Intelligent network connection vehicle test environment design method

Abstract

The invention discloses an Internet of Things vehicle test environment design method, which integrates a digital twinning technology and a real scene digital technology, establishes a virtual test field completely matched with an intelligent network connection vehicle entity scene, controls the entity scene adjustment through a digital twinning control platform, and generates a real scene consistent with a virtual world. The method comprises the following steps: fully automatically reasoning an intelligent networked vehicle test environment meeting user requirements according to multi-modal requirements of a user by utilizing a generative AI technology, and controlling an entity scene to complete a test according to the test environment through a digital twin control platform, thereby simulating the complexity and uncertainty of a real world in a full dimension, and improving the test efficiency. And various scene elements can be switched and self-defined in real time, and a complex entity environment can be flexibly arranged.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent networked vehicle testing, and in particular to a method for designing an intelligent networked vehicle testing environment. Background Art

[0002] Intelligent Connected Vehicle (ICV) refers to the organic combination of the Internet of Vehicles and smart cars. It is equipped with advanced on-board sensors, controllers, actuators and other devices, and integrates modern communication and network technologies to achieve intelligent information exchange and sharing between vehicles and people, vehicles, roads, and the cloud. It has functions such as complex environment perception, intelligent decision-making, and collaborative control. It can achieve "safe, efficient, comfortable, and energy-saving" driving, and will ultimately realize a new generation of cars that can replace human operation.

[0003] With the continuous advancement of technology and the changing needs of consumers, smart connected vehicle technology is showing a rapid development trend, which will bring people a more convenient, safe and environmentally friendly travel experience.

[0004] The intelligent connected vehicle test field is a site facility designed for the testing and verification of intelligent connected vehicles. It has closed physical boundaries and various roads, network facilities and environmental conditions required for the testing of the autonomous driving functions of intelligent connected vehicles. The main function of the intelligent connected vehicle test field is to provide comprehensive support and services for the research and development, testing, verification and evaluation of autonomous driving vehicles. The test field usually contains a variety of different types of roads, such as highways, rural roads and special roads, as well as various simulation scenarios, such as gas stations, toll stations, tunnels, roundabouts, railways and rural roads. The setting of these facilities and scenarios is designed to simulate the real traffic environment so as to conduct comprehensive and in-depth testing of the autonomous driving functions of intelligent connected vehicles. Intelligent connected vehicle test fields are usually also equipped with cutting-edge technical equipment, such as V2X communication, high-precision maps and positioning, cloud control platforms, etc., which can provide more comprehensive support and services for autonomous driving vehicles. In addition, some test fields are also equipped with simulation test platforms, which can provide simulations of tens of millions of scenarios, with the characteristics of high efficiency, safety, repeatability and scalability, which effectively complement the closed area test, thereby achieving effective verification of the safety of autonomous driving vehicles.

[0005] However, the current test scenarios of intelligent connected vehicle test sites are not diverse enough to meet the diverse testing needs of intelligent connected vehicles; nor can the test site facilities and layout be adjusted at any time to adapt to different testing needs. It is difficult to provide personalized services and cannot meet more diverse testing needs.

[0006] At the same time, as a combination of software and hardware, intelligent connected vehicles also involve human life safety. They need to be tested in all directions and from multiple angles to evaluate the safety performance of intelligent connected vehicles, such as situational response capabilities, control accuracy, control stability, and fault response. Designing the test environment for intelligent connected vehicles has become very important. At present, the test plans for intelligent connected vehicles are mostly written by designers based on experience and then discussed. Then, the test is carried out according to the test plan, and the test data is collected for evaluation. However, this test plan design method is too subjective, highly dependent on designers, lacks systematicity, and cannot fully consider various factors, scenarios, and risks. It is easy to cause some important links to be omitted, thus affecting the comprehensiveness and completeness of the test.

[0007] In order to avoid the above problems, the present invention provides a scientific and systematic method for designing a test environment for intelligent connected vehicles. Summary of the invention

[0008] The purpose of the present invention is to provide a method for designing a test environment for an intelligent connected vehicle, which is used to solve the problems that the test scenario of the intelligent connected vehicle is single, cannot meet personalized services, and the test plan is highly dependent on designers, lacks systematicity, and cannot fully consider various factors.

[0009] To achieve the above object, the present invention adopts the following technical solutions: A method for designing a test environment for an intelligent connected vehicle according to the present invention comprises the following steps: S1, build a digitalized intelligent connected vehicle test site, including various road types, weather conditions, obstacles, geographical features, and traffic signals; S2, building a digital twin platform matching the intelligent connected vehicle test field based on the intelligent connected vehicle test field, and connecting it to the Metaverse platform to form a virtual test platform; S3, generates dynamic test scenarios based on the generative AI model according to user test requirements; S4, projecting the dynamic test scene onto the virtual test platform; and controlling the intelligent connected vehicle test field to match the dynamic test scene in real time through the digital twin platform; S5, real-time acquisition of data of intelligent connected vehicles in the test field and feedback to the digital twin platform; S6, analyzes and understands the collected data based on artificial intelligence algorithms, infers the rationality, safety, and environmental adaptability of the intelligent connected vehicle's behavior, and generates test reports and performance evaluation indicators.

[0010] Furthermore, the intelligent connected vehicle test field includes intelligent target objects and non-intelligent target objects; the non-intelligent target objects realize information collection and scene control through sensor interfaces, identification and recognition interfaces, and mechanical or electrical interfaces; the intelligent target objects realize information collection and scene control through API and protocol interfaces, network communication interfaces, and embedded controller interfaces.

[0011] Furthermore, the road of the intelligent connected vehicle test field is divided into several sections, and the road parameters in each section can be independently and dynamically adjusted in real time; the road parameters include road surface material, slope, curvature, and lane width.

[0012] Furthermore, the digital twin platform also includes high-precision roads, buildings, and transportation facilities constructed through high-definition map data, drone aerial photography, and ground laser scanning.

[0013] Furthermore, the generative AI model constructs a knowledge graph by collecting natural driving data, driving scenario data, regulatory data, and accident scenario data, and uses deep learning technology to establish a traffic behavior model, and generates dynamic test scenarios based on reasoning according to user test needs.

[0014] Furthermore, the dynamic test scenario includes the timing and actions of the interaction between the road surface type, road parameters, weather conditions, obstacles, geographical features, traffic signals and the tested vehicle in the intelligent connected vehicle test field.

[0015] Furthermore, step S4 also includes the digital twin platform and the intelligent connected vehicle test field interacting with each other through virtual and real situations, so that the tested vehicle travels on real roads and in digital twin environments, thereby realizing a combination of virtual and real-scene verification.

[0016] Furthermore, the dynamic test scenario is fed back to the user through the Metaverse platform and is subject to user-defined adjustments.

[0017] Furthermore, the data acquired in step S5 includes scene data of the test site and data of the vehicle being tested in the test site.

[0018] Furthermore, step S6 specifically includes performing content reasoning on the collected test site scene data, judging future development trends in combination with historical data and current situations; and evaluating the rationality, safety, and adaptability of the tested vehicle's behavior to the environment based on the trends and the tested vehicle data.

[0019] The advantage of the present invention is that it integrates digital twin technology and real-scene digitization technology to establish a virtual test site that fully matches the physical scene of intelligent networked vehicles, and controls the physical scene adjustment through the digital twin control platform to generate a real scene consistent with the virtual world. Generative AI technology is used to fully automatically infer the intelligent networked vehicle test environment that meets the user's needs based on the user's multimodal needs, and the physical scene is controlled by the digital twin control platform to complete the test according to the test environment, thereby simulating the complexity and uncertainty of the real world in all dimensions, and can switch and customize various scene elements in real time, and flexibly arrange complex physical environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The intelligent connected vehicle test environment design method of the present invention specifically includes the following steps: S1, build a digital intelligent connected vehicle test site, including but not limited to various road types, weather conditions, obstacles, geographical features, and traffic signals.

[0023] Road surface types include asphalt roads, gravel roads, slippery roads, icy roads, etc. For example, by spraying / flushing related materials to change the road friction coefficient, asphalt roads, gravel roads, slippery roads or icy roads can be simulated.

[0024] Weather conditions include rain, snow, sandstorms, different wind speed levels, fog, changes in brightness between day and night, temperature differences, etc. For example: Changes in wind force levels can be simulated in small venues such as wind tunnel laboratories by adjusting the wind tunnel structure and operating parameters.

[0025] The generation and elimination of rain, fog and snow can be artificially created through a spraying system. Measures can also be taken to reduce and eliminate them, such as spraying water mist to increase humidity and help the precipitation of particles.

[0026] Light and temperature simulation can simulate the changes in light intensity and temperature in different seasons and time periods through the lighting system and intelligent temperature control system.

[0027] Large parks or urban landscapes can change the light and temperature of local areas to a certain extent through technologies such as smart lighting systems and solar reflective materials.

[0028] Geographical features include common geographical landforms such as villages, towns, mountains, open areas, and various building heights, densities, and styles, which can be controlled in the intelligent connected vehicle test field through retractable, upright / lying, and movable landscape objects.

[0029] Obstacles mainly refer to natural and human factors, including tree distribution, bridges, tunnels, pedestrians, animals and other possible sudden obstacles or interference sources. In the intelligent connected vehicle test field, they can also be controlled through retractable, movable, upright / lying down, and movable landscape objects.

[0030] Traffic signals include traffic lights, road signs, signboards, isolation belts, etc., which can be controlled by remotely controlling their lifting, standing / lying down, and movement in the intelligent connected vehicle test field.

[0031] All target objects in the intelligent networked vehicle test field can be digitally reconstructed and controlled. This application divides the target objects in the intelligent networked vehicle test field into intelligent target objects and non-intelligent target objects.

[0032] Smart target objects realize information collection and scene control through API and protocol interfaces, network communication interfaces, and embedded controller interfaces. For example, the network communication capabilities built into smart target objects can exchange data with IoT systems through wireless communication technologies such as 5G, Wi-Fi, Bluetooth, ZigBee, NB-IoT, LoRa, or wired communication technologies such as Ethernet, RS-485, and CAN bus. Or build smart target objects that support standard or customized application programming interfaces (APIs), which follow IoT protocols such as MQTT, CoAP, and HTTP, allowing the system to remotely read its status data, execute instructions, or update firmware. It is also possible to use the interface provided by the microprocessor or embedded controller inside the smart target object to achieve the purpose of external control systems controlling or monitoring its internal logic through programming instructions.

[0033] Non-intelligent target objects realize information collection and scene control through sensor interfaces, identification and recognition interfaces, and mechanical or electrical interfaces. For example, various sensors (such as temperature and humidity sensors, pressure sensors, photoelectric sensors, infrared sensors, sound sensors, vibration sensors, etc.) are deployed on non-intelligent target objects to capture the physical state information of non-intelligent target objects, such as environmental parameters, object position, motion state, etc. Or use radio frequency identification (RFID) technology, barcodes or QR codes and other automatic identification and data capture technologies to deploy identification and recognition interfaces on non-intelligent target objects to read and track the identity information and related attributes of non-intelligent target objects. Or directly detect and change the state of non-intelligent target objects through physical contact or electrical coupling such as switch control and start and stop control of mechanical transmission devices.

[0034] The intelligent connected vehicle test field is divided into several independent sections according to specific lengths or functions. Road parameters in each section, such as surface material, slope, curvature, lane width, etc., can be adjusted independently and dynamically in real time. The specific adjustment methods are: First, separate various factors from various scenes in the test site, such as weather factors, road conditions, obstacles, geographical features, etc. Label the state parameters that may appear under each factor. For example, weather factors include sunny days, rainy days, snowy days, foggy days, etc.; road conditions include dry land, wet land, icy and snowy roads, gravel roads, etc.; obstacles can be divided into static obstacles, mobile obstacles, pedestrians, animals, etc.; geographical features are divided into cities, villages, tunnels, bridges, etc. Give each state parameter that may appear under each factor a unique identifier. By selecting and combining different labels, complex and changeable test scenarios can be quickly generated. For example, the scene of a certain section can be set to snow, tunnels, pedestrians appearing randomly, etc., thereby providing a comprehensive test scenario for intelligent connected vehicles.

[0035] At the same time, each section is combined to form a more complex traffic environment scene. For example, the connection section between urban traffic and highways, the combination of rural roads and bridges, etc. The traffic environment scene composed of these mixed sections can more realistically reflect the actual road traffic conditions and provide a more comprehensive and flexible scene for the testing of intelligent connected vehicles. By dividing the test road into several sections and flexibly adjusting and labeling the adjustable state environment of the scene in each section, different test requirements can be met more efficiently and the accuracy and reliability of the test can be improved. This will help promote the research and development and application of intelligent connected vehicles and improve the safety and efficiency of road traffic.

[0036] S2, based on the intelligent connected vehicle test field, a digital twin platform matching the intelligent connected vehicle test field is constructed, and connected to the Metaverse platform to form a virtual test platform.

[0037] The digital twin platform first uses high-definition map data, drone aerial photography, ground laser scanning and other methods to digitize real-world elements such as roads, buildings, and transportation facilities with high precision, constructing a three-dimensional model that is highly consistent with the real world.

[0038] Secondly, the digital twin platform is deeply integrated with the constructed intelligent connected vehicle test field to realize the comprehensive dynamic simulation of the intelligent connected vehicle test field, and further connected to the Metaverse platform to create a real-time interactive, highly realistic virtual test platform. The digital twin platform can be used to control the digital target objects of the intelligent connected vehicle test field, realize customized and flexible arrangement of the test field scene elements, and realize complex physical test scenarios.

[0039] S3, generates dynamic test scenarios based on the generative AI model according to user test requirements.

[0040] The present invention uses a generative AI model of real scenes. Users can output their multiple test requirements in this AI model. Such as "I want to test driving on a highway on a rainy night", "Simulate a busy urban traffic scene", etc. Through the user's description, the generative AI model will automatically infer a rich variety of complex traffic scenarios based on user predefined rules, historical data, real-world statistical characteristics, etc., including but not limited to road surface types, road parameters, weather conditions, obstacles, geographical features, traffic signals and the timing and actions of the interaction between the tested vehicles, traffic flow, road layout, pedestrian behavior, emergencies and other dynamic test scenarios in the intelligent connected vehicle test field, ensuring wide coverage and high authenticity of the test scenarios.

[0041] Generative AI models are driven by diverse data sets, including massive amounts of traffic data such as natural driving data, driving scenario data, regulatory data, and accident scenario data. These data can provide reference and basis for the generation of test scenarios for autonomous vehicles. These data are combined to create a knowledge graph for training and verifying generative AI models. Based on these data, a model that can accurately reflect real road traffic behaviors and rules is generated to generate various complex traffic scenarios for test vehicles to deal with.

[0042] Generative AI models can construct pseudo-random roads, climates and road conditions to provide a more comprehensive environment for the testing of autonomous vehicles. Obstacles, distractions and personnel can be set up pseudo-randomly in the test site to simulate emergencies in real traffic, so as to test the response and handling capabilities of autonomous vehicles to these situations. The physical scenes, road surfaces, gates, rain, snow, sand, wind, fog, cold and heat, colors, light and dark, rural towns, and building heights can be flexibly changed to provide a diverse environment for the testing of intelligent connected vehicles. Various elements in road traffic can also be taken into account, and elements such as people, trees, bridges and tunnels can be introduced into the test site to enrich the test environment.

[0043] S4, projecting the dynamic test scene onto the virtual test platform; and controlling the intelligent connected vehicle test field to match the dynamic test scene in real time through the digital twin platform.

[0044] The dynamic test scene constructed by the generative AI model is projected onto the virtual test platform to adjust various parameters in the digital twin environment in real time, such as dynamically changing the behavior patterns of traffic participants, simulating road construction, introducing temporary traffic control, etc. The digital twin platform controls the intelligent connected vehicle test field to make real-time adjustments so that the virtual dynamic test scene matches the physical scene, and the matching results are fed back to the user. After the user confirms the actual matching effect or further adjusts and confirms it, the vehicle test can begin.

[0045] The digital twin platform can access the data of the vehicle under test, perceive the vehicle's status and environmental variables in real time, and inject these data into the digital twin platform to synchronize the physical scene with the twin scene, enhancing the reality and effectiveness of the test. At the same time, through the virtual interactive interface, the digital twin platform can also switch or add virtual scenes in some test phases, so that the vehicle under test can travel on the real road and in the digital twin environment, realizing the combination of virtual and real scene verification.

[0046] The anti-warfare of 5G communication technology and the realization of real-time and efficient V2X (Vehicle to Everything) communication can enable intelligent connected vehicles to seamlessly switch or hybrid operation in the virtual or physical environment of the test site, and ensure high-speed transmission and synchronous updating of test data, greatly improving the ability of intelligent connected vehicles to test in different test scenarios, fully examining the instant response of intelligent connected vehicles in different scenarios, obtaining more effective test data, and promoting the development of intelligent connected vehicles and unmanned driving technology.

[0047] The method of this application can also dynamically adjust the difficulty and complexity of the digital twin scenario according to test requirements and system performance to achieve adaptive test sequence design.

[0048] S5, real-time acquisition of data of intelligent connected vehicles in the test field, and feedback to the digital twin platform. The data acquired by the digital twin platform includes the scene data of the test field and the data of the vehicle under test in the test field. The scene data includes information such as weather, roads, obstacles, traffic signals, and geographical features in the test field collected by various sensors in the test field. Vehicle data includes the path, speed, obstacle avoidance, response time, braking, vehicle system parameters, etc. of the vehicle under test.

[0049] S6, analyzes and understands the collected data based on artificial intelligence algorithms, infers the rationality, safety, and environmental adaptability of the intelligent connected vehicle's behavior, and generates test reports and performance evaluation indicators.

[0050] Specifically include: By real-time processing of multimodal sensor data such as images, videos, radars, and lidars collected during the test, various elements in the road environment, such as lane lines, traffic signs, pedestrians, other vehicles, obstacles, etc., can be accurately identified, and their position, posture, motion status, and potential interaction relationships can be understood.

[0051] By comparing the driving decisions (such as path planning, speed control, obstacle avoidance strategies, etc.) output by the autonomous driving system with the actual road conditions, traffic rules and driving ethics, the system can infer whether the decisions made by the vehicle under test are reasonable and safe. For example, it can assess whether the vehicle brakes or turns in a timely and appropriate manner in the face of emergencies, or whether it follows the right-of-way rules in complex traffic scenarios.

[0052] Predict potential traffic risks and safety hazards based on historical data, current situations, and possible future development trends. For example, predict whether the vehicle ahead may suddenly change lanes or brake, whether pedestrians may enter the lane, or whether road conditions may make it difficult for the vehicle to operate. Based on these predictions, it is possible to evaluate whether the autonomous driving system has taken appropriate preventive measures and provide improvement suggestions accordingly.

[0053] By monitoring and analyzing the real-time response time, decision stability, and adaptability of the autonomous driving system to environmental changes, for example, assessing how quickly the vehicle adjusts its driving plan after receiving new information, or maintaining consistency in safe driving strategies in continuously changing traffic scenarios.

[0054] Conduct compliance checks on the behavior of autonomous vehicles in accordance with relevant laws and regulations, industry standards, and internal corporate regulations, including but not limited to complying with traffic rules, respecting the rights of pedestrians and other road users, and properly handling emergencies.

[0055] Based on the collected data and artificial intelligence algorithms, the decision-making process of the autonomous driving system in complex traffic environments and the accuracy, adaptability and safety of vehicle execution control can be comprehensively and deeply evaluated and verified. The core of this function is to simulate or reproduce various real-world driving scenarios, and use AI technology to verify the decision logic and control effect of the autonomous driving system when facing different road conditions, traffic conditions and emergencies. Specifically include: By simulating complex and ever-changing traffic scenarios, such as intersections, roundabouts, congested roads, construction areas, and bad weather, we observe how the autonomous driving system makes real-time decisions based on sensor input data. The evaluation includes but is not limited to path selection, speed planning, obstacle avoidance strategies, and right of way priority judgment, to ensure that the decision-making process complies with traffic rules, driving ethics, and safety standards.

[0056] Reaction speed, decision-making flexibility, and risk avoidance effectiveness when facing unexpected situations (such as the sudden appearance of pedestrians, animals, obstacles, irregular driving of other vehicles, etc.). Evaluate whether the autonomous driving system can quickly identify abnormal conditions, accurately determine the risk level, and take effective measures to avoid or mitigate potential accidents.

[0057] Simulate interaction scenarios with other road users (such as pedestrians, bicycles, motorcycles, other autonomous vehicles and manned vehicles) to check the rationality and courtesy of autonomous vehicles in situations such as approaching, overtaking, merging, and yielding. Evaluate whether the system can accurately predict the behavioral intentions of other road users and whether it can make appropriate responses to ensure road traffic safety and smoothness.

[0058] Through simulated driving tasks, the autonomous driving system is tested for its understanding of the vehicle dynamics model and the control accuracy of the vehicle actuators (such as steering, throttle, and brakes). It is also evaluated whether the vehicle can accurately drive along the planned path under different speeds, accelerations, turning radii, and other conditions, as well as its stability control capabilities when it encounters factors such as uneven roads and crosswinds.

[0059] Simulate extreme situations such as sensor failure, communication interruption, hardware failure, etc. to test the effectiveness of the fault detection, isolation, recovery mechanism and redundant design of the autonomous driving system. Evaluate whether the system can maintain safe driving capabilities when some functions are lost, or whether it can smoothly transition to a safe mode (such as downgrading to assisted driving or emergency parking).

[0060] In addition, artificial intelligence algorithms can also continuously optimize the decision-making algorithms and control strategies of the autonomous driving system based on test results, and self-learn and self-adjust on a large scale for continuous improvement.

Claims

1. A method for designing a test environment for an intelligent connected vehicle, characterized in that: The following steps are involved: S1, build a digitalized intelligent connected vehicle test site, including various road types, weather conditions, obstacles, geographical features, and traffic signals; S2, building a digital twin platform matching the intelligent connected vehicle test field based on the intelligent connected vehicle test field, and connecting it to the Metaverse platform to form a virtual test platform; S3, generates dynamic test scenarios based on the generative AI model according to user test requirements; S4, projecting the dynamic test scene onto the virtual test platform; and controlling the intelligent connected vehicle test field to match the dynamic test scene in real time through the digital twin platform; S5, real-time acquisition of data of intelligent connected vehicles in the test field and feedback to the digital twin platform; S6, analyzes and understands the collected data based on artificial intelligence algorithms, infers the rationality, safety, and environmental adaptability of the intelligent connected vehicle's behavior, and generates test reports and performance evaluation indicators.

2. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: The intelligent connected vehicle test field includes intelligent target objects and non-intelligent target objects; the non-intelligent target objects realize information collection and scene control through sensor interfaces, identification and recognition interfaces, and mechanical or electrical interfaces; the intelligent target objects realize information collection and scene control through API and protocol interfaces, network communication interfaces, and embedded controller interfaces.

3. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: The road of the intelligent connected vehicle test field is divided into several sections, and the road parameters in each section can be adjusted independently and dynamically in real time; the road parameters include road surface material, slope, curvature, and lane width.

4. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: The digital twin platform also includes high-precision roads, buildings, and transportation facilities constructed through high-definition map data, drone aerial photography, and ground laser scanning.

5. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: The generative AI model constructs a knowledge graph by collecting natural driving data, driving scenario data, regulatory data, and accident scenario data, and uses deep learning technology to establish a traffic behavior model, and generates dynamic test scenarios based on reasoning based on user test needs.

6. The method for designing a test environment for an intelligent connected vehicle according to claim 5, characterized in that: The dynamic test scenario includes the timing and actions of the interaction between the road surface type, road parameters, weather conditions, obstacles, geographical features, traffic signals and the tested vehicle in the intelligent connected vehicle test field.

7. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: Step S4 also includes the digital twin platform and the intelligent connected vehicle test field interacting with each other through virtual and real situations, so that the tested vehicle travels on real roads and in digital twin environments, thereby realizing a combination of virtual and real-scene verification.

8. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: The dynamic test scenario is fed back to the user through the Metaverse platform and is subject to user customized adjustments.

9. The method for designing a test environment for an intelligent connected vehicle according to claim 1, characterized in that: The data acquired in step S5 includes the scene data of the test site and the data of the vehicle being tested in the test site.

10. The method for designing a test environment for an intelligent connected vehicle according to claim 9, characterized in that: Step S6 specifically includes performing content reasoning on the collected test site scene data, combining historical data and current situations to determine future development trends; and evaluating the rationality, safety, and environmental adaptability of the tested vehicle's behavior based on the trends and the tested vehicle data.

Images