Vehicle simulation scene modeling method and device, computer device and storage medium

Abstract

This invention provides a method, apparatus, computer device, and storage medium for vehicle simulation scene modeling. The method includes acquiring static scene data and constructing a static scene model based on the static scene data; loading the static scene model and a vehicle kinematic model to construct an initial simulation model; acquiring dynamic environment data and performing coordinate transformation on the dynamic environment data based on the initial simulation model to obtain a dynamic environment model; acquiring dynamic master vehicle data and driving the vehicle based on the initial simulation model using the dynamic master vehicle data; detecting whether the vehicle speed has reached a preset speed, and when the vehicle speed reaches the preset speed, constructing a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed; and generating a dynamic scene model based on the dynamic environment model and the dynamic master vehicle model. Based on the dynamic scene model reconstruction using data playback, dynamic traffic scene reconstruction and fault scene reproduction are achieved, thereby realizing data-driven modeling of key scenarios for intelligent driving of vehicles.

Description

Technical Field

[0001] This invention relates to the technical field of simulation modeling, and in particular to a method, apparatus, computer equipment, and storage medium for modeling vehicle simulation scenes. Background Technology

[0002] With economic development, especially the automotive industry, the number of vehicles in society is increasing. As people's living standards improve, using cars as a means of transportation has become very common, and the requirements for vehicle configurations are also rising. Currently, vehicle driving methods are shifting from manned to autonomous driving, mainly achieved through autonomous driving algorithms. To reach the level of human drivers, these algorithms need to be refined with a large amount of data, requiring at least 17.7 billion kilometers of driving data. If a fleet of 100 autonomous driving test vehicles were deployed, conducting road tests 24 hours a day at an average speed of 25 miles per hour (40 kilometers per hour), it would take over 500 years to reach the target mileage. As can be seen, the time and cost involved are prohibitively high.

[0003] Among related technologies, simulation testing based on virtual scenarios is a crucial approach to addressing the high costs, long cycles, and low coverage of autonomous driving testing. By constructing virtual scenarios, closed-loop simulation testing of algorithms for autonomous driving perception, decision-making, planning, and control can be achieved. Specifically, simulation testing of autonomous vehicles first requires simulating a vehicle operation scenario consistent with the real world and then operating the autonomous vehicle within that scenario. However, when critical scenarios arise during autonomous driving, such as safety driver intervention or abnormal stopping, the ability to model the dynamic and static aspects of these scenarios and apply them to closed-loop data simulation testing becomes particularly important. Summary of the Invention

[0004] Therefore, it is necessary to provide a vehicle simulation scene modeling method, device, computer equipment, and storage medium to address the aforementioned technical problems.

[0005] A method for modeling a vehicle simulation scene, comprising:

[0006] Acquire static scene data and construct a static scene model based on the static scene data;

[0007] Load the static scene model and vehicle kinematics model to construct the initial simulation model;

[0008] Acquire dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model;

[0009] Acquire dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data;

[0010] Detect whether the vehicle's speed has reached a preset speed. When the vehicle's speed reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed.

[0011] A dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model.

[0012] In one embodiment, the step of acquiring static scene data and constructing a static scene model based on the static scene data includes:

[0013] Collect feature parameter data of preset road marking lines, and perform segmented fitting on the feature parameter data to obtain multiple pre-aiming paths, each pre-aiming path having multiple discrete points;

[0014] The discrete points of each segment of the pre-aimed path are integrated to form a discrete point sample set;

[0015] The road curve is obtained by globally fitting the discrete point sample set, and a static scene model is constructed based on the road curve.

[0016] In one embodiment, the step of acquiring static scene data and constructing a static scene model based on the static scene data further includes:

[0017] The feature parameter data is preprocessed;

[0018] The system detects whether the number of feature parameter data within a preset time is equal to a preset number. When the number of feature parameter data within the preset time is equal to the preset number, the system performs a global fitting on the discrete point sample set to obtain the road curve.

[0019] A static scene model is constructed based on the road curve.

[0020] In one embodiment, the step of acquiring dynamic environment data and performing coordinate transformation on the dynamic environment data based on the initial simulation model to obtain a dynamic environment model includes:

[0021] The dynamic environment data is acquired, and the dynamic environment data is arranged according to the acquisition time to obtain dynamic environment sequence data;

[0022] The dynamic environment sequence data is transformed using coordinates to generate dynamic environment element data;

[0023] The dynamic environment element data is input into the initial simulation model to generate multiple first simulation dynamic scenes;

[0024] The system sequentially checks whether each of the first simulated dynamic scenes matches the preset scene. If the first simulated dynamic scene matches the preset scene, then the first simulated dynamic scene is determined to be the first preparatory dynamic scene.

[0025] The dynamic environment model is constructed based on multiple first preliminary dynamic scenarios.

[0026] In one embodiment, the step of inputting the dynamic environment element data into the initial simulation model to generate multiple first dynamic simulation scenes includes:

[0027] The initial values ​​of the dynamic environment element data are input into the initial simulation model, and the vehicle is run in the initial simulation model.

[0028] The system detects whether the vehicle state has reached a preset state. When the vehicle state reaches the preset state, it acquires the vehicle's state data in the initial simulation model, thereby generating multiple first simulation dynamic scenarios.

[0029] In one embodiment, after the step of sequentially detecting whether each of the first simulated dynamic scenes matches a preset scene, the method further includes:

[0030] If the first simulated dynamic scene does not match the preset scene, the dynamic environment element data is determined to be supplementary element data, and the supplementary element data is added.

[0031] The supplemented dynamic environment element data is input into the initial simulation model to generate multiple second simulation dynamic scenes;

[0032] The system sequentially checks whether each of the second simulated dynamic scenes matches the preset scene. If the second simulated dynamic scene matches the preset scene, the dynamic environment model is constructed based on the multiple second simulated dynamic scenes and the first preparatory dynamic scene.

[0033] In one embodiment, the step of acquiring dynamic master vehicle data and driving the vehicle based on the initial simulation model using the dynamic master vehicle data includes:

[0034] The dynamic master vehicle data is acquired, and the dynamic master vehicle data is arranged according to the acquisition time to obtain dynamic master vehicle sequence data;

[0035] A vehicle acceleration model is established based on the dynamic master vehicle sequence data;

[0036] The vehicle is driven by the vehicle acceleration model.

[0037] A vehicle simulation scene modeling device, comprising:

[0038] The static data acquisition module is used to acquire static scene data;

[0039] The static model building module is used to build static scene models based on static scene data.

[0040] The initial model building module is used to load the static scene model and the vehicle kinematics model to build the initial simulation model;

[0041] The dynamic data acquisition module is used to acquire dynamic environmental data and dynamic main vehicle data;

[0042] The dynamic model building module is used to build a dynamic scene model based on the initial simulation model and the dynamic environment data and the dynamic main vehicle data.

[0043] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to perform the following steps:

[0044] Acquire static scene data and construct a static scene model based on the static scene data;

[0045] Load the static scene model and vehicle kinematics model to construct the initial simulation model;

[0046] Acquire dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model;

[0047] Acquire dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data;

[0048] Detect whether the vehicle's speed has reached a preset speed. When the vehicle's speed reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed.

[0049] A dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model.

[0050] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0051] Acquire static scene data and construct a static scene model based on the static scene data;

[0052] Load the static scene model and vehicle kinematics model to construct the initial simulation model;

[0053] Acquire dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model;

[0054] Acquire dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data;

[0055] Detect whether the vehicle's speed has reached a preset speed. When the vehicle's speed reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed.

[0056] A dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model.

[0057] The aforementioned vehicle simulation scenario modeling method, apparatus, computer equipment, and storage medium acquire static scene data, construct a static scene model based on the static scene data, and load the static scene model and vehicle kinematic model into the simulation platform to construct the initial simulation model. Furthermore, dynamic environment data and dynamic master vehicle data are acquired. Based on the initial simulation model, a dynamic environment model is constructed based on the dynamic environment data, and a dynamic master vehicle model is constructed based on the dynamic master vehicle data. Finally, a dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model. The dynamic scene model is reconstructed based on the playback of radar recognition result data, realizing the reconstruction of dynamic traffic scenes and the reproduction of fault scenes. This enables data-driven modeling of key scenarios in intelligent driving, improves the realism of scene modeling, and expands the effective data sources for intelligent driving simulation test scenarios. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating a vehicle simulation scene modeling method in one embodiment.

[0059] Figure 2 A structural block diagram of a vehicle simulation scene modeling device in one embodiment;

[0060] Figure 3 This is an internal structural diagram of a computer device in one embodiment;

[0061] Figure 4 A flowchart illustrating the construction process of a static scene model in one embodiment;

[0062] Figure 5 A schematic diagram illustrating the process of constructing the initial simulation model in one embodiment;

[0063] Figure 6 A flowchart illustrating the construction process of a dynamic environment model in one embodiment;

[0064] Figure 7 This is a schematic diagram illustrating the process of constructing a dynamic master vehicle model in one embodiment. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0066] Example 1

[0067] In this embodiment, as Figure 1 As shown, a method for modeling a vehicle simulation scene is provided, which includes:

[0068] Step 110: Obtain static scene data and construct a static scene model based on the static scene data.

[0069] In this embodiment, static scene data refers to environmental data in the scenario of vehicle travel, such as road width, road markings, road length, and curvature. Static scene data is primarily obtained from road markings collected during vehicle travel, including parameters such as the position, length, width, and curvature of the road markings. In practical applications, monitoring devices such as cameras can be installed directly in front of the vehicle, along with positioning sensors, so that the cameras continuously collect feature parameter data from different positions of the road markings as the vehicle travels on the road, generating static scene data. This static scene data is then input into a simulation platform to construct a static scene model. In some embodiments, static scene data can be acquired using high-precision maps, LiDAR, or other sensors. In this embodiment, static scene data is collected by a camera and used to construct a static scene model, which is then used to reconstruct the road markings on a simulation platform. The reconstruction of the static scene model based on the playback of camera recognition data reduces the reliance on high-precision maps or preset static scene models found in existing technologies and achieves static scene reproduction.

[0070] Step 120: Load the static scene model and vehicle kinematics model to construct the initial simulation model.

[0071] After constructing the static scene model, the static scene model and the vehicle kinematics model are loaded onto the simulation platform, allowing the simulation platform to initialize and build the initial simulation model. In this embodiment, the vehicle kinematics model describes the motion characteristics of the vehicle in a specific coordinate system. Depending on the application scenario and requirements, different coordinate systems can be selected, such as the global coordinate system or the road coordinate system. In the field of autonomous driving, the vehicle kinematics model is used to plan and predict the vehicle's trajectory, as well as control the vehicle's driving under various road and traffic conditions. This is existing technology, so it will not be elaborated on here.

[0072] Step 130: Obtain dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model.

[0073] In this embodiment, the master vehicle is the vehicle currently in an autonomous driving state for which a simulation model needs to be built, and the slave vehicles refer to other vehicles in the same road or traffic flow as the master vehicle. It can be understood that dynamic environment data refers to vehicle parameter data collected from slave vehicles during operation. Dynamic environment data is obtained from slave vehicles during operation. This dynamic environment data includes real-time data such as the slave vehicle's position, speed, acceleration, steering angle, and roll angle, as well as vehicle state estimation data (such as perception and judgment of the environment). Dynamic environment data can be acquired through monitoring equipment. When acquiring dynamic environment data, devices such as LiDAR, GPS positioning systems, cameras, and sensors can be used to detect slave vehicles during operation to obtain dynamic environment data. For example, if the master vehicle scrapes against a taxi, this is considered a fault scenario. In this scenario, the taxi and other vehicles in the traffic flow are slave vehicles. Obtaining the parameter data of the slave vehicles constitutes the dynamic environment data to be acquired. After obtaining the dynamic environment data, based on the constructed initial simulation model, the dynamic environment data is subjected to coordinate transformation on the simulation platform. The dynamic environment data after coordinate transformation is input into the slave vehicle in the simulation platform, so that the slave vehicle runs on the simulation platform. At this time, the dynamic environment model is obtained according to the parameters of the slave vehicle.

[0074] Step 140: Obtain dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data.

[0075] In this embodiment, dynamic master vehicle data refers to vehicle parameter data collected from the master vehicle in an autonomous driving state during operation. This dynamic master vehicle data, obtained from the master vehicle in motion, includes real-time data such as the master vehicle's position, speed, acceleration, steering angle, and roll angle, as well as vehicle state estimation data (such as environmental perception and judgment). Dynamic master vehicle data can be acquired through monitoring equipment. When acquiring dynamic master vehicle data, vehicle state parameters are collected based on the fusion data of the vehicle's Global Navigation Satellite System (GNSS) and Inertial Navigation System (INS), thereby obtaining the dynamic master vehicle data. It should be noted that the fusion data of GNSS and INS can provide high-precision positioning and navigation information. In the field of autonomous driving, the fusion of GNSS and INS can provide more accurate parking information, helping vehicles to perform high-precision navigation in parking lots or complex urban environments. Furthermore, during the operation of the autonomous vehicle, this fused data can also provide real-time information such as the vehicle's speed, direction, and acceleration, assisting the vehicle in making autonomous decisions and control.

[0076] In this embodiment, after obtaining the dynamic master vehicle data, the vehicle is driven based on the constructed initial simulation model and the dynamic master vehicle data, so that the vehicle performs kinematic simulation on the simulation platform. It should be noted that the vehicle here is the master vehicle being simulated.

[0077] Step 150: Detect whether the speed of the vehicle has reached a preset speed. When the speed of the vehicle reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed.

[0078] In this embodiment, after obtaining dynamic master vehicle data, the master vehicle is driven based on the dynamic master vehicle data, enabling kinematic simulation of the master vehicle on the simulation platform. It should be understood that during this process, the speed of the master vehicle needs to be detected. When the speed of the master vehicle reaches a preset speed, a dynamic master vehicle model can be constructed based on the parameters of the master vehicle that has reached the preset speed. When collecting dynamic master vehicle data from the master vehicle, if the actual collected data is truncated, initial states such as dynamic information, typically represented by speed, may not be instantly reached at the start of initialization and cannot be directly input into the simulation platform. Therefore, it is necessary to detect whether the master vehicle has reached the preset speed required by the simulation platform. When the speed of the master vehicle reaches the preset speed, a dynamic master vehicle model can be constructed based on the parameters of the master vehicle that has reached the preset speed. However, if the speed of the master vehicle has not reached the preset speed, in step 140, the master vehicle needs to be driven based on the initial simulation model using dynamic master vehicle data until the speed of the master vehicle reaches the preset speed, and then a dynamic master vehicle model is constructed based on the parameters of the vehicle that has reached the preset speed.

[0079] Step 160: Generate a dynamic scene model based on the dynamic environment model and the dynamic master vehicle model.

[0080] In this embodiment, after obtaining the dynamic environment model and the dynamic master vehicle model, the dynamic environment model and the dynamic master vehicle model are loaded onto the simulation platform to generate a dynamic scene model. An initial simulation model is constructed based on the static scene model and the vehicle kinematics model. Based on the initial simulation model, dynamic environment data is collected using LiDAR and used to construct a dynamic environment model, thereby reconstructing the vehicle on the simulation platform. Simultaneously, dynamic master vehicle data is collected and used to construct a dynamic master vehicle model. Based on the dynamic scene model reconstruction using radar recognition result data playback, dynamic traffic scene reconstruction and fault scene reproduction are achieved. This enables data-driven modeling of key intelligent driving scenarios, improves the realism of scene modeling, and expands the effective data sources for intelligent driving simulation test scenarios.

[0081] In the above embodiments, static scene data is acquired, and a static scene model is constructed based on this data. After obtaining the static scene model, it and the vehicle kinematics model are loaded into the simulation platform to construct the initial simulation model. Dynamic environment data and dynamic master vehicle data are acquired. Based on the initial simulation model, coordinate transformation is performed on the dynamic environment data to obtain the dynamic environment model. Based on the initial simulation model, the master vehicle is driven according to the dynamic master vehicle data, and it is detected whether the master vehicle reaches the preset speed required by the simulation platform. When the master vehicle reaches the preset speed, a dynamic master vehicle model can be constructed based on the parameters of the master vehicle that has reached the preset speed. Finally, a dynamic scene model is generated based on the constructed dynamic environment model and dynamic master vehicle model. The reconstruction of the dynamic scene model based on the playback of radar recognition result data realizes the reconstruction of dynamic traffic scenes and the reproduction of fault scenes, thereby achieving data-driven modeling of key intelligent driving scenarios, improving the realism of scene modeling, and expanding the effective data sources for intelligent driving simulation test scenarios.

[0082] In one embodiment, the step of acquiring static scene data and constructing a static scene model based on the static scene data includes:

[0083] Collect feature parameter data of preset road marking lines, and perform segmented fitting on the feature parameter data to obtain multiple pre-aiming paths, each pre-aiming path having multiple discrete points;

[0084] The discrete points of each segment of the pre-aimed path are integrated to form a discrete point sample set;

[0085] The road curve is obtained by globally fitting the discrete point sample set, and a static scene model is constructed based on the road curve.

[0086] In this embodiment, the feature parameter data of road markings can be collected by installing a camera directly in front of the vehicle and fusing it with positioning sensors. This allows the camera to collect data from different positions of the road markings as the vehicle travels on the road, and then calculate the feature parameter data after recognition. Alternatively, it can be considered as the process of using sensors and algorithms to scan and detect road markings during autonomous driving to obtain their feature parameter data and predict the vehicle's driving path. For each coordinate position of the main vehicle and the corresponding feature parameter data of the road markings identified by the camera, the road markings are segmented and fitted at preset distances to obtain multiple pre-planning paths. Segmented fitting divides the feature parameter data into multiple segments, and then fits each segment. Its main purpose is to fit a set of discrete data points using a piecewise function, ensuring that the fitting result best matches the characteristics of the original data. Therefore, each pre-planning path has multiple discrete data points (discrete points). The discrete points of all pre-planning paths are then integrated to form a discrete point sample set for the static scene model. Finally, a global fitting is performed on the discrete point sample set to obtain the road curve. Global fitting refers to a data fitting technique that can simultaneously perform a holistic fitting of a set of data. Specifically, global fitting first requires merging multiple datasets into a single dataset, and then using certain mathematical methods to fit this dataset holistically, resulting in a fitting function or model. Specifically, global fitting of a discrete point sample set can be performed using a smooth spline curve fitting method to fit the discrete sample point set and thus obtain the road curve. It should be noted that before performing piecewise fitting of the feature parameter data, coordinate transformation is required. Since the camera collects feature parameter data based on its own coordinate system (with the camera's location as the origin), the collected data represents the relative position of road markings. Therefore, before piecewise fitting, the feature parameter data collected by the camera needs to be transformed into a vehicle coordinate system (with the front or rear axle of the vehicle as the origin). This transformation can be performed using existing mathematical formulas.

[0087] In this embodiment, after obtaining the road curves, a static scene model is constructed based on the road curves using the OPENDRIVE format. The OPENDRIVE format is a format for describing the static road traffic network required for autonomous driving simulation applications. This format mainly covers the description of modeling elements such as roads, lanes, and intersections.

[0088] In one embodiment, the step of acquiring static scene data and constructing a static scene model based on the static scene data further includes:

[0089] The feature parameter data is preprocessed;

[0090] The system detects whether the number of feature parameter data within a preset time is equal to a preset number. When the number of feature parameter data within the preset time is equal to the preset number, the system performs a global fitting on the discrete point sample set to obtain the road curve.

[0091] A static scene model is constructed based on the road curve.

[0092] In this embodiment, after collecting the feature parameter data of the preset road marking lines through the camera, the feature parameter data needs to be preprocessed. Preprocessing includes: filtering the feature parameter data, and then cleaning the filtered feature parameter data by deduplication, outlier removal, downsampling, etc., to eliminate data errors and noise and improve the accuracy and reliability of the data.

[0093] It should be noted that the filtering process requires filtering feature parameter data based on the camera's valid flag bits. Before filtering the feature parameter data based on the camera's valid flag bits, the reliability of the feature parameter data needs to be assessed. When assessing data reliability, firstly, the data source needs to be verified: the reliability of the data source needs to be confirmed; secondly, the data quality needs to be checked: the quality and clarity of the data need to be assessed; then, data bias needs to be considered: whether there are any biases that may affect the data reliability needs to be assessed. Next, other data sources need to be compared: if possible, the reliability can be assessed by comparing data from other sources. Finally, statistical methods can be applied: some statistical methods can be used to assess the reliability of the data. It should be understood that the process of filtering feature parameter data based on the camera's valid flag bits can include the following steps: First, data collection: image or video data is acquired from the camera, and the valid flag bit information for each data point is recorded. Second, filtering criteria are determined: based on the application scenario and objective, the criteria for filtering feature parameters are determined. These criteria can include the range, trend of change, correlation, etc. of the parameters. Then, feature parameters are extracted: each data point is preprocessed to extract feature parameters related to the target. These parameters can include color, texture, shape, etc. Next, feature parameters are filtered: each feature parameter is filtered according to the valid flag and filtering criteria. For example, if the valid flag indicates that the camera view corresponding to the data point is partially obstructed, then some feature parameters of that data point can be ignored. Finally, the filtered data is processed: the filtered feature parameters are processed to achieve the application goal. For example, the filtered feature parameters can be used for object recognition, behavior analysis, scene understanding, etc.

[0094] In this embodiment, during the preprocessing of feature parameter data, if factors such as backlighting or blurred lane lines are found to cause partial loss of feature parameter data, it is necessary to detect whether the preprocessed feature parameter data is missing. Specifically, by detecting whether the number of feature parameter data within a preset time is equal to a preset number, if the number of feature parameter data within the preset time is equal to the preset number, a global fitting is performed on the discrete point sample set to obtain the road curve, and a static scene model is constructed based on the road curve according to the OPENDRIVE format. However, if the number of feature parameter data within the preset time is not equal to the preset number, multi-sensor data interpolation and abnormal scene reconstruction are performed. First, the feature parameter data within the preset time is interpolated and supplemented based on the existing feature parameter data. Then, the positioning sensor data is fused. For road marking lines with missing data, the interpolated data is corrected using the vehicle driving path coordinate data as a reference path to ensure that the path filling fitting error is within the allowable range. Subsequently, for road marking lines with missing data, the cause analysis of the scene can be referenced to construct abnormal scene models such as blurred or missing lane lines and rut interference in the simulation scene to restore the actual static scene in the static scene model. After interpolating and correcting the feature parameter data, the supplemented feature parameter data is used for global fitting to obtain the road curve. Then, a static scene model is constructed based on the road curve according to the OPENDRIVE format.

[0095] like Figure 6 As shown, in one embodiment, the step of acquiring dynamic environment data and performing coordinate transformation on the dynamic environment data based on the initial simulation model to obtain a dynamic environment model includes:

[0096] The dynamic environment data is acquired, and the dynamic environment data is arranged according to the acquisition time to obtain dynamic environment sequence data;

[0097] The dynamic environment sequence data is transformed using coordinates to generate dynamic environment element data;

[0098] The dynamic environment element data is input into the initial simulation model to generate multiple first simulation dynamic scenes;

[0099] The system sequentially checks whether each of the first simulated dynamic scenes matches the preset scene. If the first simulated dynamic scene matches the preset scene, then the first simulated dynamic scene is determined to be the first preparatory dynamic scene.

[0100] The dynamic environment model is constructed based on multiple first preliminary dynamic scenarios.

[0101] In this embodiment, after acquiring dynamic environment data via LiDAR, the dynamic environment data is arranged according to a time series based on the simulation platform model. This time series can be the acquisition time. A nearest neighbor data association algorithm is used to preprocess the dynamic environment data, generating dynamic environment sequence data. After obtaining the dynamic environment sequence data, coordinate transformation is performed on the dynamic environment sequence data according to the global coordinate system, and dynamic environment element data is generated according to the OpenScenario standard. Here, the global coordinate system is the vehicle's coordinate system. Then, the dynamic environment element data is input into the initial simulation model to generate the first simulated dynamic scene. Each first simulated dynamic scene is then checked against a preset scene. If a first simulated dynamic scene matches a preset scene, it is determined as the first preliminary dynamic scene. A dynamic environment model is constructed based on multiple first preliminary dynamic scenes to recreate the actual dynamic scene within the dynamic environment model. For example, in reality, an autonomous vehicle and a taxi collide on the road. The autonomous vehicle is the master vehicle, and the taxi is the slave vehicle. Parameter data from the slave vehicle is collected as dynamic environment data and input into the initial simulation model to generate the first simulated dynamic scene. If the simulation on the simulation platform shows that the first simulated dynamic scene matches the preset scene (a collision between an autonomous vehicle and a taxi on the road), then the first simulated dynamic scene is determined to be the first preparatory dynamic scene.

[0102] In one embodiment, the step of inputting the dynamic environment element data into the initial simulation model to generate multiple first dynamic simulation scenarios includes:

[0103] The initial values ​​of the dynamic environment element data are input into the initial simulation model, and the vehicle is run in the initial simulation model.

[0104] The system detects whether the vehicle state has reached a preset state. When the vehicle state reaches the preset state, it acquires the vehicle's state data in the initial simulation model, thereby generating multiple first simulation dynamic scenarios.

[0105] In this embodiment, when collecting dynamic environmental data, if the actual collected data is truncated, the initial state of some dynamic information, such as speed, may not be reached instantaneously at the start of initialization and cannot be directly input into the simulation platform. Therefore, it is necessary to input the dynamic environmental element data into the simulation initial model. Specifically, a scene initialization road segment can be added to the static scene model. First, the initial values ​​of the dynamic environmental element data are input into the simulation initial model, and the initial values ​​of the dynamic environmental element data are used as the control target to perform pre-preparation processing on the vehicle in the initialization road segment. Here, the pre-preparation processing involves setting the starting position and attitude of the vehicle in the simulation scene. After the vehicle is pre-prepared, it is necessary to check whether the vehicle state has reached the preset state. When the vehicle state reaches the preset state, the control of the vehicle is switched back to using the dynamic environmental element data as the control target, and the state data of the vehicle in the simulation initial model is obtained, thereby generating multiple first simulation dynamic scenes and realizing the restoration of the actual dynamic scene in the dynamic environment model.

[0106] In one embodiment, after the step of sequentially detecting whether each of the first simulated dynamic scenes matches a preset scene, the method further includes:

[0107] If the first simulated dynamic scene does not match the preset scene, the dynamic environment element data is determined to be supplementary element data, and the supplementary element data is added.

[0108] The supplemented dynamic environment element data is input into the initial simulation model to generate multiple second simulation dynamic scenes;

[0109] The system sequentially checks whether each of the second simulated dynamic scenes matches the preset scene. If the second simulated dynamic scene matches the preset scene, the dynamic environment model is constructed based on the multiple second simulated dynamic scenes and the first preparatory dynamic scene.

[0110] In this embodiment, when the first simulated dynamic scene does not match the preset scene, multi-source data supplementation is required for the dynamic environment element data. First, the dynamic environment element data is determined to be the element data to be supplemented. The data to be supplemented is then supplemented, and the supplemented dynamic environment element data is input into the initial simulation model to generate multiple second simulated dynamic scenes. Then, each second simulated dynamic scene is sequentially checked to see if it matches the preset scene. When the second simulated dynamic scene matches the preset scene, a dynamic environment model is constructed based on the multiple second simulated dynamic scenes and the first preset dynamic scene, thereby recreating the actual dynamic scene in the dynamic environment model.

[0111] like Figure 7As shown, in one embodiment, the step of acquiring dynamic master vehicle data and driving the vehicle based on the initial simulation model using the dynamic master vehicle data includes:

[0112] The dynamic master vehicle data is acquired, and the dynamic master vehicle data is arranged according to the acquisition time to obtain dynamic master vehicle sequence data;

[0113] A vehicle acceleration model is established based on the dynamic master vehicle sequence data;

[0114] The vehicle is driven by the vehicle acceleration model.

[0115] In this embodiment, after acquiring dynamic master vehicle data, based on the initial simulation model, the dynamic master vehicle data is arranged according to the acquisition time. Preprocessing steps such as data cleaning, data format consistency processing, and missing data fitting interpolation are performed on the dynamic master vehicle data to generate dynamic master vehicle sequence data. A vehicle acceleration model is established based on the dynamic master vehicle sequence data to drive the vehicle kinematic model in the initial simulation model. Based on the initial simulation model, the master vehicle is switched to manual driving mode, and the vehicle acceleration model drives the master vehicle to achieve kinematic simulation in the simulation platform. Subsequently, it is detected whether the master vehicle's speed has reached a preset speed. When the master vehicle's speed reaches the preset speed, the master vehicle is switched to autonomous driving mode. A dynamic master vehicle model is constructed based on the parameters of the master vehicle in autonomous driving mode. Finally, a dynamic scene model is constructed based on the dynamic environment model and the dynamic master vehicle model. It should be noted that if the master vehicle's speed has not reached the preset speed, the vehicle acceleration model drives the master vehicle to perform kinematic simulation until the master vehicle's speed reaches the preset speed.

[0116] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0117] Example 2

[0118] This embodiment provides a vehicle simulation scene modeling method, including static scene restoration and dynamic scene reconstruction.

[0119] like Figure 4As shown, firstly, static scene restoration: To address the issue of the authenticity of static scene restoration in the absence of high-precision maps, this invention proposes a static scene restoration modeling method based on camera data playback, and on this basis, constructs a simulation platform initialization model.

[0120] The process of restoring a static scene mainly involves the following steps:

[0121] Data Acquisition and Preprocessing: A camera sensor is installed directly in front of the vehicle. Combined with positioning sensor data, the system collects feature parameter data from different camera positions as the vehicle travels on open roads, focusing on the identification of specific road markings. The feature parameter data is filtered based on the effective marker positions of the cameras. Then, the data undergoes cleaning processes such as deduplication, outlier removal, and downsampling to eliminate errors and noise, improving accuracy and reliability. The data preprocessing process requires an assessment of data reliability.

[0122] Local path segmentation fitting and global fitting: For each coordinate position of the vehicle and the feature parameter data of the road markings identified by the corresponding camera, the road markings are segmented and fitted at a preset distance to obtain multiple pre-aimed paths. Discrete points of each pre-aimed path are obtained. All segmented fitting results are integrated to establish a discrete point sample set for the static scene model. The road curve is obtained by fitting the discrete point sample set with a smooth spline curve fitting method. Then, the static scene model is constructed based on the road curve according to the OPENDRIVE format.

[0123] Multi-source data supplementation: Specifically, during data cleaning, it was discovered that some feature parameter data was missing due to factors such as backlighting and blurred lane lines. The following solutions are proposed to address this issue: multi-sensor data interpolation and anomalous scene reconstruction. First, interpolation is performed based on existing feature parameter data. Then, positioning sensor data is fused. For road sections with missing data, vehicle travel path coordinate data is used as a reference path to correct the interpolated data, ensuring that the path filling fitting error is within acceptable limits. Then, for road sections with missing data, based on on-site cause analysis, anomalous scene models such as blurred or missing lane lines and rut interference are constructed in the simulation scenario to recreate the actual static scene.

[0124] like Figure 5 As shown, the initialization model construction for the simulation platform mainly involves loading the reconstructed static scene model and vehicle motion / dynamics model based on the existing simulation platform, thereby allowing the simulation platform to enter the initialization phase.

[0125] Subsequently, the dynamic scene was recreated:

[0126] Dynamic scene restoration includes two parts: dynamic environmental scene restoration and dynamic main vehicle scene restoration.

[0127] like Figure 6 As shown, for the dynamic environment scene reconstruction part, the initial scene construction based on the above simulation platform is used as the foundation. First, radar identification dynamic environment data is collected, and the dynamic environment data of each frame is arranged according to the time series. An improved nearest neighbor data association algorithm is used to preprocess the dynamic environment data to generate dynamic environment sequence data. After the dynamic environment sequence data is transformed according to the global coordinate system, dynamic environment element data is generated according to the OpenScenario standard.

[0128] In particular, to address the issue of data truncation in actual data collection, where initial states of dynamic information, such as speed, may not be instantly available at the start of initialization and therefore cannot be directly input into the simulation system, an initialization construction scheme needs to be established. Specifically, a scene initialization section can be added to the static road reconstruction. Before the actual data is input, the initial values ​​of the dynamic environmental element data need to be used as the control target, and the simulation vehicle needs to be pre-prepared in the initialization section. Once the simulation vehicle's state meets the initialization requirements of the dynamic environmental element data, the simulation vehicle switches to using the dynamic environmental element data as the control target, thus achieving dynamic scene reconstruction.

[0129] like Figure 7 As shown, the reconstruction of the dynamic main vehicle model is also based on the initialization scenario construction of the simulation platform mentioned above, with data-driven simulation as the main technical means.

[0130] First, based on the fusion data of the real vehicle's Global Navigation Satellite System (GNSS) and Inertial Navigation System, vehicle state parameters are collected as dynamic master vehicle data. The dynamic master vehicle data is then preprocessed, including data cleaning, data format consistency processing, and missing data fitting interpolation, to obtain dynamic master vehicle sequence data in the scene coordinate system. A data-driven vehicle acceleration model is then established to drive the simulation platform to initialize the vehicle kinematic model in the scene model.

[0131] Then, the scenario model is initialized based on the simulation platform, and the vehicle is switched to manual driving mode. Data-driven simulation is achieved by importing external data, similar to manual driving. Subsequently, it is determined whether the vehicle speed at the main target point meets the requirements. If the requirements are met, the vehicle is switched to autonomous driving mode, indicating that the simulation platform has successfully modeled the dynamic and static scenarios when key events occur.

[0132] Example 3

[0133] In this embodiment, as Figure 2 As shown, a vehicle simulation scene modeling device is provided, comprising:

[0134] The static data acquisition module 310 is used to acquire static scene data;

[0135] Static model building module 320 is used to build a static scene model based on static scene data;

[0136] The initial model building module 330 is used to load the static scene model and the vehicle kinematics model to build the initial simulation model;

[0137] The dynamic data acquisition module 340 is used to acquire dynamic environmental data and dynamic main vehicle data;

[0138] The dynamic model building module 350 is used to build a dynamic scene model based on the initial simulation model and the dynamic environment data and the dynamic main vehicle data.

[0139] In this embodiment, the static data acquisition module 310 is used to acquire feature parameter data of road markings. This feature parameter data can be acquired through monitoring devices such as radar and cameras. After acquiring the static scene data, the static scene data is filtered and cleaned to eliminate data errors and noise, thereby improving the accuracy and reliability of the data. The static data acquisition module 310 inputs the static scene data into the static model construction module 320, which then constructs the static scene model. Subsequently, the initial model construction module 330 loads the static scene model and the vehicle kinematics model into the simulation platform, thereby constructing the initial simulation model.

[0140] The dynamic data acquisition module 340 acquires dynamic environment data, arranges the dynamic environment data according to time series, and preprocesses the dynamic environment data to generate dynamic environment sequence data. The dynamic environment sequence data is then transformed using a global coordinate system to generate dynamic environment element data. Additionally, the dynamic data acquisition module 340 acquires dynamic master vehicle data, arranges the dynamic master vehicle data according to acquisition time, and preprocesses the dynamic master vehicle data to generate dynamic master vehicle sequence data. Then, the dynamic data acquisition module 340 inputs the dynamic environment element data and the dynamic master vehicle sequence data into the dynamic model construction module 350, which then constructs the dynamic scene model.

[0141] Specific limitations regarding the vehicle simulation scene modeling device can be found in the limitations of the vehicle simulation scene modeling method described above, and will not be repeated here. Each unit in the aforementioned vehicle simulation scene modeling device can be implemented entirely or partially through software, hardware, or a combination thereof. These units can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each unit.

[0142] Example 4

[0143] In this embodiment, a computer device is provided. Its internal structure diagram can be shown as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs, and also deploys a database for storing static scene data, dynamic environment data, and dynamic vehicle data. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with other computer devices that have deployed application software. When the computer program is executed by the processor, it implements a vehicle simulation scene modeling method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0144] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0145] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to perform the following steps:

[0146] Acquire static scene data and construct a static scene model based on the static scene data;

[0147] Load the static scene model and vehicle kinematics model to construct the initial simulation model;

[0148] Acquire dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model;

[0149] Acquire dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data;

[0150] Detect whether the vehicle's speed has reached a preset speed. When the vehicle's speed reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed.

[0151] A dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model.

[0152] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0153] Collect feature parameter data of preset road marking lines, and perform segmented fitting on the feature parameter data to obtain multiple pre-aiming paths, each pre-aiming path having multiple discrete points;

[0154] The discrete points of each segment of the pre-aimed path are integrated to form a discrete point sample set;

[0155] The road curve is obtained by globally fitting the discrete point sample set, and a static scene model is constructed based on the road curve.

[0156] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0157] The feature parameter data is preprocessed;

[0158] The system detects whether the number of feature parameter data within a preset time is equal to a preset number. When the number of feature parameter data within the preset time is equal to the preset number, the system performs a global fitting on the discrete point sample set to obtain the road curve.

[0159] A static scene model is constructed based on the road curve.

[0160] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0161] The dynamic environment data is acquired, and the dynamic environment data is arranged according to the acquisition time to obtain dynamic environment sequence data;

[0162] The dynamic environment sequence data is transformed using coordinates to generate dynamic environment element data;

[0163] The dynamic environment element data is input into the initial simulation model to generate multiple first simulation dynamic scenes;

[0164] The system sequentially checks whether each of the first simulated dynamic scenes matches the preset scene. If the first simulated dynamic scene matches the preset scene, then the first simulated dynamic scene is determined to be the first preparatory dynamic scene.

[0165] The dynamic environment model is constructed based on multiple first preliminary dynamic scenarios.

[0166] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0167] The initial values ​​of the dynamic environment element data are input into the initial simulation model, and the vehicle is run in the initial simulation model.

[0168] The system detects whether the vehicle state has reached a preset state. When the vehicle state reaches the preset state, it acquires the vehicle's state data in the initial simulation model, thereby generating multiple first simulation dynamic scenarios.

[0169] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0170] If the first simulated dynamic scene does not match the preset scene, the dynamic environment element data is determined to be supplementary element data, and the supplementary element data is added.

[0171] The supplemented dynamic environment element data is input into the initial simulation model to generate multiple second simulation dynamic scenes;

[0172] The system sequentially checks whether each of the second simulated dynamic scenes matches the preset scene. If the second simulated dynamic scene matches the preset scene, the dynamic environment model is constructed based on the multiple second simulated dynamic scenes and the first preparatory dynamic scene.

[0173] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0174] The dynamic master vehicle data is acquired, and the dynamic master vehicle data is arranged according to the acquisition time to obtain dynamic master vehicle sequence data;

[0175] A vehicle acceleration model is established based on the dynamic master vehicle sequence data;

[0176] The vehicle is driven by the vehicle acceleration model.

[0177] Example 5

[0178] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps:

[0179] Acquire static scene data and construct a static scene model based on the static scene data;

[0180] Load the static scene model and vehicle kinematics model to construct the initial simulation model;

[0181] Acquire dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model;

[0182] Acquire dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data;

[0183] Detect whether the vehicle's speed has reached a preset speed. When the vehicle's speed reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed.

[0184] A dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model.

[0185] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0186] Collect feature parameter data of preset road marking lines, and perform segmented fitting on the feature parameter data to obtain multiple pre-aiming paths, each pre-aiming path having multiple discrete points;

[0187] The discrete points of each segment of the pre-aimed path are integrated to form a discrete point sample set;

[0188] The road curve is obtained by globally fitting the discrete point sample set, and a static scene model is constructed based on the road curve.

[0189] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0190] The feature parameter data is preprocessed;

[0191] The system detects whether the number of feature parameter data within a preset time is equal to a preset number. When the number of feature parameter data within the preset time is equal to the preset number, the system performs a global fitting on the discrete point sample set to obtain the road curve.

[0192] A static scene model is constructed based on the road curve.

[0193] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0194] The dynamic environment data is acquired, and the dynamic environment data is arranged according to the acquisition time to obtain dynamic environment sequence data;

[0195] The dynamic environment sequence data is transformed using coordinates to generate dynamic environment element data;

[0196] The dynamic environment element data is input into the initial simulation model to generate multiple first simulation dynamic scenes;

[0197] The system sequentially checks whether each of the first simulated dynamic scenes matches the preset scene. If the first simulated dynamic scene matches the preset scene, then the first simulated dynamic scene is determined to be the first preparatory dynamic scene.

[0198] The dynamic environment model is constructed based on multiple first preliminary dynamic scenarios.

[0199] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0200] The initial values ​​of the dynamic environment element data are input into the initial simulation model, and the vehicle is run in the initial simulation model.

[0201] The system detects whether the vehicle state has reached a preset state. When the vehicle state reaches the preset state, it acquires the vehicle's state data in the initial simulation model, thereby generating multiple first simulation dynamic scenarios.

[0202] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0203] If the first simulated dynamic scene does not match the preset scene, the dynamic environment element data is determined to be supplementary element data, and the supplementary element data is added.

[0204] The supplemented dynamic environment element data is input into the initial simulation model to generate multiple second simulation dynamic scenes;

[0205] The system sequentially checks whether each of the second simulated dynamic scenes matches the preset scene. If the second simulated dynamic scene matches the preset scene, the dynamic environment model is constructed based on the multiple second simulated dynamic scenes and the first preparatory dynamic scene.

[0206] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0207] The dynamic master vehicle data is acquired, and the dynamic master vehicle data is arranged according to the acquisition time to obtain dynamic master vehicle sequence data;

[0208] A vehicle acceleration model is established based on the dynamic master vehicle sequence data;

[0209] The vehicle is driven by the vehicle acceleration model.

[0210] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0211] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0212] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for modeling a vehicle simulation scene, characterized in that, include: Acquire static scene data and construct a static scene model based on the static scene data; Load the static scene model and vehicle kinematics model to construct the initial simulation model; Acquire dynamic environment data, and perform coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model; Acquire dynamic master vehicle data, and drive the vehicle based on the initial simulation model using the dynamic master vehicle data; Detect whether the vehicle's speed has reached a preset speed. When the vehicle's speed reaches the preset speed, construct a dynamic master vehicle model based on the parameters of the vehicle that has reached the preset speed. A dynamic scene model is generated based on the dynamic environment model and the dynamic master vehicle model.

2. The vehicle simulation scene modeling method according to claim 1, characterized in that, The steps of acquiring static scene data and constructing a static scene model based on the static scene data include: Collect feature parameter data of preset road marking lines, and perform segmented fitting on the feature parameter data to obtain multiple pre-aiming paths, each pre-aiming path having multiple discrete points; The discrete points of each segment of the pre-aimed path are integrated to form a discrete point sample set; The road curve is obtained by globally fitting the discrete point sample set, and a static scene model is constructed based on the road curve.

3. The vehicle simulation scene modeling method according to claim 2, characterized in that, The step of acquiring static scene data and constructing a static scene model based on the static scene data further includes: The feature parameter data is preprocessed; The system detects whether the number of feature parameter data within a preset time is equal to a preset number. When the number of feature parameter data within the preset time is equal to the preset number, the system performs a global fitting on the discrete point sample set to obtain the road curve. A static scene model is constructed based on the road curve.

4. The vehicle simulation scene modeling method according to claim 1, characterized in that, The steps of acquiring dynamic environment data and performing coordinate transformation on the dynamic environment data based on the initial simulation model to obtain the dynamic environment model include: The dynamic environment data is acquired, and the dynamic environment data is arranged according to the acquisition time to obtain dynamic environment sequence data; The dynamic environment sequence data is transformed using coordinates to generate dynamic environment element data; The dynamic environment element data is input into the initial simulation model to generate multiple first simulation dynamic scenes; The system sequentially checks whether each of the first simulated dynamic scenes matches the preset scene. If the first simulated dynamic scene matches the preset scene, then the first simulated dynamic scene is determined to be the first preparatory dynamic scene. The dynamic environment model is constructed based on multiple first preliminary dynamic scenarios.

5. The vehicle simulation scene modeling method according to claim 4, characterized in that, The step of inputting the dynamic environment element data into the initial simulation model to generate multiple first dynamic simulation scenes includes: The initial values ​​of the dynamic environment element data are input into the initial simulation model, and the vehicle is run in the initial simulation model. The system detects whether the vehicle state has reached a preset state. When the vehicle state reaches the preset state, it acquires the vehicle's state data in the initial simulation model, thereby generating multiple first simulation dynamic scenarios.

6. The vehicle simulation scene modeling method according to claim 5, characterized in that, After the step of sequentially detecting whether each of the first simulated dynamic scenes matches the preset scene, the method further includes: If the first simulated dynamic scene does not match the preset scene, the dynamic environment element data is determined to be supplementary element data, and the supplementary element data is added. The supplemented dynamic environment element data is input into the initial simulation model to generate multiple second simulation dynamic scenes; The system sequentially checks whether each of the second simulated dynamic scenes matches the preset scene. If the second simulated dynamic scene matches the preset scene, the dynamic environment model is constructed based on the multiple second simulated dynamic scenes and the first preparatory dynamic scene.

7. The vehicle simulation scene modeling method according to claim 5, characterized in that, The step of acquiring dynamic master vehicle data and driving the vehicle based on the initial simulation model using the dynamic master vehicle data includes: The dynamic master vehicle data is acquired, and the dynamic master vehicle data is arranged according to the acquisition time to obtain dynamic master vehicle sequence data; A vehicle acceleration model is established based on the dynamic master vehicle sequence data; The vehicle is driven by the vehicle acceleration model.

8. A vehicle simulation scene modeling device, characterized in that, include: The static data acquisition module is used to acquire static scene data; The static model building module is used to build static scene models based on static scene data. The initial model building module is used to load the static scene model and the vehicle kinematics model to build the initial simulation model; The dynamic data acquisition module is used to acquire dynamic environmental data and dynamic main vehicle data; The dynamic model building module is used to build a dynamic scene model based on the initial simulation model and the dynamic environment data and the dynamic main vehicle data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to 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 by a processor, it implements the steps of the method described in any one of claims 1-7.

Images