information, and collision scene feature information.

[0104] For example, the obstacle characteristic information of the collision may include type, size, speed, acceleration, position relative to the vehicle, time of appearance, confidence level, sensor source, and obstacle trajectory. Specifically, the type may include a vehicle or a pedestrian, the size may include length, width, or height, the position relative to the vehicle may include distance or azimuth, and the sensor source may include a single sensor or multiple sensors.

[0105] For example, the vehicle's characteristic information in the event of a collision may include the time of the collision and the vehicle's state. Specifically, the vehicle's state may include speed, acceleration, or steering.

[0106] For example, the collision scenario feature information may include road type, traffic rules, weather, and time. Specifically, road type may include highway or urban area, traffic rules may include traffic lights or stop and yield, weather may include sunny or rainy, and time may include day or night.

[0107] Optionally, the feature information of collision events can be integrated into a collision event database, which records the feature information of collision events and collision results.

[0108] Understandably, the causes of collisions can be categorized based on whether they are due to inaccurate obstacle perception. For example, false obstacle detection, abnormal motion parameters, or contradictory features may lead to subsequent collisions, and these causes can be considered as being due to inaccurate obstacle perception. Conversely, collisions caused by incorrect vehicle decisions or normal obstacle motion parameters can be considered as not being due to inaccurate obstacle perception.

[0109] Therefore, the core features of the colliding obstacle can be determined based on the feature information of the collision event. These core features include at least one of confidence level, velocity, acceleration, and location. Specifically, the colliding obstacle includes the obstacles mentioned above where a collision event occurs due to detection errors. For example, false obstacle detection may include the appearance of an obstacle in an area not present on the map; abnormal obstacle motion parameters may include instantaneous velocity exceeding physical limits; contradictory obstacle features may include size and type mismatch.

[0110] In some implementations, the characteristic information of collision events can be analyzed using expert experience to determine the collision obstacle. Optionally, a machine learning model can be pre-built to select the collision obstacle from among the obstacles based on the characteristic information of the collision event; this application does not impose specific limitations on the embodiments.

[0111] Optionally, obstacles that do not cause collisions due to detection errors, i.e., non-collision obstacles, can be recorded as normal collisions.

[0112] Furthermore, the core features of the collision obstacle can be obtained. For example, the core features include confidence level, velocity, and location. Specifically, the confidence level is 0.3, the velocity is 50 m / s, and the location is that it appears in the green belt.

[0113] Then, the core features of the collision obstacles can be statistically analyzed, and the filtering rules for obstacles in the simulation scene can be updated based on the algorithm logic corresponding to the core features.

[0114] It should be noted that the filtering rules in the simulation scenario are used to characterize the removal of obstacles that satisfy the rules.

[0115] In some implementations, the core feature includes confidence level. Therefore, step S380 may include steps S381 to S383.

[0116] Step S381: Obtain the current confidence threshold, wherein the confidence threshold is used to characterize the critical threshold for filtering obstacles.

[0117] Step S382: Determine the first proportion of collision obstacles with a confidence level lower than the current confidence threshold.

[0118] Step S383: If the first proportion is greater than or equal to the first threshold, increase the current confidence threshold.

[0119] The current confidence threshold can be obtained, whereby the confidence threshold is used to characterize the critical threshold for filtering obstacles. For example, the current velocity threshold can be stored in a file within the simulation scene. It is understood that the confidence threshold is the filtering rule corresponding to the confidence level.

[0120] Furthermore, the confidence level of a collision obstacle can be compared with the current confidence level to determine the first proportion of collision obstacles with a confidence level lower than the current confidence threshold. It should be noted that the first proportion represents the percentage of collision obstacles with a confidence level lower than the current confidence threshold out of the total number of collision obstacles.

[0121] Therefore, if the first proportion is greater than or equal to the first threshold, the current confidence threshold can be increased. The first threshold can be a pre-set proportion value. When the first proportion is greater than or equal to the first threshold, it indicates that even if the current confidence threshold is exceeded, some obstacles may still collide. In this case, the current confidence threshold can be appropriately increased. For example, if the current confidence threshold is 0.5, it can be increased to 0.6.

[0122] In some implementations, the core feature includes speed. Therefore, step S380 may also include steps S384 to S386.

[0123] Step S384: Obtain the current speed threshold, wherein the speed threshold is used to characterize the critical speed for filtering the corresponding type of obstacle.

[0124] Step S385: Determine the second proportion of collision obstacles whose speed is greater than the current speed threshold.

[0125] Step S386: If the second percentage is greater than or equal to the second threshold, reduce the current speed threshold.

[0126] The current speed threshold can be obtained, whereby the speed threshold characterizes the critical speed for filtering corresponding types of obstacles. This current speed threshold can be stored in a file within the simulation scene. It can be understood that the speed threshold is the filtering rule corresponding to the speed.

[0127] Furthermore, the speed of the colliding obstacle can be compared with the current speed threshold to determine a second proportion of the colliding obstacles whose speed exceeds the current speed threshold. It should be noted that the second proportion represents the percentage of all colliding obstacles whose speed exceeds the current speed threshold. The fact that the speed exceeds the current speed threshold may indicate that the current threshold is set too high, causing unreasonable high-speed false targets to be included in the second proportion.

[0128] Therefore, if the second proportion is greater than or equal to the second threshold, the current speed threshold can be reduced. The second threshold can be a pre-set proportion value. When the second proportion is greater than or equal to the second threshold, it indicates that an obstacle has caused collisions multiple times due to exceeding the current speed threshold, and the current speed threshold can be appropriately reduced. For example, for pedestrian obstacles, if the current speed threshold is 3 m / s, it can be reduced to 2.5 m / s.

[0129] Similarly, core features can include acceleration. It's understandable that the acceleration threshold is the filtering rule corresponding to acceleration.

[0130] Then, the current acceleration threshold can be obtained, and the acceleration of the colliding obstacle can be compared with the current acceleration threshold to determine the fourth proportion of colliding obstacles whose acceleration is greater than the current acceleration threshold. Therefore, if the fourth proportion is greater than or equal to the fourth threshold, the current acceleration threshold can be lowered. The fourth threshold can be a pre-set proportion value.

[0131] In some implementations, the core feature includes location. Therefore, step S380 may also include steps S387 and S388.

[0132] Step S387: Determine the third proportion of collision obstacles whose positions are located in a specified position area, wherein the specified position area is used to characterize the position of filtering obstacles that meet the target size.

[0133] Step S388: If the third proportion is greater than or equal to the third threshold, update the rule corresponding to the position to filter obstacles in the specified position area.

[0134] Understandably, the simulation scenario can store relevant data for a specified location area. This specified location area represents the location where obstacles meeting the target size are filtered out. For example, the specified scenario includes a green belt; if an obstacle's size exceeds the target size, that obstacle can be filtered out.

[0135] In some implementations, each collision obstacle may have a corresponding location, and based on the location of each collision obstacle, it can be determined whether the collision obstacle is within a specified location area. Then, collision obstacles whose locations are within the specified location area can be searched to obtain a third proportion of collision obstacles whose locations are within the specified location area. It should be noted that the third proportion is used to characterize the percentage of the number of collision obstacles whose locations are within the specified location area out of the total number of collision obstacles.

[0136] Furthermore, if the third proportion is greater than or equal to the third threshold, the rule corresponding to the position can be updated to filter obstacles in the specified position area. The third threshold can be a pre-set proportion value. When the third proportion is greater than or equal to the third threshold, it indicates that a certain specified position area is more prone to falsely detected obstacles. In this case, the filtering rule corresponding to the position can be directly updated to filter obstacles in the specified position area. That is, it is not necessary to consider the size of the obstacle in the specified position area; rather, once an obstacle is detected as being in the specified position area, that obstacle can be filtered out.

[0137] Optionally, the system can analyze whether a collision obstacle is a high-risk obstacle based on its core features, thereby increasing the confidence threshold for high-risk collision obstacles. For example, if a collision obstacle's sensor source is a single sensor and the obstacle's speed is relatively high, it can be determined as a high-risk obstacle. Then, the confidence threshold for this high-risk collision obstacle can be increased, for example, by 20%.

[0138] In some implementations, steps S350 to S380 can be automatically triggered after a specified number of simulations at a set interval; alternatively, steps S350 to S380 can be automatically triggered after the cumulative number of detected collision events exceeds a fourth threshold. This application does not impose any specific limitations on these implementations.

[0139] For example, the specified number of times can be 100 times; the fourth threshold can be 50 times.

[0140] Optionally, to avoid negative impacts from updated filtering rules, after updating the filtering rules in a specific simulation scenario, the effectiveness of the updated filtering rules can be verified. Only after verifying effectiveness can the filtering rules be applied to all simulation scenarios. For example, if updating the filtering rules in a simulation scenario reduces the probability of a collision event by more than a threshold probability, then the filtering rule can be considered effective. Specifically, for example, the threshold probability might be 10%.

[0141] Optionally, in the simulation scene update method provided in this application embodiment, if a certain type of parameter results in a high probability of collision, the process can be re-executed to determine the collision event between the vehicle and the obstacle and subsequent steps. For example, if a certain type of parameter causes a collision to occur more than or equal to 20 times per week, it can be determined that the parameter results in a high probability of collision.

[0142] Therefore, in this embodiment, the filtering rules for obstacles in the simulation scene are updated based on the algorithm logic corresponding to each item in the core features. This essentially achieves adaptive filtering, reducing labor costs while enabling seamless updating of the filtering rules.

[0143] Please see Figure 4 , Figure 4 A flowchart of a simulation scene update method provided in an embodiment of this application is shown. This simulation scene update method can be applied to electronic devices, specifically including steps S410 to S490.

[0144] Step S410: Is the obstacle in the current frame a new obstacle?

[0145] As described above, a simulation scene can correspond to multiple frames. Therefore, it can be determined whether the obstacle in the current frame is a new obstacle, that is, whether the obstacle in the current frame is appearing for the first time. If it is appearing for the first time, the process can proceed to step S420; if it is not appearing for the first time, the process can proceed to step S430.

[0146] Step S420: Initialize the fourth motion parameters of the obstacle using the first motion parameters of the first number of frames prior to the current frame.

[0147] Step S430: Update the fourth motion parameter based on the historical motion parameters of the obstacle to obtain the new fourth motion parameter.

[0148] If the obstacle is identified as a new obstacle, its fourth motion parameters can be initialized using the first motion parameters from a first number of frames preceding the current frame. Conversely, if the obstacle is not identified as a new obstacle, its fourth motion parameters can be updated based on its historical motion parameters to obtain new fourth motion parameters. These historical motion parameters may include the motion parameters from each frame preceding the current frame. For detailed descriptions, please refer to the foregoing embodiments.

[0149] Step S440: Predict the third motion parameters of the obstacle.

[0150] Step S450: The difference between the third motion parameter of the obstacle and the second motion parameter of the obstacle.

[0151] Furthermore, the third motion parameters of the obstacle can be predicted by combining the fourth motion parameter with the uniform acceleration model.

[0152] Then, the difference between the obstacle's third motion parameter and its second motion parameter is obtained. The second motion parameter may include the measured value of the obstacle in the current frame.

[0153] Step S460: Does the difference meet the target condition?

[0154] Then, it can be determined whether the difference meets the target condition. If the target condition is met, the process can proceed to step S480; if the target condition is not met, the process can proceed to step S470. Meeting the target condition can be used to characterize a large difference.

[0155] Step S470: Update model parameters.

[0156] Therefore, if the target conditions are not met, the model parameters can be updated. Specifically, the model parameters can be updated using the second motion parameters. These second motion parameters are the relevant parameters of the model used to predict the subsequent third motion parameters.

[0157] Alternatively, after updating the model parameters, you can return to step S410.

[0158] Step S480: Whether the specified number of times the second quantity is detected consecutively.

[0159] Step S490: Identify the obstacle as the target obstacle.

[0160] If the target conditions are met, it can also be determined whether a second number of specified situations are detected continuously. The specified situations include a first target situation or a second target situation. The first target situation includes a situation where at least one type of parameter has a difference greater than a difference threshold. The second target situation includes a situation where at least one type of parameter has a second motion parameter greater than its corresponding maximum value.

[0161] Therefore, if the specified number of consecutive detections for the second number of times is not detected, the process can proceed to step S470; if the specified number of consecutive detections for the second number of times is detected, the process can proceed to step S490 to determine the obstacle as the target obstacle.

[0162] For a detailed description of each of the above steps, please refer to the description in the foregoing embodiments, which will not be repeated here.

[0163] Please see Figure 5 , Figure 5 A flowchart of a simulation scene update method provided in an embodiment of this application is shown. This simulation scene update method can be applied to electronic devices, specifically including steps S510 to S570.

[0164] Step S510: Detect collision events.

[0165] First, collision detection can be performed. In some implementations, a collision event can be determined based on the vehicle's motion parameters in the simulation scenario and the updated motion parameters of the obstacles in the simulation scenario.

[0166] Step S520: Obtain the feature information of the collision event.

[0167] Therefore, characteristic information of the collision event can be obtained. This characteristic information includes obstacle characteristic information, vehicle characteristic information, and collision scene characteristic information.

[0168] Step S530: Whether a collision event occurred due to an error in the detection of the obstacle.

[0169] It is understandable that the cause of a collision event can be classified based on whether it is due to inaccurate perception of the obstacle. Therefore, it is possible to detect whether the collision event is caused by an obstacle detection error. If it is determined that the collision event is caused by an obstacle detection error, the process jumps to step S550; if it is determined that the collision event is not caused by an obstacle detection error, the process jumps to step S540.

[0170] Step S540: Record as normal collision.

[0171] If it is determined that the collision event was not caused by an error in the detection of the obstacle, the collision event can be recorded as a normal collision.

[0172] Step S550: Determine the core features of the collision obstacle.

[0173] If a collision event is determined to have occurred due to an obstacle detection error, the obstacle can be identified as the collision obstacle, thereby further determining the core features of the collision obstacle. These core features may include at least one of confidence level, velocity, acceleration, and position.

[0174] Step S560: Update the obstacle filtering rules for the simulation scene based on the algorithm logic corresponding to the core features.

[0175] Step S570: Apply to the next round of simulation.

[0176] Then, based on the algorithm logic corresponding to the core features, the filtering rules for obstacles in the simulation scene are updated. For example, the filtering rules may include filtering rules corresponding to confidence level, velocity, acceleration, and position. The updated filtering rules are then applied to the next round of simulation.

[0177] For a detailed description of each of the above steps, please refer to the foregoing embodiments; they will not be repeated here.

[0178] Please see Figure 6 , Figure 6 This illustration shows a structural block diagram of an electronic device according to an embodiment of this application. The electronic device 110 can be a smartphone, desktop computer, in-vehicle computer, server, or tablet computer, etc. Therefore, the guidance methods provided in the above embodiments of this application can be applied to the field of mobile devices such as mobile phones, and also to the field of intelligent mobile devices such as in-vehicle systems and drones; this application does not impose specific limitations.

[0179] The electronic device 110 in this application may include one or more of the following components: a processor 111, a memory 112, and one or more application programs, wherein the processor 111 is electrically connected to the memory 112, and the one or more programs are configured to perform the methods as described in the foregoing embodiments.

[0180] Processor 111 may include one or more processing cores. Processor 111 connects to various parts within the electronic device 110 using various interfaces and lines, and performs various functions and processes data of the electronic device 110 by running or executing instructions, programs, code sets, or instruction sets stored in memory 112, and by calling data stored in memory 112. Optionally, processor 111 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). Processor 111 may integrate one or more of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and computer programs; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into processor 111 and may be implemented separately through a communication chip. Specifically, the methods described in the foregoing embodiments can be executed by one or more processors 111.

[0181] In some implementations, memory 112 may include random access memory (RAM) or read-only memory (ROM). Memory 112 can be used to store instructions, programs, code, code sets, or instruction sets. Memory 112 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the various method embodiments described below, etc. The data storage area may also store data created by the electronic device 110 during use.

[0182] Please see Figure 7 This diagram illustrates a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. The computer-readable medium 700 stores program code that can be called by a processor to execute the methods described in the above method embodiments.

[0183] The computer-readable storage medium 700 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 700 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 700 has storage space for program code 710 that performs any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code 710 may be compressed, for example, in a suitable form.

[0184] Please refer to Figure 8 The diagram illustrates a structural block diagram of a computer program product according to an embodiment of this application. The computer program product 800 includes a computer program / instructions 810, which, when executed by a processor, implements the steps of the aforementioned method.

[0185] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 this application.

Claims

1. A simulation scene update method, characterized in that, include: Obtain the first motion parameters of each obstacle in the simulation scene for the first number of frames prior to the current frame, and the second motion parameters corresponding to the current frame; Predict the third motion parameters of the obstacle in the current frame based on the first motion parameters of each obstacle; Based on the difference between the second motion parameter and the third motion parameter of each obstacle, the target obstacle whose difference satisfies the target condition is identified; Remove the target obstacle from the simulation scene to update the simulation scene.

2. The method according to claim 1, characterized in that, The second motion parameter includes the second position information and / or second heading angle information of the obstacle; the third motion parameter includes the third position information and / or third heading angle information of the obstacle predicted based on the first motion parameters of each obstacle; the step of determining the target obstacle whose difference satisfies the target condition based on the difference between the second motion parameters and the third motion parameters of each obstacle includes: Obtain the difference thresholds corresponding to the position parameters and / or heading angle parameters of the obstacle; Acquire the difference between the second and third position information of the obstacle, and / or acquire the difference between the second and third heading angle information of the obstacle; When the difference between the second and third position information of an obstacle is greater than the difference threshold corresponding to the position parameter, and / or the difference between the second and third heading angle information of an obstacle is greater than the difference threshold corresponding to the heading angle parameter, the obstacle is determined to be a target obstacle.

3. The method according to claim 1 or 2, characterized in that, The second motion parameters include the second velocity and / or second acceleration of the obstacle, and the method further includes: Obtain the maximum values ​​of the velocity and / or acceleration parameters of the obstacle; When the second speed is detected to be greater than the maximum value of the speed parameter, and / or the second acceleration is greater than the maximum value of the acceleration parameter, the obstacle is determined to be a target obstacle.

4. The method according to claim 3, characterized in that, The obstacles include various types of obstacles, each type of obstacle corresponding to its own position parameter difference threshold, heading angle parameter difference threshold, maximum value of velocity parameter, and maximum value of acceleration parameter.

5. The method according to claim 1, characterized in that, After removing the target obstacle from the simulation scene to update the simulation scene, the method further includes: Based on the motion parameters of the vehicle and the motion parameters of the obstacles in the updated simulation scenario, a collision event in which the vehicle collides with the obstacle is determined. The feature information of the collision event is obtained, wherein the feature information includes obstacle feature information, vehicle feature information, and collision scene feature information; The core features of the colliding obstacle are determined based on the feature information of the collision event, wherein the core features include at least one of confidence, velocity, acceleration and position, and the colliding obstacle includes the obstacle that caused the collision event due to the detection error of the obstacle; Based on the algorithm logic corresponding to the core features, the filtering rules for obstacles in the simulation scene are updated.

6. The method according to claim 5, characterized in that, The core feature includes confidence level, and the algorithm logic based on the core feature updates the obstacle filtering rules for the simulation scene, including: Obtain the current confidence threshold, wherein the confidence threshold is used to characterize the critical threshold for filtering obstacles; Determine the first proportion of collision obstacles with a confidence level lower than the current confidence threshold; If the first proportion is greater than or equal to the first threshold, the current confidence threshold is increased.

7. The method according to claim 5, characterized in that, The core feature includes speed, and the algorithm logic based on the core feature updates the obstacle filtering rules for the simulation scene, including: Obtain the current speed threshold, wherein the speed threshold is used to characterize the critical speed for filtering the corresponding type of obstacle; Determine the second proportion of obstacles whose speed is greater than the current speed threshold; If the second percentage is greater than or equal to the second threshold, the current speed threshold is reduced.

8. The method according to claim 5, characterized in that, The core feature includes location, and the algorithm logic based on the core feature updates the obstacle filtering rules for the simulation scene, including: Determine the third proportion of collision obstacles whose positions are located in a specified position region, wherein the specified position region is used to characterize the position of filtering obstacles that meet the target size; If the third proportion is greater than or equal to the third threshold, the rule corresponding to the updated position is to filter out obstacles in the specified position area.

9. An electronic device, characterized in that, include: One or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to perform the method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores program code that can be invoked by a processor to execute the method as described in any one of claims 1-8.

Images