A communication middleware system for intelligent networked vehicle joint simulation

Abstract

The application discloses a communication middleware system for intelligent networked vehicle joint simulation, and relates to the field of intelligent networked vehicle joint simulation, and comprises a communication middleware and a simulation platform; data interaction and collaborative operation are carried out between the simulation platforms through the communication middleware; the communication middleware comprises a basic communication capability module and an intelligent collaborative enhancement module; the basic communication capability module comprises a plug-in interface module, a data communication module and a protocol conversion module, and is used for realizing flexible access, data transmission synchronization and protocol unification between multiple simulation platforms; the intelligent collaborative enhancement module comprises a dynamics collaborative module and a scene deduction module; the dynamics collaborative module models target vehicles and environment vehicles through a hierarchical modeling mode; the scene deduction module performs dynamic interaction and behavior evolution between multiple subjects based on a traffic participant behavior model. The application realizes efficient collaborative operation and unified management of multiple simulation systems.

Description

Technical Field

[0001] This invention relates to the field of joint simulation technology for intelligent connected vehicles, and more specifically to a communication middleware system for joint simulation of intelligent connected vehicles. Background Technology

[0002] In the development of intelligent connected vehicles, software simulation testing plays an irreplaceable role. While traditional road testing and closed-loop testing can verify vehicle performance, they suffer from drawbacks such as low testing efficiency, high costs, and difficulty in reproducing dangerous scenarios. In contrast, simulation testing can not only complete large-scale testing quickly and at low cost, but also safely simulate various dangerous scenarios, providing an efficient and safe verification environment for the research and development of intelligent connected vehicles. For example, Waymo's autonomous driving system completed 27 million miles of testing in one year through simulation testing, a distance that would require 100 autonomous vehicles operating 24 hours a day for 250 years in the real world to achieve. This data clearly demonstrates the importance of simulation testing in the development of intelligent connected vehicles, accelerating technological iteration and maturity while reducing R&D costs and risks.

[0003] However, current simulation testing of intelligent connected vehicles mainly relies on various simulation platforms, including vehicle simulation platforms (such as CarMaker and CarSim), traffic flow simulation platforms (such as VISSIM and SUMO), and scenario simulation platforms (such as CARLA and VTD). Each of these platforms has its advantages and disadvantages: vehicle simulation platforms offer highly realistic vehicle dynamics models, capable of simulating real vehicle control systems, but lack diversity in traffic flow and scenarios; traffic flow simulation platforms offer fast testing speeds and large scales, but the scenarios are fixed and scripted, and the vehicle dynamics are not realistic enough; scenario simulation platforms excel in scenario modeling and vehicle perception algorithm testing, but suffer from unrealistic vehicle dynamics and fixed traffic flow scenarios. Furthermore, existing simulation platforms generally lack network communication simulation capabilities, making it impossible to verify the reliability of vehicle-road-cloud communication networks for intelligent connected vehicles.

[0004] Therefore, how to build a joint simulation platform to conduct comprehensive simulation of intelligent connected vehicles is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, and in response to the problems existing in the joint simulation of intelligent connected vehicles, such as independent simulation platforms, inconsistent communication interfaces, poor system scalability, and difficulty in balancing simulation efficiency and accuracy, this invention provides a communication middleware system for joint simulation of intelligent connected vehicles, so as to achieve efficient collaborative operation and unified management of multiple simulation systems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: This invention discloses a communication middleware system for joint simulation of intelligent connected vehicles, comprising: a communication middleware and a simulation platform; the simulation platforms interact and operate collaboratively through the communication middleware. The communication middleware includes a basic communication capability module and an intelligent collaboration enhancement module; The basic communication capability module includes a plug-in interface module, a data communication module, and a protocol conversion module, which are used to enable flexible access between multiple simulation platforms, data transmission synchronization, and protocol unification. The intelligent collaboration enhancement module includes a dynamics collaboration module and a scene interpretation module; The dynamics collaboration module models the target vehicle and the environmental vehicle separately using a hierarchical modeling approach; The scenario interpretation module is based on the traffic participant behavior model to enable dynamic interaction and behavioral evolution among multiple subjects.

[0007] Furthermore, the simulation software of the simulation platform includes: vehicle simulation software, traffic flow simulation software, scene simulation software, and communication simulation software.

[0008] Furthermore, the plug-in interface module provides unified data access for different simulation software through standardized interface definitions; The plug-in interface module adopts a publish / subscribe mechanism for data interaction, and accesses data from an external simulation platform by subscribing to a specified data channel; The plugin interface module processes the received data through a callback function mechanism, decoupling the data reception and business processing logic.

[0009] Furthermore, the data communication module encapsulates the traffic flow status data obtained from the simulation platform through the message subscription mechanism into a unified communication message and writes it into the message buffer queue; the data communication module performs asynchronous reception and sequential processing of data through the queue mechanism and time synchronization mechanism, and obtains the synchronized original data set after completing time synchronization.

[0010] Furthermore, the protocol conversion module parses and processes the received multi-source data and converts it into a unified data structure, specifically including: Data parsing: Decode the received raw communication data and extract traffic participant status information from the communication data packets; Data mapping: According to the preset data mapping rules, the data fields of traffic participant status information extracted from different simulation platforms are mapped to a unified data model; Structure encapsulation: Encapsulate the parsed and mapped data into a unified vehicle object data structure; The data parsing process also includes filtering out abnormal data or handling errors.

[0011] Furthermore, the dynamics collaboration module is constructed based on a hierarchical dynamics modeling mechanism, including: Automotive simulation software is used to model and calculate key vehicles to obtain high-precision dynamic models of the key vehicles. Traffic flow simulation software was used to model the environmental vehicles, resulting in a simplified dynamic model of the environmental vehicles.

[0012] Furthermore, the scene deduction module includes four parts: scene state perception, behavior decision generation, random event triggering, and scene dynamic update, which are used for dynamic interaction and multi-agent behavior evolution in traffic scenes.

[0013] As can be seen from the above technical solution, compared with the prior art, the present invention provides a communication middleware system for joint simulation of intelligent connected vehicles, which has the following beneficial effects: (1) Breaking down technical barriers between simulation platforms and achieving multi-simulation software integration: The communication middleware of this invention breaks down the barriers between traditional simulation platforms, achieving a powerful integration between automotive simulation software (such as CarMaker), traffic flow simulation software (such as VISSIM), scene simulation software (such as CARLA), and communication simulation software (such as OMNeT++). Automotive simulation software provides high-precision vehicle dynamics, traffic flow simulation software provides diverse traffic scenarios, scene simulation software provides realistic driving scenarios, and communication simulation software simulates data latency, packet loss, and other characteristics, reflecting the features of vehicle-road-cloud communication. Furthermore, through an LCM-based communication framework, data synchronization and interaction between the various simulation software are achieved, thus forming a highly integrated joint simulation platform.

[0014] (2) Support for plug-in simulation software replacement: The system provides a complete simulation software communication interface and supports plug-in simulation software replacement. Users can choose different simulation software to combine according to their needs to meet different simulation requirements.

[0015] (3) Supports real-time simulation of large-scale traffic flow: The system uses multi-threading and other computational optimization techniques, which greatly improves its simulation speed. Real-time simulation of large-scale traffic flow can be achieved on ordinary laptops, and it has strong simulation efficiency.

[0016] (4) Achieving synergistic optimization of dynamic modeling accuracy and simulation efficiency: This invention uses a hierarchical dynamic modeling mechanism to simulate the target vehicle using a high-precision dynamic model and to calculate the environmental vehicle using a lightweight dynamic model. This ensures the simulation accuracy of key vehicles while reducing the overall computational burden and communication overhead, thus achieving a balance between simulation accuracy and efficiency.

[0017] (5) Realizing dynamic evolution and interactive generation of traffic scenarios to improve simulation realism: This invention enables traffic participants to make autonomous decisions and generate interactive behaviors based on environmental conditions through a self-deductive mechanism of traffic scenarios, thereby realizing non-preset, dynamically evolving traffic scenarios. Compared with traditional script-based scenario generation methods, this method can generate traffic environments with randomness, interactivity, and complexity, significantly improving the realism and diversity of simulation results. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 This invention provides a schematic diagram of a communication middleware system architecture for joint simulation of intelligent connected vehicles.

[0020] Figure 2 This invention provides a schematic diagram of a vehicle dynamics coordination and state consistency mechanism for co-simulation.

[0021] Figure 3 This is a schematic diagram of a behavior-driven self-deductive method for traffic scenarios provided by the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention discloses a communication middleware system for co-simulation of intelligent connected vehicles, such as... Figure 1 As shown, it includes: communication middleware and simulation platforms; the simulation platforms interact and operate collaboratively with each other through the communication middleware, including exchanging vehicle status data, traffic flow status data, control data and communication network data; The communication middleware includes a basic communication capability module and an intelligent collaboration enhancement module; The basic communication capability module includes a plug-in interface module, a data communication module, and a protocol conversion module, which are used to enable flexible access between multiple simulation platforms, data transmission synchronization, and protocol unification. The intelligent collaboration enhancement module includes a dynamics collaboration module and a scene interpretation module; The dynamics collaboration module models the target vehicle and the environment vehicle separately using a hierarchical modeling approach; The scenario rendering module is based on the traffic participant behavior model to enable dynamic interaction and behavioral evolution among multiple subjects.

[0024] In one specific embodiment, the simulation software of the simulation platform includes: vehicle simulation software, traffic flow simulation software, scene simulation software, and communication simulation software.

[0025] In one specific embodiment, the plug-in interface module is used to achieve unified access and communication interface expansion for different simulation software, thereby improving the system's flexibility and scalability. Specifically, the plug-in interface module provides unified data access for different simulation software through standardized interface definitions; each simulation platform publishes simulation data to the communication middleware by registering the corresponding data channel, and the communication middleware performs unified reception and processing.

[0026] The plug-in interface module adopts a publish / subscribe mechanism for data interaction, allowing access to external simulation platform data by subscribing to specified data channels; for example, by subscribing to the vehicle status data channel, it receives vehicle status information sent by automotive simulation software.

[0027] The plugin interface module processes the received data through a callback function mechanism, decoupling data reception from business processing logic, thereby improving the modularity of the system.

[0028] Specifically, for any connected simulation platform, its published data channel can be represented as: ; in, Indicates the first One data publishing channel.

[0029] The set of communication middleware maintenance subscription channels is represented as follows: ; in, Indicates the first One subscription channel.

[0030] Furthermore, the matching relationship between publishing and subscribing channels is determined based on channel identifier, message topic, and data type. When the message topics of publishing and subscribing channels are consistent, they are considered to meet the matching condition. The matching function is expressed as follows: ; in, This indicates the message topic for the corresponding data channel. When the matching conditions are met, the communication middleware establishes a data connection and performs data forwarding; otherwise, data forwarding is not performed.

[0031] Furthermore, when the publishing channel generates data messages At that time, the communication middleware sends data to the corresponding subscriber based on the channel mapping relationship. Its data distribution process can be represented as follows: ; in, This represents the set of data received by the subscriber.

[0032] Meanwhile, the plugin interface module uses a callback function mechanism to decouple data processing from business logic. Let the callback function set be: ; in, Indicates subscription channel The corresponding callback function.

[0033] When the subscriber receives the data message When the time comes, the corresponding callback function will be automatically triggered: ; in, This indicates the output result after the callback function has processed it.

[0034] When a new simulation platform is added, only the corresponding data channel and callback function registration are required to join the communication middleware management system. No modification to the existing simulation platform interface configuration is needed, thus enabling dynamic expansion and plug-and-play functionality. Through the aforementioned publish / subscribe mechanism and callback function mechanism, data decoupling and unified management between different simulation platforms are achieved. This allows the communication middleware to support the dynamic access, replacement, and expansion of simulation software, improving the flexibility and scalability of the co-simulation system.

[0035] In one specific embodiment, the data communication module is used to realize data reception, caching, and time synchronization between multiple simulation platforms. Specifically, the data communication module encapsulates traffic flow state data obtained from the simulation platforms through a message subscription mechanism into communication messages and writes them into a message buffer queue, thereby achieving data decoupling between different simulation platforms. The data communication module uses a message queue as a data buffer structure to cache received traffic participant status information, including vehicle ID, location coordinates, speed information, and heading angle. The data communication module uses a queue mechanism and a time synchronization mechanism to asynchronously receive and sequentially process data. After time synchronization, a synchronized original data set is obtained, thereby avoiding data loss or blocking problems caused by inconsistent communication frequencies between different simulation platforms. Specifically: First, data is received from traffic flow simulation software, scenario simulation software, vehicle simulation software, and communication simulation software via a publish / subscribe mechanism, and the received raw data is represented as follows: ;in, Indicates the traffic participant number. and Indicates the vehicle's position coordinates. Indicates vehicle speed. Vehicle heading angle, Represents the data timestamp.

[0036] Subsequently, the data communication module writes the received data into the message buffer queue, constructing a message set: ; The messages are arranged in timestamp order, satisfying the following: This ensures the consistency of data processing order across different simulation platforms.

[0037] Furthermore, to ensure time consistency during the operation of multiple simulation platforms, the data communication module establishes a unified simulation step size, and its synchronization period is defined as: ; in, This indicates the traffic flow simulation step size. This indicates the step size for vehicle dynamics simulation. Indicates the simulation step size. This represents the simulation step size for the communication network. The unified simulation clock is represented as: ;in, Indicates the simulation cycle number.

[0038] After time synchronization is completed, the data communication module sorts and organizes the raw data in the message queue according to a unified simulation clock, forming a synchronized raw data set: The original synchronous dataset serves as the input data source for the protocol conversion module, used for subsequent data parsing, field mapping, and structure encapsulation.

[0039] In one specific embodiment, the protocol conversion module is used to perform data parsing and unified format conversion between different simulation platforms to solve the inconsistency problem in data structure and communication protocol of multi-source heterogeneous data. Specifically, the protocol conversion module parses and processes multi-source data received by the communication middleware, converting the raw communication data from traffic flow simulation software, vehicle simulation software, and scene simulation software into a unified data structure, thereby realizing data sharing and collaborative computing across simulation platforms.

[0040] Furthermore, the protocol conversion module parses and processes the received multi-source data and converts it into a unified data structure, specifically including: Data parsing: Decode the received raw communication data and extract traffic participant status information from the communication data packets, including vehicle identification, location coordinates, speed information, and heading angle; Data mapping: According to the preset data mapping rules, the data fields of traffic participant status information extracted from different simulation platforms are mapped to a unified data model. For example, the vehicle status data in traffic flow simulation software is mapped to the vehicle objects in scene simulation software. Structure encapsulation: The parsed and mapped data is encapsulated into a unified vehicle object data structure for subsequent simulation modules to call, thereby standardizing the data interface; The data parsing process also includes filtering or handling abnormal data to ensure the stability of system operation.

[0041] Specifically: First, the raw data from different simulation platforms are analyzed. Let the... The raw data output by each simulation platform is represented as follows: ;in, Indicates the target number. and Indicates the vehicle's position coordinates. Indicates vehicle speed. Indicates the vehicle's heading angle. Represents a timestamp.

[0042] Because different simulation platforms use different data formats, a unified data model is established: ; in, ID , x , y , v , θ and t These represent the unified target number, position coordinates, speed, heading angle, and time information, respectively.

[0043] Furthermore, the protocol conversion module completes field conversion according to a preset mapping relationship, and the mapping process is represented as follows: ;in, This represents a data mapping function used to map data fields from different simulation platforms to a unified data model. For example, for vehicle state data in a traffic flow simulation platform: After mapping, we get: ; This enables data unification between different simulation platforms.

[0044] After completing the field mapping, the conversion result is structurally encapsulated to form a unified vehicle object, represented as follows: ; For simulation platforms that can directly output vehicle acceleration information, the protocol conversion module directly reads the corresponding acceleration data; for simulation platforms that cannot directly output vehicle acceleration information, the vehicle acceleration is calculated based on the rate of change of velocity between adjacent time points. The calculation formula is as follows: ;in, This indicates the time interval between adjacent sampling times.

[0045] Furthermore, to ensure data quality, anomaly detection is performed on the parsed data. Let the received data be... Data is considered abnormal when any of the following conditions are met: ;or ;or ; in, This indicates the maximum permissible displacement threshold for a single vehicle cycle. This indicates the maximum permissible speed change threshold for a single vehicle cycle.

[0046] For detected abnormal data, the protocol conversion module executes corresponding processing strategies based on the abnormality type. Specifically, for timestamp-abnormal data, the data packet is discarded directly; for location-jumping abnormal data, interpolation correction is performed using the vehicle's state at the preceding and following times, as follows: ; For abrupt changes in velocity anomalies, a moving average method is used for smoothing, specifically: ; in, This indicates the corrected vehicle speed; Indicates the historical sampling rate; Indicates the length of the sliding window.

[0047] After anomaly handling, the standardized vehicle object The corresponding status parameters are updated, and the updated vehicle object is written to the data cache of the communication middleware for unified management and deployment by subsequent modules.

[0048] In one specific embodiment, the dynamics coordination module is used to enable collaborative operation between vehicle dynamics models of different accuracies in a multi-simulation system, thereby improving overall simulation efficiency while ensuring simulation accuracy. The dynamics coordination module is built based on a hierarchical dynamics modeling mechanism and includes: The key vehicle (i.e. the target vehicle in the simulation) is modeled and calculated using automotive simulation software to obtain a high-precision dynamic model of the key vehicle, thereby obtaining a high-precision vehicle operating state. The high-precision dynamic model adopts a joint modeling method of vehicle longitudinal dynamics, lateral dynamics and tire dynamics to solve the state, so as to improve the accuracy of vehicle motion state calculation. Traffic flow simulation software is used to model environmental vehicles, resulting in a simplified dynamic model of the vehicles. The simplified dynamic model achieves efficient calculation of large-scale vehicles through embedding. A kinematic monorail model or a traffic flow car-following model is used for state updates to reduce computational complexity and improve simulation efficiency.

[0049] Specifically, such as Figure 2 As shown, the left side of the figure shows a high-precision dynamic model of the key vehicle, while the middle side shows a simplified dynamic model of the environmental vehicle. Figure 2 As shown in the lower middle section, the key vehicle and the environmental vehicles interact with each other through a dynamic coordination and state synchronization mechanism in the communication middleware, achieving unified scheduling and coordinated operation among different dynamic models. The interacting data includes information such as vehicle position, speed, acceleration, and control commands. Simultaneously, as... Figure 2 As shown on the right, the dynamics collaboration module unifies and summarizes the vehicle state data output by different models, and realizes state synchronization and result consistency verification between the high-precision dynamics model and the simplified dynamics model through a multi-model result alignment mechanism.

[0050] Specifically, let the vehicle state output by the high-precision dynamics model at time k be: ; in, , These represent the vehicle's position coordinates. Indicates vehicle speed. Indicates vehicle acceleration. Indicates the vehicle's heading angle.

[0051] Let the simplified dynamics model output the vehicle state at the same time as follows: ; To eliminate state deviations caused by differences in sampling periods between different models, a unified synchronization step size is first established: Furthermore, the unified simulation time is represented as: ;in, This is the step size for a high-precision dynamic model; To simplify the step size of the dynamic model; This represents the simulation cycle number. Both types of model outputs are mapped to a unified time point. .

[0052] Then, the state error and position error between the two types of models are calculated. for: ; The speed error is expressed as: ; The heading angle error is expressed as: ; Furthermore, a comprehensive consistency evaluation index is constructed: ; in, , , , These represent the weighting coefficients corresponding to position error, velocity error, and heading angle error, respectively. To avoid misjudgment caused by instantaneous state fluctuations, a continuous periodic consistency evaluation mechanism is adopted. When N consecutive simulation cycles satisfy: When the results of the high-precision dynamic model and the simplified dynamic model are consistent, it is determined that the results are consistent; when the following conditions are met: When this occurs, the state correction mechanism is triggered.

[0053] Furthermore, the state correction employs a weighted fusion approach to update the simplified dynamic model state: ; in, These are the state fusion coefficients, used to control the strength of the correction from the high-precision dynamic model to the simplified dynamic model; when As the accuracy increases, the correction result becomes closer to the output of the high-precision dynamics model. The corrected vehicle state is then rewritten into the environmental vehicle model for calculation in the next simulation cycle, thus ensuring consistency in the results across different precision dynamics models.

[0054] Through the aforementioned multi-model result alignment mechanism and consistency evaluation mechanism, the collaborative operation between the high-precision dynamics model of key vehicles and the simplified dynamics model of environmental vehicles is realized, which reduces the overall computational overhead while ensuring simulation accuracy and improves the operating efficiency of the joint simulation system in large-scale traffic scenarios.

[0055] In a specific embodiment, the scene derivation module comprises four parts: scene state perception, behavior decision generation, random event triggering, and scene dynamic updating. It is used for dynamic interaction and multi-agent behavior evolution in traffic scenarios, thereby improving the realism, complexity, and test coverage of the joint simulation scenario. In traditional script-based scenarios, each traffic participant operates according to predefined trajectories and behavioral rules, resulting in a fixed and uninterrupted scene evolution process. However, in the self-derivation scenario of this invention, each traffic participant (including intelligent connected vehicles, randomly appearing vehicles, bicycles, and pedestrians) makes dynamic decisions based on the current environmental state and the behavior of surrounding traffic participants, achieving real-time interaction and behavior evolution among traffic participants. The behavior-driven traffic scene self-derivation method is as follows: Figure 3 As shown.

[0056] Specifically: (1) Scene state perception part First, obtain the state information of all traffic participants in the current simulation scenario and construct a scenario state set: ; in, Indicates time The scene state, Indicates the first Traffic participants include intelligent connected vehicles, private vehicles, bicycles, and pedestrians. The state of each traffic participant is defined as follows: ; in, , These represent the position coordinates, Indicates the speed of motion. Indicates the heading angle. This indicates the lane where the current traffic participant is located. The lane information includes the current lane number, the relationship between adjacent lanes, and lane attribute information, which is used to characterize the road environment in which the traffic participant is located.

[0057] (2) Behavioral decision generation part

[0058] Traffic participants make behavioral decisions based on their own state and the state of their surrounding environment. Let the set of executable behaviors for traffic participants be: ; in, Indicates keeping the lane open; Indicates acceleration; Indicates deceleration; Indicates a lane change; This indicates yielding. Based on the current scenario and the historical behavioral characteristics of traffic participants, a probabilistic behavioral decision model is established for traffic participants. The behavioral decision-making results are expressed as follows: That is, select the action with the highest probability of occurrence based on the current scenario state.

[0059] (3) Random event triggering part

[0060] To enhance the diversity and randomness of scenarios, a random event generation mechanism is introduced. Furthermore, to improve the coverage of dangerous and long-tail scenarios, when the time distance, collision time, or lateral safety distance between traffic participants meets preset danger conditions, the scenario rendering module increases the trigger probability of the corresponding random event to generate dangerous traffic scenarios such as cutting in line, sudden deceleration, crossing the road, and sudden appearance of obstacles.

[0061] Let the set of random events be: ;in, This indicates the appearance of a random vehicle; This indicates the appearance of random pedestrians; This indicates that a random bicycle will appear; This indicates random obstacle generation. To assess the potential collision risk between traffic participants, a collision time metric is introduced: ;in, Indicates the relative distance between two traffic participants. Represents relative velocity. When: When a potential collision risk is identified in the current scene, the probability of triggering the corresponding random event is increased. The probability of triggering a random event is expressed as: ;in, Indicates traffic density. Indicates the road environment condition. When the following conditions are met: When the event occurs, a corresponding random event is triggered and the scene state is updated.

[0062] (4) Scene dynamic update section

[0063] The scene state is updated based on the behavioral decisions of traffic participants and the results of random event generation: ;in, This represents the scene state update function.

[0064] Furthermore, the updated traffic participant status is re-input into the behavior decision module to realize the behavior prediction for the next cycle, thus forming a continuous closed-loop traffic scenario evolution process.

[0065] Through the above steps, traffic participants no longer operate according to a preset script, but make autonomous decisions and interact dynamically based on the environmental conditions, realizing the transformation from traditional script-driven scenarios to behavior-driven scenarios, thereby significantly improving the realism, complexity, and coverage of dangerous scenarios in traffic scenarios.

[0066] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0067] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A communication middleware system for joint simulation of intelligent connected vehicles, characterized in that, include: Communication middleware and simulation platform; the simulation platforms interact and operate collaboratively with each other through the communication middleware. The communication middleware includes a basic communication capability module and an intelligent collaboration enhancement module; The basic communication capability module includes a plug-in interface module, a data communication module, and a protocol conversion module, which are used to enable flexible access between multiple simulation platforms, data transmission synchronization, and protocol unification. The intelligent collaboration enhancement module includes a dynamics collaboration module and a scene interpretation module; The dynamics collaboration module models the target vehicle and the environmental vehicle separately using a hierarchical modeling approach; The scenario interpretation module is based on the traffic participant behavior model to enable dynamic interaction and behavioral evolution among multiple subjects.

2. The communication middleware system for joint simulation of intelligent connected vehicles according to claim 1, characterized in that, The simulation platform's simulation software includes: vehicle simulation software, traffic flow simulation software, scene simulation software, and communication simulation software.

3. The communication middleware system for joint simulation of intelligent connected vehicles according to claim 1, characterized in that, The plug-in interface module provides unified data access for different simulation software through standardized interface definitions; The plug-in interface module adopts a publish / subscribe mechanism for data interaction, and accesses data from an external simulation platform by subscribing to a specified data channel; The plugin interface module processes the received data through a callback function mechanism, decoupling the data reception and business processing logic.

4. A communication middleware system for joint simulation of intelligent connected vehicles according to claim 1, characterized in that, The data communication module will encapsulate the traffic flow status data obtained from the simulation platform through the message subscription mechanism into a unified communication message and write it into the message buffer queue. The data communication module will perform asynchronous reception and sequential processing of data through the queue mechanism and the time synchronization mechanism, and obtain the synchronized original data set after completing the time synchronization.

5. A communication middleware system for joint simulation of intelligent connected vehicles according to claim 1, characterized in that, The protocol conversion module parses and processes the received multi-source data and converts it into a unified data structure, specifically including: Data parsing: Decode the received raw communication data and extract traffic participant status information from the communication data packets; Data mapping: According to the preset data mapping rules, the data fields of traffic participant status information extracted from different simulation platforms are mapped to a unified data model; Structure encapsulation: Encapsulate the parsed and mapped data into a unified vehicle object data structure; The data parsing process also includes filtering out abnormal data or handling errors.

6. A communication middleware system for joint simulation of intelligent connected vehicles according to claim 1, characterized in that, The dynamics collaboration module is built based on a hierarchical dynamics modeling mechanism and includes: Automotive simulation software is used to model and calculate key vehicles to obtain high-precision dynamic models of the key vehicles. Traffic flow simulation software was used to model the environmental vehicles, resulting in a simplified dynamic model of the environmental vehicles.

7. A communication middleware system for joint simulation of intelligent connected vehicles according to claim 1, characterized in that, The scenario interpretation module comprises four parts: scenario state perception, behavior decision generation, random event triggering, and scenario dynamic updating, which are used for dynamic interaction and multi-agent behavior evolution in traffic scenarios.

Images