CN115830255B

CN115830255B - Simulation scene generation method and device, electronic equipment and storage medium

Info

Publication number: CN115830255B
Application number: CN202211512732.XA
Authority: CN
Inventors: 周辰霖; 常立乐
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-11-28
Filing date: 2022-11-28
Publication date: 2023-11-21
Anticipated expiration: 2042-11-28
Also published as: CN115830255A

Abstract

The disclosure provides a simulation scene generation method, a simulation scene generation device, electronic equipment and a storage medium, and relates to the field of artificial intelligence, in particular to the field of automatic driving and intelligent traffic. The specific implementation scheme is as follows: acquiring road network data; constructing a region model of a target region and a road network element model in the target region according to the road network data; determining an out-of-road-network element model according to the region model and the road network element model in the target region; detecting a target element model in a unit model included in the road network external element model according to the element real space characteristics of the target element; modeling rendering is carried out on the target element model, and a simulation scene is generated. The embodiment of the disclosure can increase the authenticity of the simulation scene, enrich the details of the simulation scene, and thereby improve the accuracy of the simulation scene.

Description

Simulation scene generation method and device, electronic equipment and storage medium

Technical Field

The disclosure relates to the field of artificial intelligence, in particular to the field of automatic driving and intelligent traffic, and particularly relates to a simulation scene generation method, a simulation scene generation device, electronic equipment and a storage medium.

Background

Along with the development of automatic driving technology, in order to ensure the safety of an automatic driving system, a simulation scene needs to be constructed for testing.

In the construction process of a simulation scene, data is generally dependent on a high-precision map.

Disclosure of Invention

The disclosure provides a simulation scene generation method, a simulation scene generation device, electronic equipment and a storage medium.

According to an aspect of the present disclosure, there is provided a simulation scene generation method, including:

acquiring road network data;

constructing a region model of a target region and a road network element model in the target region according to the road network data;

determining an out-of-road-network element model according to the region model and the road network element model in the target region;

detecting a target element model in a unit model included in the road network external element model according to the element real space characteristics of the target element;

modeling rendering is carried out on the target element model, and a simulation scene is generated.

According to an aspect of the present disclosure, there is provided a simulation scene generating apparatus including:

the data acquisition module is used for acquiring road network data;

the road network element construction module is used for constructing a region model of a target region and a road network element model in the target region according to the road network data;

The road network external element construction module is used for determining a road network external element model according to the region model and the road network element model in the target region;

the target element determining module is used for detecting a target element model in unit models included in the road network external element model according to the element real space characteristics of the target element;

and the rendering module is used for modeling and rendering the target element model to generate a simulation scene.

According to another aspect of the present disclosure, there is provided an electronic device including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the simulation scenario generation method of any one embodiment of the present disclosure.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the simulation scenario generation method of any one of the embodiments of the present disclosure.

According to another aspect of the present disclosure, there is provided a computer program object comprising a computer program which, when executed by a processor, implements the simulation scenario generation method of any one of the embodiments of the present disclosure.

The embodiment of the disclosure can increase the authenticity of the simulation scene and enrich the details of the simulation scene, thereby improving the accuracy and richness of the simulation scene.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1a is a flow chart of a simulation scenario generation method disclosed in accordance with an embodiment of the present disclosure;

FIG. 1b is a schematic illustration of a region model of a target region disclosed in accordance with an embodiment of the present disclosure;

FIG. 1c is a schematic diagram of a road network element model in a target area according to an embodiment of the present disclosure;

FIG. 2a is a flow chart of another simulation scenario generation method disclosed in accordance with an embodiment of the present disclosure;

FIG. 2b is a schematic diagram of an extended road network element model disclosed in accordance with an embodiment of the present disclosure;

FIG. 2c is a schematic illustration of a building element model disclosed in accordance with an embodiment of the present disclosure;

FIG. 2d is a schematic illustration of a smooth intersection model disclosed in accordance with an embodiment of the present disclosure;

FIG. 3 is a flow chart of yet another simulation scenario generation method disclosed in accordance with an embodiment of the present disclosure;

FIG. 4a is a flow chart of another simulation scenario generation method disclosed in accordance with an embodiment of the present disclosure;

FIG. 4b is a schematic illustration of a greening element model disclosed according to an embodiment of the present disclosure;

FIG. 4c is a schematic illustration of a pavement element model disclosed in accordance with an embodiment of the present disclosure;

FIG. 4d is a schematic diagram of a cell model disclosed in accordance with an embodiment of the present disclosure;

FIG. 5a is a schematic application diagram of a simulation scenario generation method disclosed in accordance with an embodiment of the present disclosure;

FIG. 5b is a schematic diagram of a simulation scenario generated in accordance with the prior art;

FIG. 5c is a schematic diagram of a simulation scenario generated by a simulation scenario generation method disclosed in an embodiment of the present disclosure;

FIG. 6 is a block diagram of a simulation scenario generating apparatus in an embodiment of the present disclosure;

fig. 7 is a block diagram of an electronic device used to implement a simulation scenario generation method of an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

Fig. 1a is a flowchart of a simulation scenario generation method disclosed in accordance with an embodiment of the present disclosure, which may be applicable to the case of creating a simulation scenario for an autonomous vehicle system. The method of the embodiment can be executed by a simulation scene generating device, the device can be realized in a software and/or hardware mode, and the device is specifically configured in an electronic device with a certain data operation capability, and the electronic device can be a vehicle-mounted device. As shown in fig. 1a, the method specifically includes:

s101, road network data are acquired.

Road network refers to a road system composed of various roads and interconnected and interlaced into net-shaped distribution, and can be understood as a network composed of roads and intersections. Road network data refers to data describing a road network. In the present disclosure, road network data may be acquired through a high-precision map.

Specifically, a map file of a high-definition map can be obtained, and the map file is loaded. Road network data is stored in the map file. The road network data at least comprises the road topology network and the types of road network elements. Wherein the layout of the road system can be determined by means of the road topology network. Road network elements refer to elements that make up a road topology network. The road network element comprises at least one of road elements, intersection elements, lane line elements, camera elements, traffic indicator light elements, traffic sign elements and the like. The types of road network elements include at least a lane type and a lane line type. The lane type includes at least one of city lane, high speed lane, bus lane, tide lane, and lane speed limit type. The lane line type includes at least one of yellow line, white line, broken line, solid line, and the like.

S102, constructing a region model of a target region and a road network element model in the target region according to the road network data.

The target region refers to a region in which a simulation scene is to be generated. A simulation scene may be generated for one city and a partial region may be designated as a target region therefrom. The simulation scene refers to a scene capable of presenting a real space, and is a model built on the basis of the real space. A model may be understood as a geometric body, and a region model refers to a geometric body established based on road network data within a certain region. The geometry may be a solid geometry or a planar geometry. The road network element model refers to a geometry for describing road network elements, and may include road geometry and intersection geometry. A road network element model may be included within the region model. The region model of the target region refers to a geometry created from road network data within the target region. In the present disclosure, the region model of the target region may be a rectangular parallelepiped capable of including the target region, and the road network element model in the target region is included in the region model of the target region. A schematic diagram of a region model of a target region according to an embodiment of the present disclosure is shown in fig. 1 b. The road network element model in the target area refers to a model of road network elements located within the target area. As shown in fig. 1c, a schematic diagram of a road network element model in a target area according to an embodiment of the present disclosure is disclosed, and in fig. 1c, road geometry and intersection geometry are shown.

Specifically, road network data in a target area can be screened out according to the road network data, an area model of the target area is built according to the road network data in the target area, and a road network element model in the target area is built in the area model; and constructing a region model and a region road network element model according to the road network data, and screening the region model of the target region and the road network element model in the target region according to the target region in the region model and the region road network element model. Wherein, the cross section of the solid geometry of the target area and parallel to the ground can be obtained to obtain an area model.

S103, determining an out-of-road-network element model according to the region model and the road-network element model in the target region.

An out-of-road element refers to an element within a region other than the out-of-road element. The extranet element model refers to a geometry used to describe the extranet element. Illustratively, the off-road network element may include at least one of: building, green belt, sidewalk, etc. In addition, off-road elements may be further subdivided, for example, to differentiate buildings into residential, commercial, public and beneficial buildings, and the like. In this regard, the setting may be performed as necessary.

Specifically, according to the region model and the road network element model in the target region, the region where the road network element is located in the target region can be removed, the remaining region is determined to be the region where the road network element is located, and the geometric body corresponding to each independent closed region in the remaining region is respectively determined to be the road network element model; and in the residual region, a region with a certain range is narrowed inwards along the boundary of the residual region to obtain a region corresponding to the narrowed range and a narrowed residual region, the region corresponding to the narrowed range and the narrowed residual region are respectively determined to be the region where the road network external element is located, and the corresponding geometric body is determined to be the road network external element model. For example, boolean operation can be performed on the area model and the road network element model in the target area, and the operation result is determined as the road network external element model; the area model and the road network element model in the target area can be subjected to Boolean operation, the model obtained by the operation result is reduced according to a preset value, and the reduced operation result and the reduced part are respectively determined as the road network external element. Illustratively, the region corresponding to the reduced range is generally referred to as an edge region, which is generally a region of a sidewalk or the like. The boolean operation is to obtain a new object shape by performing a union, difference or intersection operation on two or more objects. In this disclosure, a boolean operation may be an operation of a difference set.

S104, detecting a target element model in unit models included in the road network external element model according to the element real space characteristics of the target element.

The target element refers to an element to be added to the simulation scene to be modeled. The target element is usually an element outside the road network and can be set according to the requirement. The element real space feature refers to the feature of an element outside the road network in real space. The element real space features are used for distinguishing different types of off-road network element models to detect the types of the off-road network element models, and exemplary element real space features include at least one of area, position, shape and the like. The target element model refers to a model to be added to a simulation scene that needs to be rendered. The target element model can be obtained by modeling the target element. The unit model can refer to a geometric body surrounded by a closed line, and a connecting line of any point inside the unit model and any point outside the unit model is intersected with the boundary of the unit model. In practice, the off-road network element model includes a plurality of unit models, one unit model represents an independent off-road network element, and rendering parameters of different types of unit models are different. Illustratively, the element model of the lawn is rendered green in color, low in rendering height, and so on. The unit model with corresponding characteristics can be detected according to the real space characteristics of the elements of some types of elements, so that the purpose of dividing the unit model included in the element model outside the road network is achieved, further detail confirmation of the element outside the road network is achieved, information of the element outside the road network is enriched, and therefore the reality and fine-grained content of a simulation scene are improved.

The determining of the target element model may be to screen a unit model included in an element model outside the road network, and for a simulation scene of testing the automatic driving system, some elements may interfere with vehicle running in the road network, and such elements are added to the simulation scene to realize more accurate and complex testing of the automatic driving system. Alternatively, the target element model may be a geometric body of an element that has an interference with the running of the vehicle in the road network, for example, a geometric body of an element such as a sidewalk, a green belt, and a building. In addition, other scenes exist, and elements which need to be detected and filled in the simulation scene and the real space characteristics of the elements of the target element model corresponding to the elements can be set according to the needs.

Specifically, according to the element real space characteristics of the target element, matching with a unit model included in the element model outside the road network, and determining the unit model meeting the element real space characteristics in the unit model included in the element model outside the road network as the target element model. Multiple types of target elements can be set, and corresponding element real space features are set corresponding to different types, so that target element models corresponding to the target elements of all types are screened out. Exemplary, the elemental real space features of an element include loops and areas less than 10m ² In the road network external element model, the annular shape and the area are simultaneously satisfied to be less than 10m ² Is determined as a target element model for the element.

And S105, modeling and rendering the target element model to generate a simulation scene.

Modeling rendering refers to building and coloring models in a simulated scene. Specifically, modeling is performed on the target element, a shape body conforming to the real situation is constructed, for example, the target element is a flower bed, the shape body of the flower bed with a certain occupied space is generated at the corresponding position, the target element is colored, the color of the target element in the real space is reflected, and a simulation scene is formed. In the present disclosure, the simulation scene may be a scene that can reflect the surroundings of the vehicle collected during the driving of the autonomous vehicle. In the simulation test of the automatic driving vehicle system, as the automatic driving vehicle runs in a simulation scene, the positions of the automatic driving vehicle are different, and the perceived simulation data are different. When the target element model is determined, no exogenous data is introduced, that is, no support of other external data is adopted, the detection of the target element model is still realized, and the detection is supplemented into a simulation scene.

In the prior art, a simulation scene is generally constructed by a high-precision map, and the high-precision map generally does not comprise areas of building blocks, greenbelts, sidewalks and other data outside a road network. For the missing area, the data of the navigation map with lower precision is generally adopted for rendering. Due to the precision problem, the problem of position conflict or mismatch often occurs in the combination process of the area with the road network area, which is obtained by rendering. In addition, part of data may be missing in the navigation map, so that after the missing region is combined with the road network region, the content is still missing.

According to the technical scheme, the regional model of the target region and the road network element model in the target region are constructed according to the acquired road network data, the road network outer model is determined according to the regional model and the road network element model in the target region, the road network outer element model is determined, and the integrity of the road network outer element model can be ensured, so that when a simulation scene is generated, the details of the simulation scene are enriched, the integrity of the simulation scene is ensured, the richness of the simulation scene is improved, the target element model is determined according to the real space characteristics of the elements in the road network outer element, the target element model is rendered, the simulation scene is generated, the authenticity of the simulation scene can be increased, the accuracy and richness of the simulation scene are improved, meanwhile, no exogenous data is introduced, namely, the support of other external data is not adopted, the detection of the target element model is still realized, the simulation scene is supplemented, and the creation cost of the simulation scene can be reduced.

Fig. 2a is a flowchart of another simulation scenario generation method disclosed in an embodiment of the present disclosure, which is further optimized and expanded based on the above technical solution, and may be combined with the above various alternative implementations. On the basis of the foregoing embodiment, the determining an out-of-road element model according to the region model and the road element model in the target region includes: extending the road network element model to obtain an extended road network element model; and calculating the difference between the extended road network element model and the region model to obtain a building element model, and determining the building element model as an out-of-road network element model. As shown in fig. 2a, the method specifically includes:

s201, road network data are acquired.

S202, constructing a region model of a target region and a road network element model in the target region according to the road network data.

S203, extending the road network element model to obtain an extended road network element model.

The extension means that the road network element model is expanded outwards, and the area and the volume of the road network element are enlarged under the condition that the shape of the road network element model is not changed. The extended road network element model is a model obtained by performing external extension on road network elements. Fig. 2b is a schematic diagram of an expanded road network element model according to an embodiment of the present disclosure, the plane area of the expanded road network element model shown in fig. 2b being larger than the plane area of the road network element model of fig. 1 c.

Specifically, according to the road network element model, the road network element model is expanded towards the direction outside the model, and the expanded model is determined as an expanded road network element model. The extended distance can be set according to practical situations, and the edge of the road network element model is extended outwards by 10m in an exemplary manner.

S204, calculating the difference between the extended road network element model and the region model to obtain a building element model, and determining the building element model as an external road network element model.

The differences refer to models corresponding to different regions between the expanded road network element model and the region model. Building elements refer to buildings that exist in real space. By way of example, the building elements may be elements of residential buildings, parks, office buildings, malls, and the like. Building element models refer to geometry used to describe building elements.

Specifically, boolean operation can be performed on the extended road network element model and the region model, and a difference set between the extended road network element and the region model is taken, so that a difference between the extended road network element model and the region model is obtained, a calculation result is determined as a building element model, and the building element model is determined as an out-of-road network element model. In the present disclosure, the difference between the extended road network element model and the region model may be calculated by other methods, which is not limited. Fig. 2c is a schematic illustration of a building element model disclosed in accordance with an embodiment of the present disclosure.

S205, detecting a target element model in unit models included in the road network external element model according to the element real space characteristics of the target element.

S206, modeling rendering is carried out on the target element model, and a simulation scene is generated.

In one embodiment, after the calculating the difference between the extended road network element model and the area model, a building element model is obtained and determined as an off-road network element model, the method further includes: and combining the building element model with the road network element model, calculating the difference between the combined model and the region model to obtain a road side element model, and determining the road side element model as an external road network element model.

Combining means that the building element model and the road network element model are combined, and as a whole, the combination means that the building element model and the road network element model form a set, and the building element model and the road network element model are not processed. The combined model is a model obtained by combining a building element model and a road network element model. The road side element refers to an element located in a certain range around a road in real space. By way of example, the road side elements may be elements of sidewalks, guardrails, greenbelts, and the like. The roadside element model refers to a model for describing a roadside element.

Specifically, the building element model and the road network element model are combined to form a whole to obtain a combined model, a difference set between the combined model and the region model can be determined through Boolean operation, so that the difference between the combined model and the region model is obtained, a calculation result is determined to be a road side element model, and the road side element model is determined to be an out-of-road network element model.

Because road side elements such as green belts and sidewalks generally exist between building elements and road network elements in real space, the building element models and the road network element models are combined, and the areas where the road side elements are located are removed from the obtained combined models, therefore, the road side element models are determined through the difference between the combined models and the area models, the determination of the road side element models is realized, the road side element models are determined to be the road network external element models, the richness of the road network external element models can be increased, and the richness of simulation scenes is improved.

In one embodiment, the extending the road network element model to obtain an extended road network element model includes: performing smoothing treatment on the intersection model included in the road network element model to obtain a smooth intersection model; and carrying out epitaxial combination on the smooth intersection model and the road model included by the road network element model to obtain an extended road network element model.

The road network elements comprise crossing elements and road elements, and it is understood that the crossing elements are connected through the road elements to form a road topology network. Correspondingly, the road network element model comprises an intersection model and a road model. The intersection model is a model for describing intersection elements, and the road model is a model for describing road elements.

The smoothing process is to smooth the boundary line in the intersection element model and reduce the sharpened shape formed by the boundary line, specifically, the convex hull shape. It will be appreciated that the geometry of the intersection element is generally concave polygonal. In the process of creating the intersection element model, the boundary of the intersection element is depicted through points and lines, at least two lines are connected at the same point, inward concave included angles which can be formed between the two lines form convex hulls among the included angles. The smoothing process is to smooth the convex hull, specifically, the two lines connected at the same point can be connected by replacing the point with a curve, or the two lines connected at the same point can be connected by replacing the point with a straight line. Fig. 2d is a schematic diagram of a smooth intersection model disclosed in accordance with an embodiment of the present disclosure. As shown in fig. 2d, the dotted line portion is a convex hull before the smoothing process, and the solid line portion is a smoothed intersection model obtained after the smoothing process. The smooth intersection model is a model obtained by performing smoothing processing on the intersection model. The extension combination means that the smooth intersection model and the road model are respectively extended and combined. Illustratively, the road network element model includes an intersection model A1 and a road model B1, and the intersection model A1 is subjected to smoothing processing to obtain a smoothed intersection model A2. And combining the road model B1 and the smooth road opening model A2 to obtain a new road network element model, and extending the road network element model to obtain an extended road network element model. Or respectively extending the road model B1 and the smooth road opening model A2, and combining the extended models to obtain an extended road network element model.

By carrying out smoothing treatment on the intersection model to obtain a smooth intersection model, the probability of occurrence of data extrusion condition can be reduced in the process of extending the intersection model, and the accuracy of the smooth intersection model is improved, so that when the smooth intersection model and the road network element model are subjected to extending combination, the accuracy of the extended road network element model is improved.

According to the technical scheme, the road network element model is extended to obtain the extended road network element model, and compared with the road network element model, the plane area of the extended road network element is larger, and because a certain space distance exists between the building element and the road network element in real space, when the building element model is determined through the extended road network element model and the area model, the area between the building element and the road network element can be removed, the accuracy of the building element model is improved, the building element model is determined to be an out-of-road-network element model, the richness of the out-of-road-network element model can be increased, and therefore the richness of simulation scenes is improved.

Fig. 3 is a flowchart of yet another simulation scenario generation method disclosed in an embodiment of the present disclosure, which is further optimized and expanded based on the above technical solution, and may be combined with the above various alternative implementations. On the basis of the above embodiment, the method further comprises: acquiring at least one unit model in the building element model; and screening the unit models included in the building element model according to the model space characteristics of each unit model, and updating the building element model. As shown in fig. 3, the method specifically includes:

S301, road network data are acquired.

S302, constructing a region model of a target region and a road network element model in the target region according to the road network data.

S303, extending the road network element model to obtain an extended road network element model.

S304, calculating the difference between the extended road network element model and the region model to obtain a building element model, and determining the building element model as an external road network element model.

S305, acquiring at least one unit model in the building element model.

A building element model is understood to mean a collection of at least one unit model, a unit model representing an individual building. The unit model of which the type is a building element is used as or added to the building element model.

S306, screening the unit models included in the building element model according to the model space characteristics of each unit model, and updating the building element model.

Model spatial features are used to describe the spatial features of the unit model. By way of example, model space features may include features such as area and planar shape. In practice, the building elements have spatial features such as large area and polygonal shape, and the unit models which do not belong to the building elements can be removed according to the comparison between the real spatial features of the building elements and the model spatial features. Specifically, according to each unit model, determining model space characteristics of each unit model, screening the unit models through the model space characteristics, and updating the building element model by using the unit models reserved after screening.

Exemplary, the model space features are areas, unit models smaller than an area threshold are removed according to the areas of the unit models, and the building element models are updated by using the reserved unit models; the model space features are plane shapes, unit models with the plane shapes of round, triangle, strip and the like are removed according to the plane shapes of the unit models, and the building element models are updated by using the reserved unit models. The area threshold and the plane shape can be set according to practical situations.

S307, combining the building element model and the road network element model, calculating the difference between the combined model and the region model, obtaining a road side element model, and determining the road side element model as an external road network element model.

S308, detecting a target element model in unit models included in the road network external element model according to the element real space characteristics of the target element.

And S309, modeling rendering is carried out on the target element model, and a simulation scene is generated.

In one embodiment, the method further comprises: and according to the generated simulation scene, performing simulation test on the automatic driving vehicle system.

An autonomous vehicle system refers to a system that controls a vehicle to perform autonomous driving. The simulation test is used for simulating the running condition of the automatic driving vehicle system in the running process of the automatic driving vehicle. Specifically, the simulation scene is used for simulating the environment of the automatic driving vehicle in the driving process. In the simulation scene, the automatic driving vehicle system is operated, so that the automatic driving vehicle system can acquire the surrounding environment of the vehicle in the simulation scene to simulate the running condition of the automatic driving vehicle in the real world, thereby testing how the automatic driving vehicle runs in the simulation scene, testing the functions of the automatic driving vehicle system and the like, and optimizing the automatic driving vehicle system.

By performing simulation test on the automatic driving vehicle system in the simulation scene, the simulation running condition of the automatic driving vehicle system can be obtained before the automatic driving vehicle system is applied to the vehicle, the cost of the automatic driving vehicle system test is reduced, the resource waste of the automatic driving vehicle system test is reduced, and the accuracy of the test and the precision of the test result can be improved by improving the authenticity of the simulation scene.

According to the technical scheme, the unit models in the building element models are obtained, the unit models are screened according to the model space characteristics of the unit models, the updating of the building element models is achieved, the unit models with errors in the building element models can be removed, the data quantity of the building element models is reduced, and the accuracy of the building element models is improved.

Fig. 4a is a flowchart of another simulation scenario generation method disclosed in an embodiment of the present disclosure, which is further optimized and expanded based on the above technical solution, and may be combined with the above various alternative implementations. On the basis of the above embodiment, the element real space feature includes an element real region shape, and the determining a target element model according to the element real space feature in the road network external element model includes: classifying at least one unit model included in the road side element model according to the shape of the element real region to obtain an element model of the annular region and an element model of the block region; determining the element model of the block area as a greening element model; determining an element model of the annular region as a pavement element model; and determining the building element model, the greening element model and the pavement element model as target element models. As shown in fig. 4a, the method specifically includes:

S401, road network data are acquired.

S402, constructing a region model of a target region and a road network element model in the target region according to the road network data.

S403, extending the road network element model to obtain an extended road network element model.

S404, calculating the difference between the extended road network element model and the region model to obtain a building element model, and determining the building element model as an external road network element model.

S405, combining the building element model and the road network element model, calculating the difference between the combined model and the region model, obtaining a road side element model, and determining the road side element model as an external road network element model.

S406, classifying at least one unit model included in the road side element model according to the shape of the element real region to obtain an element model of the annular region and an element model of the block region. The elemental real space features include elemental real region shapes.

The roadside element model may describe at least one roadside element in real space. A roadside element model may be understood as a collection comprising at least one unit model, one unit model representing an individual roadside element. The unit model with the type of the road side element is used as or added to the road side element model. The element real area shape refers to the shape of the roadside element in real space. The elemental real area shape may include a ring-shaped area and a block-shaped area. The element model of the annular region is used to describe the element in which the element real region is annular in shape. The element model of the block-shaped region is used to describe the element in which the element real region is shaped as a block.

Specifically, in at least one unit model included in the roadside element model, the unit models are classified according to the element real region shape, the unit model with the element real region shape being annular is determined as the element model of the annular region, and the unit model with the element real region shape being block is determined as the element model of the block region.

S407, determining the element model of the block area as a greening element model.

The greening element refers to an element for greening the environment in the vicinity of a road (between two sides or two-way lanes), and may be a green belt, for example. The greening element model is used for describing the greening elements. In the real space, the shape of the greening element is generally a block, and therefore, the element model of the block area is determined as a greening element model. Fig. 4b is a schematic diagram of a greening element model disclosed according to an embodiment of the present disclosure. As shown in fig. 4b, the model of the narrow and long block-shaped closed area in the middle of the road is the greening element model.

S408, determining the element model of the annular region as a pavement element model.

The pavement element refers to an area for pedestrians to walk on both sides of the lane. Illustratively, the pavement element is an area where waterproof tables (or steps) on both sides of the road are located. The pavement element model is used to describe pavement elements. In real space, the pavement elements are generally distributed along the lanes and the building element models to form a loop-shaped region, and thus, the element model of the loop-shaped region is determined as a pavement element model. Fig. 4c is a schematic illustration of a pavement element model disclosed in accordance with an embodiment of the present disclosure. As shown in fig. 4c, the closed area of the long and narrow ring shape around the building element model (model of the massive area) is the pavement element model.

S409, determining the building element model, the greening element model and the sidewalk element model as target element models.

S409, modeling rendering is carried out on the target element model, and a simulation scene is generated.

In one embodiment, in the at least one unit model included in the roadside element model, classification is performed according to a shape of an element real region to obtain an element model of a ring-shaped region and an element model of a block-shaped region, including: acquiring nodes of unit models included in the road side element model; detecting whether a target node has at least three adjacent points; determining that the unit model is an element model of a ring-shaped region, if any; in the absence, the unit model is determined to be an element model of a block region.

Nodes refer to points used to build a cell model. The cell model may be constructed by interconnecting the nodes and lines. Fig. 4d is a schematic diagram of a cell model disclosed in accordance with an embodiment of the present disclosure. As shown in fig. 4d, at least two lines are connected to one node in the cell model. In the constructed cell model, a connection can be established between any two nodes through at least one line. The adjacent point means two nodes connected by a line. As shown in fig. 4b, nodes a, b, c, d, e, f, g and h are included in model a. The closed region formed between the curve abcd and the curve efgh is the model a. Nodes a, B, c, and d are included in model B. The closed area formed by the curve abcd is model B. In model a (the ring model), node a has adjacent points b, d, and e. In model B (block model), node a has adjacent points B and d. The target node refers to a node having at least three neighboring nodes.

Specifically, nodes of a unit model included in the road side element model are obtained, all the nodes are traversed, and the number of adjacent points of each node is determined. A node having at least three neighboring nodes is determined as a target node. Detecting whether a target node exists in each unit model, and determining the unit model with the target node as an element model of the annular region; the unit model without the target node is determined as an element model of the block-shaped region. As shown in fig. 4b, the model a has a target node a, and the model a is defined as an element model of the annular region. In the model B, there is no target node, and the model B is determined as an element model of the block region.

Further, the area of the unit model included in the road side element model can be obtained. Detecting nodes and area of each unit model, and determining the unit model as an element model of the annular area under the condition that a target node exists and the area of the area is smaller than a preset threshold value; under the condition that a target node exists and the area of the area is larger than a preset threshold value, determining that the unit model is an element model of a block-shaped area; in the case where there is no target node, the unit model is determined to be an element model of the block region. The preset threshold value can be set according to actual conditions.

The element models of the annular region and the element models of the block region are determined according to the existence condition by acquiring the nodes of the unit models included in the road side element models, so that the unit models can be classified by utilizing the nodes of the unit models, the classification process is simplified, and the classification efficiency of the unit models is improved.

According to the technical scheme, at least one unit model included in the road side element model is classified according to the shape of the element real area, so that a classification method can be simplified, and the accuracy of determining the greening element model and the pavement element model is improved.

Fig. 5a is an application schematic diagram of a simulation scenario generating method according to an embodiment of the present disclosure, as shown in fig. 5a, the method includes:

s501, loading a map file of a high-precision map, and constructing a region model of a target region and a road network element model in the target region according to road network data in the map file.

And calculating a bounding box of the road network covered by the target area, and extracting one cross section of the bounding box to obtain an area model of the target area and a road network element model in the target area. Indeed, the model in the embodiments of the present disclosure is a planar geometry.

S502, performing smoothing treatment on the road junction element to obtain a smoothing treatment model.

S503, carrying out extension combination on the road model included by the smooth intersection model and the road network element model to obtain an extended road network element model.

S504, calculating the difference between the extended road network element model and the area model to obtain a building element model.

S505, combining the building element model and the road network element model, and calculating the difference between the combined model and the region model to obtain the road side element model.

S506, classifying at least one unit model included in the road side element model according to the shape of the element real region to obtain an element model of the annular region and an element model of the block region.

Specifically, classification can be performed by the following algorithm:

int point[]＝primpoints(0,@primnum)；

for(int i＝0；i<len(points)；i++){

int neighbours[]＝neighbours(0,i)；

if(len(neighbours)>2&&@area<ch(“max”)){

@side_walk＝1；

}

the number of adjacent points len (neighbours) and the area of the unit model can be detected through the algorithm, and the unit model is determined to be the element model side_walk of the annular region under the condition that the number of adjacent points is at least three and the area is smaller than the preset threshold max.

S507, determining the element model of the block area as a greening element model.

S508, determining the element model of the annular region as a pavement element model.

S509, modeling and rendering the building element model, the greening element model and the pavement element model to generate a simulation scene.

A schematic diagram of a simulation scenario generated according to the prior art is shown in fig. 5 b. Shown in fig. 5c is a schematic diagram of a simulation scenario generated by the simulation scenario generation method disclosed in an embodiment of the present disclosure. By comparing fig. 5b with fig. 5c, it can be seen that the simulation scene graph obtained by the simulation scene generating method disclosed by the present disclosure has richer details, and can reflect the scene in real space more.

Fig. 6 is a block diagram of a simulation scenario generation apparatus in an embodiment of the present disclosure, which is applicable to a case of running a simulation scenario generation method, according to an embodiment of the present disclosure. The device is realized by software and/or hardware, and is specifically configured in the electronic equipment with certain data operation capability.

A simulation scene generating apparatus 600 as shown in fig. 6, comprising: a data acquisition module 601, a road network element construction module 602, an off-road network element construction module 603, a target element determination module 604 and a rendering module 605; wherein,

a data acquisition module 601, configured to acquire road network data;

the road network element construction module 602 is configured to construct a region model of a target region and a road network element model in the target region according to the road network data;

an extranet element construction module 603, configured to determine an extranet element model according to the region model and the road network element model in the target region;

a target element determining module 604, configured to detect a target element model in a unit model included in the off-road network element model according to an element real space feature of the target element;

and the rendering module 605 is used for modeling and rendering the target element model to generate a simulation scene.

Further, the road network external element model includes: a building element model and a roadside element model;

the out-of-road-network element construction module 603 includes:

the extension unit is used for extending the road network element model to obtain an extended road network element model;

The difference calculation unit is used for calculating the difference between the extended road network element model and the region model to obtain a building element model and determining the building element model as an external road network element model;

further, the off-road network element construction module further includes:

and the combination unit is used for calculating the difference between the extended road network element model and the area model to obtain a building element model, determining the building element model as an external road network element model, combining the building element model with the road network element model, calculating the difference between the combination model and the area model to obtain a road side element model, and determining the building element model as the external road network element model.

Further, the epitaxial unit includes:

the smoothing subunit is used for carrying out smoothing treatment on the intersection model included in the road network element model to obtain a smooth intersection model;

and the combining subunit is used for carrying out epitaxial combination on the smooth intersection model and the road model included by the road network element model to obtain an extended road network element model.

Further, the device further comprises:

the unit model acquisition module is used for acquiring at least one unit model in the building element models;

And the updating module is used for screening the unit models included in the building element model according to the model space characteristics of each unit model and updating the building element model.

Further, the elemental real space features include elemental real area shapes;

the target element determination module 604 includes:

the classification unit is used for classifying at least one unit model included in the road side element model according to the shape of the element real region to obtain an element model of the annular region and an element model of the block region;

a first model determination unit configured to determine an element model of the block-shaped region as a greening element model;

a second model determining unit configured to determine an element model of the annular region as a pavement element model;

and the target element determining unit is used for determining the building element model, the greening element model and the pavement element model as target element models.

Further, the classification unit includes:

a node obtaining subunit, configured to obtain a node of a unit model included in the roadside element model;

an adjacent point detection subunit, configured to detect whether there are target nodes with at least three adjacent points;

A first model determination subunit configured to determine, if any, that the unit model is an element model of an annular region;

and the second model determining subunit is used for determining that the unit model is an element model of the block area in the absence of the unit model.

Further, the device further comprises:

and the test module is used for carrying out simulation test on the automatic driving vehicle system according to the generated simulation scene.

The simulation scene generating device can execute the simulation scene generating method provided by any embodiment of the disclosure, and has the corresponding functional modules and beneficial effects of executing the simulation scene generating method.

In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program object.

Fig. 7 shows a schematic area diagram of an example electronic device 700 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the apparatus 700 includes a computing unit 701 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 may also be stored. The computing unit 701, the ROM702, and the RAM 703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Various components in device 700 are connected to I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, etc.; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, an optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The computing unit 701 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the respective methods and processes described above, for example, the simulation scene generation method. For example, in some embodiments, the simulation scenario generation method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 700 via ROM702 and/or communication unit 709. When a computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the simulation scenario generation method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the simulation scenario generation method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application specific standard objects (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or region diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A simulation scene generation method, comprising:

acquiring road network data;

Modeling and rendering the target element model to generate a simulation scene;

wherein the determining an out-of-road element model according to the region model and the road element model in the target region includes:

extending the road network element model to obtain an extended road network element model; the extension refers to enlarging the area and volume of the road network element under the condition that the shape of the road network element model is not changed;

and calculating the difference between the extended road network element model and the region model to obtain a building element model, and determining the building element model as an out-of-road network element model.

2. The method of claim 1, further comprising, after said calculating the difference between the extended road network element model and the region model, obtaining a building element model and determining as an off-road network element model:

combining the building element model and the road network element model, calculating the difference between the combined model and the region model to obtain a road side element model, and determining the road side element model as an external road network element model; the road side element model is used for describing elements positioned in a certain range around a road in real space.

3. The method of claim 2, wherein the extending the road network element model to obtain an extended road network element model comprises:

performing smoothing treatment on the intersection model included in the road network element model to obtain a smooth intersection model;

performing epitaxial combination on the smooth intersection model and a road model included in the road network element model to obtain an extended road network element model; the extension combination means that the smooth intersection model and the road model are respectively extended and combined.

4. The method of claim 2, further comprising:

acquiring at least one unit model in the building element model;

screening the unit models included in the building element model according to the model space characteristics of each unit model, and updating the building element model; wherein the model spatial features are used to describe the spatial features of the unit model.

5. The method of claim 2, wherein the elemental real space features comprise elemental real area shapes;

the detecting the target element model in the unit model included in the road network external element model according to the element real space characteristics of the target element comprises the following steps:

Classifying at least one unit model included in the road side element model according to the shape of the element real region to obtain an element model of the annular region and an element model of the block region;

determining the element model of the block area as a greening element model;

determining an element model of the annular region as a pavement element model;

and determining the building element model, the greening element model and the pavement element model as target element models.

6. The method according to claim 5, wherein the classifying according to the element real region shape in the at least one unit model included in the roadside element model to obtain an element model of a ring region and an element model of a block region includes:

acquiring nodes of unit models included in the road side element model;

detecting whether a target node has at least three adjacent points;

determining that the unit model is an element model of a ring-shaped region, if any;

in the absence, the unit model is determined to be an element model of a block region.

7. The method of claim 1, further comprising:

and according to the generated simulation scene, performing simulation test on the automatic driving vehicle system.

8. A simulation scene generating apparatus comprising:

the data acquisition module is used for acquiring road network data;

the rendering module is used for modeling and rendering the target element model to generate a simulation scene;

the road network external element construction module comprises:

the extension unit is used for extending the road network element model to obtain an extended road network element model; the extension refers to enlarging the area and volume of the road network element under the condition that the shape of the road network element model is not changed;

the difference calculation unit is used for calculating the difference between the extended road network element model and the region model to obtain a building element model and determining the building element model as an external road network element model.

9. The apparatus of claim 8, the off-road network element construction module further comprising:

the combination unit is used for calculating the difference between the extended road network element model and the area model to obtain a building element model, determining the building element model as an external road network element model, combining the building element model with the road network element model, calculating the difference between the combination model and the area model to obtain a road side element model, and determining the building element model as the external road network element model; the road side element model is used for describing elements positioned in a certain range around a road in real space.

10. The apparatus of claim 9, wherein the epitaxial cell comprises:

the combining subunit is used for carrying out epitaxial combination on the smooth intersection model and the road model included by the road network element model to obtain an extended road network element model; the extension combination means that the smooth intersection model and the road model are respectively extended and combined.

11. The apparatus of claim 9, further comprising:

the updating module is used for screening the unit models included in the building element model according to the model space characteristics of each unit model and updating the building element model; wherein the model spatial features are used to describe the spatial features of the unit model.

12. The apparatus of claim 9, wherein the elemental real space features comprise elemental real area shapes;

the target element determining module includes:

13. The apparatus of claim 12, wherein the classification unit comprises:

14. The apparatus of claim 8, further comprising:

15. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the simulation scenario generation method of any one of claims 1-7.

16. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the simulation scenario generation method according to any one of claims 1-7.