Digital twin virtual-real multi-vehicle mixed traffic simulation method and device
Through digital twin technology and virtual vehicle model, the difficulty of updating parameters and multi-vehicle experiments in autonomous driving simulation software is solved, and the low-cost and high-flexible virtual and real multi-vehicle mixed simulation is realized, which verifies the effectiveness of the multi-vehicle collaborative algorithm. .
Patent Information
- Application Number
- CN202110936056.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-16
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-08-16
AI Technical Summary
The existing autonomous driving simulation software is difficult to update dynamically after the parameters are preset, resulting in model inaccuracy and simulation results lag, unable to effectively deal with the rapid changes in complex systems, and it is difficult to carry out multi-vehicle experiments, which are costly and have low flexibility.
Introduce digital twin technology, build dynamic characteristics similar to physical space in computers through virtual vehicle models, establish information interaction channels with intelligent connected vehicles, build multi-vehicle collaboration scenarios in the information space, and adopt parallel digital twin simulation Process to realize mixed simulation of virtual and real multi-vehicles.
Reduces time costs, improves scene flexibility, provides a multi-vehicle experiment platform, verifies the effectiveness of multi-vehicle collaborative algorithm, solves the problem of difficult multi-vehicle experiments, and realizes real-time feedback on the results of virtual and real multi-vehicle mixed simulation .
Smart Images

Figure CN113642177B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a digital twin virtual-real multi-vehicle mixed traffic simulation method and device. Background Art
[0002] The intelligentization and networking of automobiles are the major trends in the development of the automobile industry today. Compared with traditional driver-driven cars, intelligent connected cars have great advantages in improving traffic efficiency. Before the commercialization of intelligent connected cars, a large number of experimental verifications of related technologies or algorithms are required to ensure safety. However, real vehicle experiments are often expensive and have many limitations, so simulations are often performed before real vehicle experiments.
[0003] Currently, the commonly used autonomous driving simulation software includes CarSim, Prescan, Carla, etc. These software or platforms have good features in vehicle dynamics modeling, sensor modeling, scene reconstruction, etc. However, the design patterns of these software have inherent defects, which are mainly reflected in the following aspects:
[0004] The above simulation software operation process mainly includes the steps of preset parameters - simulation operation - obtaining results - parameter adjustment. This serial execution process makes it difficult to effectively update the parameters dynamically during a single operation. When the model and the objective situation are very different, as the simulation time increases, the impact of model inaccuracy on the simulation results will become more and more significant; simulation always has a lag, the simulation of complex systems takes a long time, the simulation results cannot effectively cope with the rapid changes of complex systems, and the dynamic changes of the system cannot be used as real-time input to the simulation system. Summary of the invention
[0005] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.
[0006] To this end, one purpose of the present invention is to propose a digital twin virtual-real multi-vehicle mixed simulation method, which solves the problems of difficulty in dynamic updating and flexible adjustment after parameter preset in related technologies, and introduces digital twin technology into the simulation field of intelligent connected vehicles. By introducing virtual vehicles, attempts are made to use them to represent the real intelligent connected vehicles in the physical system, and multi-vehicle experiments are carried out based on virtual vehicles, which has the advantages of low time cost and high scene flexibility.
[0007] Another object of the present invention is to propose a digital twin virtual-real multi-vehicle mixed traffic simulation device.
[0008] To achieve the above-mentioned objectives, an embodiment of the present invention proposes a digital twin virtual-real multi-vehicle mixed running simulation method, comprising the following steps: based on the digital twin modeling technology, constructing a virtual vehicle model with similar dynamic characteristics to the intelligent connected vehicle in the physical space in a computer; based on communication, constructing an information interaction channel between the virtual vehicle model and the intelligent connected vehicle, and constructing a multi-vehicle collaborative scene in the information space; using the virtual vehicle model and the intelligent connected vehicle in the multi-vehicle collaborative scene to perform mixed running simulation, and obtain the simulation operation results of the virtual-real multi-vehicle mixed running.
[0009] The digital twin virtual-real multi-vehicle mixed simulation method of the embodiment of the present invention replaces the traditional simulation software / platform serial simulation process with a parallel digital twin solution, and has the characteristics of low time cost and high scene flexibility. By introducing a virtual car, it simulates an intelligent connected car, which solves the problem that real car experiments requiring multiple vehicles are difficult to carry out. It provides a possible experimental platform for research requiring multiple vehicles, such as queue control and multi-vehicle formation, based on which the effectiveness of the multi-vehicle collaborative algorithm can be verified.
[0010] In addition, the digital twin virtual-real multi-vehicle mixed driving simulation method according to the above embodiment of the present invention may also have the following additional technical features:
[0011] Optionally, in one embodiment of the present invention, based on the virtual vehicle model and the communication channel, the information space is mapped to the real-time operating status of the physical system.
[0012] Optionally, in one embodiment of the present invention, the physical system includes one or more of roadside settings, traffic infrastructure and moving vehicles, and the model includes one or more of a roadside facility model, a traffic infrastructure model and a moving vehicle model.
[0013] Optionally, in one embodiment of the present invention, when constructing a corresponding model that reflects the real-time operating status of the physical system, it further includes: while modeling based on the static information of the physical system, building a state update interface in the data layer of the model, and using the state update interface to receive the state true value, and updating and optimizing the model parameters according to the received state true value.
[0014] Optionally, in one embodiment of the present invention, it also includes: using a communication channel to feed back the simulation operation results as reference decision information to the intelligent connected vehicles contained in the physical space.
[0015] To achieve the above-mentioned purpose, another embodiment of the present invention proposes a digital twin virtual-real multi-vehicle mixed running simulation device, including: a model construction module, which is used to construct a virtual vehicle model with similar dynamic characteristics to the smart connected vehicle in the physical space in a computer based on digital twin modeling technology; a scene construction module, which is used to construct an information interaction channel between the virtual vehicle model and the smart connected vehicle based on communication, and to construct a multi-vehicle collaborative scene in the information space; a result analysis and feedback module, which is used to use the virtual vehicle model and the smart connected vehicle in the multi-vehicle collaborative scene to perform mixed running simulation to obtain the simulation operation results of the virtual-real multi-vehicle mixed running.
[0016] The digital twin virtual-real multi-vehicle mixed simulation device of the embodiment of the present invention replaces the serial simulation process of the traditional simulation software / platform with a parallel digital twin solution, and has the characteristics of low time cost and high scene flexibility. By introducing a virtual car, it simulates an intelligent connected car, which solves the problem that real car experiments requiring multiple vehicles are difficult to carry out. It provides a possible experimental platform for research requiring multiple vehicles, such as queue control and multi-vehicle formation, based on which the effectiveness of the multi-vehicle collaborative algorithm can be verified.
[0017] In addition, the digital twin virtual-real multi-vehicle mixed driving simulation device according to the above embodiment of the present invention may also have the following additional technical features:
[0018] Optionally, in one embodiment of the present invention, the model building module is further used to map the real-time operating status of the physical system in the information space based on the virtual vehicle model and the communication channel.
[0019] Optionally, in one embodiment of the present invention, the physical system includes one or more of roadside settings, traffic infrastructure and moving vehicles, and the model includes one or more of a roadside facility model, a traffic infrastructure model and a moving vehicle model.
[0020] Optionally, in one embodiment of the present invention, it also includes: an update module, which is used to build a state update interface in the data layer of the model while modeling based on the static information of the physical system when constructing a corresponding model that reflects the real-time operating status of the physical system, and use the state update interface to receive the state true value, and update and optimize the model parameters according to the received state true value.
[0021] Optionally, in one embodiment of the present invention, the result analysis and feedback module is further used to use a communication channel to feed back the simulation operation results as reference decision information to the intelligent connected vehicles contained in the physical space.
[0022] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0024] Figure 1 This is a flow chart of a digital twin virtual-real multi-vehicle mixed driving simulation method according to an embodiment of the present invention;
[0025] Figure 2 A schematic diagram of the main components of a digital twin simulation system for an intelligent connected vehicle according to an embodiment of the present invention;
[0026] Figure 3 A schematic diagram of the internal information interaction form of a physical system according to an embodiment of the present invention;
[0027] Figure 4 A schematic diagram of information space component modules according to an embodiment of the present invention;
[0028] Figure 5 A schematic diagram of interactive information flow of a virtual car according to an embodiment of the present invention;
[0029] Figure 6 A schematic diagram of a physical / cyber space information interaction process according to an embodiment of the present invention;
[0030] Figure 7 A schematic diagram of a virtual car simulation deceleration process according to an embodiment of the present invention;
[0031] Figure 8 It is a structural schematic diagram of a digital twin virtual-real multi-vehicle mixed traffic simulation device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0032] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.
[0033] Digital Twins (DTs) is a key technology for mapping physical systems (physical space) to digital models of information space (virtual space). It makes full use of sensors placed in various parts of the system to analyze and model physical entities, and reflects the entire life cycle of physical systems in different real scenarios. With the help of various high-performance sensors and reliable communications, digital twin technology can inherit the data of multi-dimensional physical entities, supplemented by data analysis and simulation, and present the actual situation of physical entities in near real time.
[0034] At present, the simulation process of simulation software / platform for intelligent networked vehicles is mostly serial, and there are problems such as difficulty in dynamic updating and flexible adjustment after parameter preset. In view of this problem, the present invention introduces digital twin technology into the simulation field of intelligent networked vehicles, and for the first time proposes a digital twin simulation system for intelligent networked vehicles for general scenarios, and explains the system composition, functions of each part, etc. In addition to the "mapping" connotation of digital twins, the existence of "virtual vehicles" is introduced, which expands the technical connotation of digital twins. By introducing virtual vehicles, attempts are made to use them to represent the real intelligent networked vehicles in the physical system, solving the problem of difficulty in conducting multi-vehicle experiments.
[0035] The following describes the digital twin virtual-real multi-vehicle mixed traffic simulation method and device proposed according to an embodiment of the present invention with reference to the accompanying drawings.
[0036] First, the digital twin virtual-real multi-vehicle mixed driving simulation method proposed according to an embodiment of the present invention will be described with reference to the accompanying drawings.
[0037] Figure 1 The present invention is a flowchart of a digital twin virtual-real multi-vehicle mixed driving simulation method according to an embodiment of the present invention.
[0038] like Figure 1 As shown, the digital twin virtual-real multi-vehicle mixed driving simulation method includes the following steps:
[0039] In step S101, a virtual vehicle model with similar dynamic characteristics to the smart connected vehicle in the physical space is constructed in a computer based on the digital twin modeling technology.
[0040] It can be understood that by building a virtual vehicle model in a computer based on digital twin modeling technology, it can have similar dynamic characteristics to the intelligent connected vehicle in the physical space. The constructed vehicle model can reflect the real-time operating status of the intelligent connected vehicle in the corresponding physical space.
[0041] Optionally, in one embodiment of the present invention, based on the virtual vehicle model and the communication channel, the information space is mapped to the real-time operating status of the physical system.
[0042] Optionally, in one embodiment of the present invention, the physical system includes one or more of roadside settings, traffic infrastructure and moving vehicles, and the model includes one or more of a roadside facility model, a traffic infrastructure model and a moving vehicle model.
[0043] Optionally, in one embodiment of the present invention, when constructing a corresponding model that reflects the real-time operating status of a physical system, it further includes: while modeling based on the static information of the physical system, building a state update interface in the data layer of the model, and using the state update interface to receive the state true value, and updating and optimizing the model parameters according to the received state true value.
[0044] First, the components of the digital twin test and verification system for intelligent connected vehicles are introduced. The digital twin system is generally composed of a physical system (physical space) and an information system (information space). The digital twin system for testing and verification of intelligent connected vehicles proposed in this invention is composed of the following components: Figure 2 As shown. The physical space mainly includes various elements in the real world and the connections between elements. Combined with specific application scenarios, the physical system referred to in the present invention is the transportation system. For the transportation system, its components mainly include roadside facilities, transportation infrastructure, and vehicles; the information space mainly includes models of the above-mentioned elements, which realize real-time mapping of the physical space through state updates, and feed back reference decisions to the corresponding entities in the physical system. Through the two-way interaction and collaboration between the physical space and the information space, the model in the information space can be gradually optimized while realizing state synchronization with the corresponding components in the physical space. The components in the physical system realize their own dynamic adjustment by executing the control instructions of the corresponding models of the information system.
[0045] like Figure 2 As shown in the figure, for the traffic system, the roadside facilities include both roadside control units such as traffic lights and various sensors and communication equipment deployed on the roadside. The roadside equipment achieves target-level perception by monitoring the dynamic changes in the area, such as the speed, position, vehicle head direction, pedestrian speed, etc., and can transmit its own perception results to the information space through communication channels to achieve the status update of each element of the information space.
[0046] In the embodiment of the present invention, the traffic infrastructure mainly includes road environment, buildings, various traffic signs, etc., which constrain the drivable path of the vehicle and affect the vehicle decision.
[0047] The vehicles in the embodiments of the present invention include not only the intelligent and connected vehicles (ICVs) in the scene, but also the legacy vehicles (LVs). The main information interaction forms of the physical system are as follows: Figure 3As shown, it includes vehicle-to-vehicle information interaction (V2V) and vehicle-to-road information interaction (V2I). Among them, the roadside communication unit can not only directly receive the driving status of the ICV through V2I, but for traditional cars, it can also monitor the dynamic changes of the area through the roadside computing unit, and calculate the real-time status of such vehicles such as speed, position, and direction. For intelligent connected vehicles, they transmit their own position, speed, direction and other information to the outside through the on-board communication unit; within the vehicle's own communication range, the status of other intelligent connected vehicles based on V2V is obtained. Based on the above networked information interaction, intelligent connected vehicles can achieve over-distance perception, break through the perception limitations of single vehicles, and achieve more reasonable decisions by integrating global information.
[0048] In step S102, an information interaction channel between the virtual vehicle model and the intelligent connected vehicle is constructed based on communication, and a multi-vehicle collaborative scenario is constructed in the information space.
[0049] The information space mainly contains the models of the elements of the physical transportation system. By receiving data from the physical system, it realizes the real-time state mapping of the physical system. When constructing the information space, that is, completing the modeling of the above elements, it is necessary to realize modeling of different granularities according to the characteristics of each element and the actual application needs.
[0050] Generally speaking, models in information space can be divided into two types: static models and dynamic models. Static models usually do not need to interact with other environmental factors or update their status at high frequency, but they directly or indirectly affect vehicle decision-making or control results. For example, the parameters of the road model, such as the road length and width, usually do not change frequently, but the road surface material affects the road adhesion coefficient, which in turn affects the braking performance of the vehicle. Compared with static models, the parameters or states of dynamic models are usually updated at high frequency and continuously changed. For example, the position and speed of the vehicle are often in a continuous change process, and the phase of traffic lights also shows periodic changes. The state update of the dynamic model represents the real-time operation status change of the corresponding physical entity in the information space. The quality of modeling is directly related to the representation and feedback capabilities of the information space for the physical system. The components in the physical system realize their own dynamic adjustment by executing the control instructions of the corresponding model of the information system. The two are in a parallel cycle of optimization and evolution.
[0051] In addition, the information space itself can also be used as an independent platform to achieve offline simulation by presetting model parameters. It should be pointed out in particular that the modeling of element entities not only includes general three-dimensional modeling, but also includes behavior and rule modeling that conforms to objective laws. The construction of the information space is mainly the modeling of the physical entities mentioned above and the construction of interaction rules between models. In particular, because the information space to be constructed by the present invention can not only realize the real-time mapping of the physical system, the model in the information space can send its results directly to the corresponding physical entity as reference information after simulation and calculation. For example, the vehicle model obtains the global information in a certain area and calculates the reference driving speed at the next moment and sends it to the corresponding intelligent connected vehicle as reference information during its driving process. That is, the digital twin simulation system including the physical system and the information space referred to in this article is a system that specifically includes two-way real-time information interaction between the two. In order to realize the generation of reference decisions and the mapping of the physical system, the present invention proposes the constituent modules of the information space such as Figure 4 shown.
[0052] The above introduces the composition of the digital twin system formed by the application of digital twin technology to the simulation of intelligent connected vehicles and the main functions of each part, taking into account the following characteristics of vehicles in the physical system: there are certain safety hazards in multiple real vehicle experiments; when the number of vehicles required for the scene is large, the available number may not meet the demand and the cost is high; the real vehicle experiment is severely restricted by weather, site and other conditions. Therefore, when constructing vehicles in the information space, the mapped vehicles are removed, and a cloud car model (virtual vehicle model) that has no corresponding relationship with the elements of the physical system is added. The characteristics of the cloud car model include: it can integrate all vehicle states in the information space, including mapped vehicles and other cloud car models, complete the acquisition of global vehicle states, and achieve global optimization; it has similar dynamic and static characteristics to real intelligent connected vehicles, such as geometric characteristics, so it can be used to simulate the movement of real intelligent connected vehicles; the model is reusable, the parameters are easy to modify, and the maintenance cost is low. For the above-mentioned scenarios involving cloud vehicles, the interactive information flow between the physical system and the information space is as follows Figure 5 shown.
[0053] like Figure 5 As shown in Figure 2, the cloud car model can achieve global perception of scene vehicle information. Figure 5As shown in (b), when the number of intelligent networked vehicles in the scene is small, by constructing a cloud car model in the information space, the effect of queuing as shown in (a) of 5 can be simulated. The advantages of introducing a cloud car model in the information space are as follows: compared with the real intelligent networked vehicles in the physical system, the cloud car model has the characteristics of low maintenance cost and adjustable parameters; multiple real car experiments are often costly and difficult, and are easily affected by the environment, so it is difficult to carry out multiple real car experiments. The present invention adds a cloud car model in the information space and feeds back the cloud car model state information to the intelligent networked vehicles in the physical system, constructs a digital object of the cloud car model in the physical space, simulates the state perception of the virtual vehicle by the intelligent networked vehicles in the physical system, and makes its own decision based on the perception results and the corresponding model reference information in the information space. Under the premise of reliable information transmission and effective modeling, the method proposed by the present invention includes both a real car verification part and a simulation part, which can effectively solve the problem that multiple real car experiments are difficult to carry out due to the insufficient number of available real intelligent networked vehicles. The proposed method takes into account the connotation of "two-way interaction" of the physical / cyber space of the digital twin and extends the boundary of "real-time mapping" of the digital twin by introducing a virtual vehicle of the non-mapped car.
[0054] In step S103, a virtual vehicle model and an intelligent connected vehicle in a multi-vehicle collaborative scenario are used to perform mixed running simulation to obtain a simulation running result of virtual and real multi-vehicle mixed running.
[0055] Optionally, in one embodiment of the present invention, simulation is performed using a virtual vehicle model and a multi-vehicle collaborative scenario, including: using a communication channel to feed back the simulation operation results as reference decision information to the intelligent connected vehicles contained in the physical space.
[0056] The above introduces the digital twin system for intelligent connected vehicle simulation and explains the specific idea of using cloud car models to simulate physical intelligent connected vehicles and constructing digital objects in physical space to form a virtual and real multi-vehicle hybrid. The following content describes the information interaction process from the cyber space and the physical system:
[0057] like Figure 6 As shown in the figure, at time t0, the physical system sends the current state of each entity to the information space, mainly including information such as location, speed, direction, and current lane of intelligent connected vehicles and traditional vehicles. After receiving the real-time state from the physical system, the information space combines historical data to calculate reference decisions that are suitable for each physical entity, such as vehicle lane change, acceleration and deceleration, etc.; in some cases, control instructions can also be sent directly, such as turning on street lights when a vehicle is approaching; for other traffic elements included in the scene, such as traffic lights, the information space can also directly give reference phases.
[0058] For the specific description of virtual and real multi-vehicle mixed traffic and the information interaction mechanism, Figure 7 The vehicle queue shown in the figure is taken as an example to illustrate. At time t2, a deceleration command is applied to the virtual car, and the deceleration intention is sent to the smart connected cars in the nearby range in the physical space. For the smart connected cars in the adjacent area, the process of decelerating the front car is simulated. Since there is no information interaction between the traditional car and the virtual car, the virtual car's deceleration intention will not affect the traditional car, which maintains a constant speed from t1 to t3. For the smart connected car behind it that receives the virtual car's deceleration intention, it starts braking to ensure safety, and the longitudinal distance with the traditional car increases after a period of time. The result of the data interaction between virtual and real vehicles in the physical space and the information space on the movement process of vehicles in the physical space is shown as follows. Figure 7 shown.
[0059] It can be understood that, according to the above introduction, the present invention proposes a method for applying digital twin technology to the field of intelligent connected vehicle testing and verification, and proposes a digital twin test system for intelligent connected vehicles. For the virtual part of the system (information space), based on the established physical system, the present invention proposes an information space modeling method based on the physical traffic system. The information space constructed by the present invention has the following characteristics. First, based on the physical traffic system, the information space can reflect its real-time operating status. Secondly, based on the operating results, the information space can generate reference decisions for real intelligent connected vehicles and realize feedback. In order to construct an information space with the above characteristics, for the physical traffic scene, when abstracting it for modeling, the traffic elements contained in the physical system can be divided into three parts: roadside facilities, traffic infrastructure and vehicles. Among them, roadside facilities and traffic infrastructure constitute the static environment for vehicle operation. Considering the complexity of scene modeling, for various types of roadside facilities, consider building a top-level data interface at the data layer to achieve the state acquisition of the target object. For example, when modeling the roadside camera, consider the vehicle status information fed back by it, rather than the image content it captures. For traffic infrastructure, in order to reduce the difficulty of modeling, when modeling it, it should be able to feedback the geometric information, association relationships, etc. of physical objects. For intelligent networked vehicles in physical systems, their real-time operating status is the decision-making input of the information space. Therefore, when modeling them, not only static information such as geometry, material, etc. is required, but also a state update interface needs to be built in the data layer so that it can reflect the real-time operating status of real intelligent networked vehicles, such as position, speed, and vehicle head direction. Based on the modeling rules of the above-mentioned traffic elements, the present invention further elaborates on the component modules of the information space, which mainly include basic modules and intelligent network-related modules to realize the generation of reference decisions.
[0060] In view of the current situation that multiple real-car experiments are difficult to carry out, the present invention proposes a method that combines real-car experiments with simulation verification. Based on the intelligent networked vehicles existing in the physical space, a virtual vehicle model is constructed in the information space (computer simulation environment) so that the two have similar dynamic characteristics. And based on communication, an information interaction channel between the virtual vehicle model and the real intelligent networked vehicle is constructed, and a multi-vehicle collaborative scene is constructed in the information space.
[0061] Specifically, a virtual vehicle model (cloud car model) and a mapping vehicle model without entity correspondence are constructed in the information space (computer simulation environment). The cloud car model has no correspondence with the elements in the physical traffic system, but it has similar dynamic characteristics to the intelligent connected car. Therefore, the cloud car model can be used to simulate the motion characteristics of the intelligent connected car. The mapping vehicle model corresponds to the intelligent connected car that actually exists in the physical system, and can reflect the real-time operation status of the corresponding vehicle through communication. The interaction between the cloud car and the mapping car can be used to simulate the multi-vehicle collaborative scenario, and generate reference decisions for real vehicles based on the scenario, and then feedback the decision information to the intelligent connected cars contained in the physical space through communication channels, solving the problem of insufficient number of available real intelligent connected cars. Among them, the intelligent connected cars and various traffic elements in the above physical space belong to the real car experiment, and the mapping content and cloud car model run by the computer simulation environment belong to the simulation part. The above two can be combined and run simultaneously through communication means to form a digital twin system in which the real object and the simulation interact and influence each other, providing a new idea for conducting multi-real car experiments.
[0062] The digital twin virtual-real multi-vehicle mixed simulation method proposed in the embodiment of the present invention replaces the traditional simulation software / platform serial simulation process with a parallel digital twin solution, which has the characteristics of low time cost and high scene flexibility. By introducing a virtual vehicle model, it simulates an intelligent connected vehicle, which solves the problem that real vehicle experiments requiring multiple vehicles are difficult to carry out. It provides a possible experimental platform for research requiring multiple vehicles, such as queue control and multi-vehicle formation, and verifies the effectiveness of the algorithm based on this.
[0063] Secondly, the digital twin virtual-real multi-vehicle mixed traffic simulation device proposed according to an embodiment of the present invention is described with reference to the accompanying drawings.
[0064] Figure 8 It is a structural schematic diagram of a digital twin virtual-real multi-vehicle mixed traffic simulation device according to an embodiment of the present invention.
[0065] like Figure 8 As shown, the digital twin virtual-real multi-vehicle mixed driving simulation device includes: a model construction module 100, a scenario construction module 200 and a simulation module 300.
[0066] The model building module 100 is used to build a virtual vehicle model with similar dynamic characteristics to the intelligent connected vehicle in the physical space in the computer based on the digital twin modeling technology. The scenario building module 200 is used to build an information interaction channel between the virtual vehicle model and the intelligent connected vehicle based on communication, and to build a multi-vehicle collaborative scenario in the information space. The result analysis and feedback module 300 is used to use the virtual vehicle model and the intelligent connected vehicle in the multi-vehicle collaborative scenario to perform mixed running simulation and obtain the simulation operation results of virtual and real multi-vehicle mixed running.
[0067] Optionally, in one embodiment of the present invention, the model building module is further used to map the real-time operating status of the physical system in the information space based on the virtual vehicle model and the communication channel.
[0068] Optionally, in one embodiment of the present invention, the physical system includes one or more of roadside settings, traffic infrastructure and moving vehicles, and the model includes one or more of a roadside facility model, a traffic infrastructure model and a moving vehicle model.
[0069] Optionally, in one embodiment of the present invention, it also includes: an update module, which is used to build a state update interface in the data layer of the model while modeling based on the static information of the physical system when constructing a corresponding model that reflects the real-time operating status of the physical system, and use the state update interface to receive the state true value, and update and optimize the model parameters according to the received state true value.
[0070] Optionally, in one embodiment of the present invention, the result analysis and feedback module is further used to use a communication channel to feed back the simulation operation results as reference decision information to the intelligent connected vehicles contained in the physical space.
[0071] It should be noted that the above explanation of the digital twin virtual-real multi-vehicle mixed traffic simulation method embodiment is also applicable to the digital twin virtual-real multi-vehicle mixed traffic simulation device of this embodiment, and will not be repeated here.
[0072] The digital twin virtual-real multi-vehicle mixed simulation device proposed in the embodiment of the present invention replaces the serial simulation process of the traditional simulation software / platform with a parallel digital twin solution, which has the characteristics of low time cost and high scene flexibility. By introducing a virtual car, it simulates an intelligent connected car, which solves the problem that real car experiments requiring multiple vehicles are difficult to carry out. It provides a possible experimental platform for research requiring multiple vehicles, such as queue control and multi-vehicle formation, based on which the effectiveness of the multi-vehicle collaborative algorithm can be verified.
[0073] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0074] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0075] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.
Claims
1. A digital twin virtual-real multi-vehicle mixed driving simulation method, characterized in that: The following steps are involved: Based on the digital twin modeling technology, a virtual vehicle model with similar dynamic characteristics to the intelligent connected vehicle in the physical space is constructed in the computer, wherein, based on the virtual vehicle model and the communication channel, the information space is mapped to the real-time operation state of the physical system, the physical system includes one or more of the roadside facilities, the traffic infrastructure and the moving vehicle, and the model includes one or more of the roadside facilities model, the traffic infrastructure model and the moving vehicle model; An information interaction channel between the virtual vehicle model and the intelligent networked vehicle is constructed based on communication, and a multi-vehicle collaborative scenario is constructed in the information space. Specifically, a digital object of the virtual vehicle model is constructed based on communication, and the state perception of the digital object of the virtual vehicle model by the intelligent networked vehicle is simulated, and self-decision is made based on the state perception and the corresponding model reference information of the information space, so as to obtain the multi-vehicle collaborative scenario; as well as The virtual vehicle model and the intelligent connected vehicles in the multi-vehicle collaborative scenario are used to perform mixed running simulation to obtain simulation results of mixed running of virtual and real multiple vehicles.
2. The method according to claim 1, characterized in that When constructing a corresponding model reflecting the real-time operating state of the physical system, it further includes: While modeling based on the static information of the physical system, a state update interface is constructed in the data layer of the model, and the state true value is received by using the state update interface, and the model parameters are updated and optimized according to the received state true value.
3. The method according to claim 1, characterized in that Also includes: The simulation operation results are fed back to the intelligent connected vehicles contained in the physical space as reference decision information through communication channels.
4. A digital twin virtual-real multi-vehicle mixed driving simulation device, characterized in that: include: A model building module, used to build a virtual vehicle model with similar dynamic characteristics to the intelligent connected vehicle in the physical space in a computer based on the digital twin modeling technology, wherein the model building module is further used to map the real-time operating state of the physical system in the information space based on the virtual vehicle model and the communication channel, wherein the physical system includes one or more of the roadside facilities, the traffic infrastructure and the moving vehicle, and the model includes one or more of the roadside facilities model, the traffic infrastructure model and the moving vehicle model; A scenario construction module, configured to construct an information interaction channel between the virtual vehicle model and the intelligent networked vehicle based on communication, and to construct a multi-vehicle collaborative scenario in the information space. The scenario construction module is further configured to construct a digital object of the virtual vehicle model based on communication, simulate the state perception of the digital object of the virtual vehicle model by the intelligent networked vehicle, and make its own decision based on the state perception and the corresponding model reference information of the information space, so as to obtain the multi-vehicle collaborative scenario; and The result analysis and feedback module is used to use the virtual vehicle model and the intelligent connected vehicle in the multi-vehicle collaborative scenario to perform mixed running simulation to obtain the simulation operation results of virtual and real multi-vehicle mixed running.
5. The device according to claim 4, characterized in that Also includes: The update module is used to build a corresponding model that reflects the real-time operating status of the physical system. While modeling based on the static information of the physical system, it builds a state update interface in the data layer of the model, and uses the state update interface to receive the state true value, and updates and optimizes the model parameters according to the received state true value.
6. The device according to claim 4, characterized in that The result analysis and feedback module is further used to use a communication channel to feed back the simulation operation results as reference decision information to the intelligent connected vehicles contained in the physical space.
