Enterprise-oriented automatic driving vehicle cloud intelligent unit

Abstract

The invention relates to the technical field of intelligent traffic management and planning, provides an enterprise-oriented automatic driving vehicle cloud intelligent unit, and aims to construct an automatic driving cooperative operation system in which a vehicle end and a cloud end are coupled in real time for an enterprise. More specifically, the enterprise-oriented automatic driving vehicle cloud intelligent unit provides one or more of the automatic driving functions of vehicle cloud linkage at different spatio-temporal levels through vehicle cloud bidirectional integration and cooperative distribution of multi-source data and computing resources, such as perception, prediction, planning and decision, control, service, management, operation and maintenance. Therefore, automatic driving travel service, automatic driving remote service and automatic driving application value-added service are supported.

Description

A cloud-based intelligent unit for enterprise-oriented autonomous vehicles Technical Field

[0001] This application relates to the field of intelligent traffic management and planning technology, and in particular to an enterprise-oriented autonomous vehicle cloud intelligent unit. Background Technology

[0002] With the rapid evolution of autonomous driving technology, enterprises are increasingly demanding autonomous vehicles for logistics, commuting, and operational scheduling. The high-speed movement and complexity of road environments necessitate a real-time closed loop between vehicle-side perception, prediction, and control units and cloud capabilities to provide stable and reliable services under varying environmental and business requirements. However, most current autonomous driving systems start with single-vehicle intelligence, lacking a systematic coupling and collaboration mechanism with the cloud. This makes it difficult to meet the efficient vehicle-cloud integration needs of enterprises operating across multiple scenarios, multiple vehicles, and large scale. Existing technologies typically focus on vehicle-side functionalities, such as onboard perception, single-vehicle decision-making, and local control. While these can support a certain degree of autonomous driving, they fall short in long-tail scenarios such as cross-regional, large-scale traffic changes, emergencies, severe weather, and construction activities, where the lack of synergy between individual vehicle capabilities and cloud empowerment remains insufficient. Furthermore, existing systems underutilize multi-source data from cloud platforms, traffic management centers, and roadside infrastructure. The lack of a unified data fusion and spatiotemporal collaborative processing mechanism between vehicles and the cloud prevents the provision of continuous and stable vehicle-cloud coupled services for autonomous vehicles. In enterprise-level applications, the large number of autonomous vehicles and the complexity of their tasks necessitate a vehicle-cloud intelligent unit platform capable of multi-temporal and spatial-level vehicle-cloud collaboration, on-demand resource allocation, and remote management and operation maintenance. However, currently, there is no complete solution encompassing vehicle-cloud collaboration, from supply-demand relationship modeling and functional fusion scheduling to event response processing. This makes it impossible to dynamically combine and flexibly invoke perception, prediction, planning, and control functions on both sides of the vehicle-cloud, nor can it provide reliable vehicle-cloud closed-loop remote monitoring, remote driving, and rescue capabilities. Based on these shortcomings, enterprises urgently need a vehicle-cloud intelligent unit capable of bidirectional integration of multi-source data on both the vehicle and cloud sides, vehicle-cloud collaboration of autonomous driving functions, closed-loop management of remote services, and multi-dimensional resource optimization to enhance the safety, reliability, and operability of autonomous vehicles in complex traffic environments. This invention aims to build an autonomous driving vehicle-cloud intelligent unit for enterprises, providing intelligent services and full lifecycle support for autonomous vehicles through a unified cloud control architecture and multi-subsystem collaborative operation. Summary of the Invention

[0003] The purpose of this application is to provide an enterprise-oriented autonomous vehicle cloud intelligent unit that can support autonomous vehicles to achieve autonomous driving functions and provide autonomous driving users with multi-temporal and spatial autonomous driving services.

[0004] To achieve the above objectives, this application adopts the following technical solution: an enterprise-oriented autonomous vehicle cloud intelligent unit (E-VCIU), wherein the enterprise-oriented autonomous vehicle cloud intelligent unit (E-VCIU) supports autonomous vehicles to realize autonomous driving functions, and provides users with multi-temporal and spatial autonomous driving services through single-source or multi-source input data in different autonomous driving scenarios based on traffic fences. The system includes a service and management subsystem, a communication subsystem, a control subsystem, and one or more of the following subsystems: (1) a perception subsystem; (2) a prediction subsystem; (3) a planning and decision-making subsystem; (4) an integration and allocation subsystem; (5) a support subsystem; the autonomous driving function includes one or more of the following functions: (1) perception function; (2) prediction function; (3) planning and decision-making function; (4) control function; (5) service function; (6) management function; (7) operation function; (8) maintenance function.

[0005] Preferably, in the aforementioned enterprise-oriented autonomous vehicle cloud intelligent unit, it is configured to provide users with collaborative operations that improve autonomous driving functions / services through autonomous vehicles and intelligent connected vehicles; wherein the collaborative operation is configured as a collaborative operation method that enables the enterprise-oriented autonomous vehicle cloud intelligent unit E-VCIU to share functions, resources and information flow with autonomous vehicles and intelligent connected vehicles through a communication subsystem.

[0006] Preferably, in the aforementioned enterprise-oriented autonomous vehicle cloud intelligent unit, a combination scheme of perception, prediction, planning and decision-making, and control functions, resources, and information flow based on the E-VCIU cloud computing and integrated distribution subsystem is generated to complete driving tasks; it is used to provide basic autonomous driving travel services, advanced autonomous driving travel services, autonomous driving application value-added services, remote services, operation services, and maintenance services to meet users' service needs.

[0007] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the traffic fence is configured to include the demand characteristics and supply characteristics of the autonomous driving system, and is used to classify autonomous driving scenarios: (1) demand characteristics, which are autonomous driving requests generated by autonomous driving service users; (2) supply characteristics, which are the functions and resources provided by each subsystem of E-VCIU, including one or more of perception, computing, storage and communication.

[0008] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the single-source or multi-source input data includes event data and / or non-event data obtained from one or more sources from the cloud platform and traffic control center: (1) Event data, including one or more of construction area data, weather data, traffic control equipment data, event data, special activity data, activity data and hazard data; (2) Non-event data, including one or more of traffic flow status data, vehicle dynamics data, travel demand data and system computing and storage resource data.

[0009] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the traffic fence and autonomous driving service are configured to be classified according to spatial and temporal dimensions: (1) The temporal dimension includes the following three levels: a. Micro level: between 1 millisecond and 10 milliseconds, including 1 millisecond and 10 milliseconds; b. Meso level: 10 milliseconds to 1000 milliseconds; c. Macro level: greater than 1000 milliseconds; (2) The spatial dimension includes the following three levels: a. Micro level: between 0.1 cm and 1 cm, including 0.1 cm and 1 cm; b. Meso level: 1 cm to 100 cm; c. Macro level: greater than 100 cm.

[0010] Preferably, in the aforementioned enterprise-oriented autonomous vehicle cloud intelligent unit, the E-VCIU receives user-sent perception, prediction, planning and decision-making, and control function requirements, and provides autonomous driving function combination solutions that meet the requirements by analyzing the characteristics of the requirements and the characteristics of the supply; the E-VCIU provides long-tail scenario solutions for autonomous vehicles through centralized and / or distributed computing methods.

[0011] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the communication subsystem is configured to establish communication channels and transmit data between various subsystems, components and users within the E-VCIU through multiple wired and / or wireless communication modes.

[0012] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the perception subsystem is configured to use cloud platform sensors and / or traffic control center sensors to acquire single-source or multi-source input data through the communication subsystem, providing positioning, identification and multi-dimensional perception functions.

[0013] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the prediction subsystem is configured to provide autonomous vehicles with prediction functions at different spatiotemporal levels by utilizing perception information and single-source and / or multi-source input data.

[0014] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligence unit, the planning and decision-making subsystem is configured to generate planning and decision-making information for autonomous vehicles using perception information, prediction information, and single-source and / or multi-source input data.

[0015] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the control subsystem is configured to receive perception, prediction, planning and decision-making information, and generate refined and timely vehicle-specific control instructions for autonomous vehicles based on single-source and / or multi-source input data; the refined and timely vehicle-specific control instructions include one or more of the following control instructions: (1) longitudinal and lateral position; (2) vehicle speed; (3) vehicle acceleration; (4) vehicle steering angle; (5) vehicle spacing; (6) lane designation; (7) vehicle following and lane changing instructions; (8) vehicle path guidance.

[0016] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the integration and allocation subsystem includes an integration module and an allocation module: (1) The integration module is used to integrate and optimize cloud platform resources including one or more of perception resources, computing resources and storage resources; the integration module is configured to improve the overall system's intelligence level by realizing the collaborative operation between different subsystems and components; (2) The allocation module is used to centrally deploy the functions, resources and information flow of each subsystem and / or component of E-VCIU in terms of perception, prediction, planning, decision and control according to user needs; the allocation module is configured to realize collaborative perception, collaborative prediction, collaborative planning and decision and collaborative control in different autonomous driving scenarios.

[0017] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the service and management subsystem includes one or more of the following modules: (1) a cloud computing module, used to provide cloud-based data analysis and computing functions at different spatiotemporal scales for each subsystem and / or component of the E-VCIU, wherein the cloud computing module includes one or more of central cloud, fog cloud and edge cloud; (2) an information service module, used to provide users with one or more of the following services: storage as a service (STaaS), control as a service (CCaaS), computing as a service (CaaS), information as a service (IaaS), perception as a service (SEaaS), operation and maintenance as a service (OMaaS), and artificial intelligence as a service (AIaaS); (3) a basic autonomous driving travel service module, used to provide basic autonomous driving travel services and remote services to autonomous vehicles at different spatiotemporal scales according to the personalized travel needs of autonomous vehicles, traffic flow status and traffic management regulations; (4) an advanced autonomous driving travel service module, used to provide advanced autonomous driving travel services and remote services to autonomous vehicles at different spatiotemporal scales according to travel needs, traffic flow status and traffic management regulations, so as to meet the personalized needs of autonomous vehicles to the greatest extent; (5) The autonomous driving application value-added service module is used to provide data analysis services for autonomous driving applications and provide autonomous driving travel data to autonomous driving application suppliers to provide further services based on different types of autonomous driving data; (6) Vehicle management module is used to provide management of autonomous vehicles at different spatiotemporal levels; (7) Event response module is used to respond to traffic events and provide autonomous driving solutions based on event analysis to achieve control optimization of autonomous vehicles at different spatiotemporal levels; (8) Operation module is used to perform one or more of the daily operation and online software updates of E-VCIU; the daily operation of E-VCIU includes monitoring and recording the working status of one or more subsystems and / or components of E-VCIU and the cloud platform; (9) Maintenance module is used to perform daily maintenance of the system, the daily maintenance of the system includes one or more of the diagnosis and troubleshooting of system emergencies and remote maintenance of E-VCIU subsystems and / or components; wherein the remote services include one or more of remote monitoring services, remote driving services, and remote rescue services: (1) Remote monitoring services are used to achieve real-time monitoring and information fusion of autonomous driving operation status through one or more of cloud platforms and / or traffic control centers, and to provide fused data for remote services; the real-time operation monitoring includes one or more of the following: a. vehicle status monitoring; b. sensor data monitoring; c. environmental perception monitoring; d. planning and decision-making monitoring; e. communication and connectivity monitoring; f.(1) Fault diagnosis and early warning; (2) Remote driving service, used to meet the remote driving needs of autonomous vehicles in different scenarios; the collected data is processed and analyzed through the cloud computing module, and detailed and timely vehicle-specific control instructions are generated for autonomous vehicles at the micro level; the vehicle-specific control instructions are sent to the autonomous vehicles through the communication subsystem to ensure their real-time execution; the remote driving service can be provided in a request-response mode or a forced provision mode according to the urgency of the scenario; the scenario includes emergency scenarios and non-emergency scenarios: a. Emergency scenarios, including severe weather conditions, sudden obstacles, unexpected traffic conditions, and external emergency events when the autonomous vehicle cannot be safely operated by human intervention; b. Non-emergency scenarios, including situations where the driver can operate safely under the premise of ensuring safety, or situations where the autonomous vehicle itself can ensure safe operation.

[0018] (3) Remote rescue service, used to allocate and schedule rescue resources when autonomous vehicles malfunction and / or encounter unexpected events requiring human intervention, thereby improving the safety and stability of autonomous vehicles; the allocation and scheduling of rescue resources includes one or more of the following: a. software fault repair; b. allocation of rescue personnel; c. provision of rescue equipment.

[0019] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the Artificial Intelligence as a Service (AIaaS) is configured to: (1) provide one or more of the following: autonomous driving database, AI model training and download, and AI large model framework, to provide AI computing services based on cloud computing modules; (2) provide end-to-end full-stack models and / or hybrid sequence models, including one or more of the following: data fusion, perception, prediction, planning, decision-making, control, integration and allocation models based on autonomous driving database data information, and send the trained customized vehicle control commands to the vehicle to complete the autonomous driving task; (3) provide a combined autonomous driving solution for each subsystem and / or component of E-VCIU to meet the needs of autonomous driving services; wherein the AI ​​large model framework is configured to include: (1) an AI platform that provides computing resources for training end-to-end full-stack models and / or hybrid sequence models to generate model parameters; (2) an AI modeling framework for: a. providing AI models, data integration, modeling platforms, simulation and / or optimization and evaluation; b. providing AI models for autonomous driving functions and services of cloud platforms and / or traffic control centers; c. Provides the model parameters required for deploying end-to-end full-stack models and / or hybrid sequence models.

[0020] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the basic autonomous driving mobility service module and the advanced autonomous driving mobility service module adopt the following operation method: (1) The autonomous vehicle sends a request to the E-VCIU, which is received by the communication subsystem and forwarded to the service and management subsystem; (2) The service and management subsystem analyzes the vehicle's needs and determines whether the request meets the requirements of the advanced service; (3) The basic autonomous driving mobility service module or the advanced autonomous driving mobility service module generates a service strategy; (4) The communication subsystem sends the service strategy to the control subsystem; (5) The control subsystem generates vehicle control commands and sends them to the autonomous vehicle.

[0021] Preferably, in the above-mentioned autonomous vehicle cloud intelligent unit for enterprises, the autonomous driving application value-added service module adopts the following operation method: (1) The autonomous driving application supplier sends a request to the E-VCIU, the communication subsystem receives the request and sends it to the service and management subsystem; (2) The autonomous driving application value-added service module provides corresponding services based on the generated service strategy through the basic autonomous driving travel service module and / or the advanced autonomous driving travel service module.

[0022] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the event response module is supported by a technical solution characterized by an event model. The event model is used to realize trajectory prediction, behavior decision-making and control of autonomous vehicles at the micro level. The event model is configured to: (1) access one or more event data through the communication subsystem, use the event data as input, convert macro event data into micro event data, and optimize the management and control of autonomous vehicles; (2) customize the management and control scheme for long-tail scenarios of autonomous vehicles to improve the user's ability to solve long-tail scenario problems; The event model adopts the following operating method: (1) The event model receives the perception information of the cloud platform and the historical and / or real-time monitoring data of the traffic management center and meteorological department through the communication subsystem, and extracts the event data as input; (2) The event model analyzes the functional and demand changes brought about by the event data to perception, prediction, planning decision-making and control; (3) The event model obtains the functional, resource and status information of users, E-VCIU subsystem and / or components, and generates collaborative technology solutions based on cloud computing, including one or more of collaborative perception, collaborative prediction, collaborative planning and decision-making and collaborative control; (4) The event model uses a cloud computing module and an AI model to generate vehicle trajectories.

[0023] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the operation module and maintenance module are configured to provide full life cycle operation of E-VCIU and full life cycle maintenance of E-VCIU users and software and / or hardware devices, wherein: (1) the hardware devices of E-VCIU include storage devices, computing devices, communication devices and / or support devices; (2) the full life cycle of autonomous driving includes the design stage, manufacturing and testing stage, operation stage and upgrade stage.

[0024] Preferably, in the above-mentioned enterprise-oriented autonomous vehicle cloud intelligent unit, the supporting subsystem is configured to provide information interface, information security, power supply, multi-engine and storage support functions to the subsystems and / or components of the E-VCIU; the supporting subsystem includes one or more of the following modules: (1) an interface module, used to interface with the E-VCIU and its internal and / or external information, the internal information including information interfaced through human-machine interface (HMI) and / or adapter plug, the external information including one or more of weather, construction, maintenance, accident and rescue information; (2) an information security module, used to provide privacy protection for one or more types of private information, and ensure the security of the E-VCIU's subsystems and / or components and their communication; (3) a multi-engine module, used to deploy multiple engines for mirroring, to realize information synchronization, backup, processing and improve computing power, to provide authorization for one or more users and subsystems and / or components of the E-VCIU, to realize specific levels of autonomous driving functions and improve system reliability and resilience; (4) a storage module, used for: a. a. Store user and / or E-VCIU configuration file information; b. Store and / or retrieve real-time sensor data, real-time forecast data, real-time planning and decision data, real-time control data, operation and maintenance data, service management log data, and / or operation log data of each E-VCIU subsystem and / or component; c. Store and manage one or more of traffic crashes, network congestion patterns, weather events, and / or road construction.

[0025] The Enterprise-Oriented Autonomous Vehicle Cloud Intelligent Unit (E-VCIU) provided in this application has at least the following beneficial effects: The enterprise-oriented E-VCIU of this application enables autonomous vehicles and / or intelligent connected vehicles to coordinate subsystems such as perception, prediction, planning and decision-making, control, and service management through the cloud. This achieves multi-source data fusion, on-demand combination of autonomous driving functions, and collaborative processing across spatiotemporal scales, thereby significantly improving the operational safety and service reliability of autonomous vehicles in complex scenarios. The system can simultaneously access event data and traffic operation data, generating optimal functional solutions for different autonomous driving scenarios through a unified supply and demand analysis and resource allocation mechanism. It also supports collaborative perception, collaborative prediction, and collaborative control, enhancing the vehicle's ability to respond to long-tail scenarios, emergencies, and environmental changes. Simultaneously, the E-VCIU provides enterprise-level services such as remote monitoring, remote driving, and remote rescue, enabling real-time management and emergency response of vehicle operating status, forming a closed loop from risk identification and function scheduling to control execution. The system supports AIaaS model training and deployment capabilities, generating customized control strategies for enterprise vehicles, improving the system's scalability and applicability. The E-VCIU of this application can be widely used in various autonomous driving scenarios such as enterprise logistics, passenger transport, and engineering vehicles, providing enterprises with safe, efficient, manageable and controllable autonomous driving service capabilities. Attached Figure Description

[0026] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0027] Figure 1 is a schematic diagram of system components for an enterprise-oriented autonomous vehicle cloud intelligent unit; Figure 2 is a schematic diagram of data sources for an enterprise-oriented autonomous vehicle cloud intelligent unit; Figure 3 is a schematic diagram of users for an enterprise-oriented autonomous vehicle cloud intelligent unit; Figure 4 is a schematic diagram of hardware devices for an enterprise-oriented autonomous vehicle cloud intelligent unit; Figure 5 is a schematic diagram of two examples of cloud platform layers for an enterprise-oriented autonomous vehicle cloud intelligent unit, where (a) is a schematic diagram of cloud platform layers; (b) is a schematic diagram of cloud platform layers; Figure 6 is an application example diagram of an enterprise-oriented autonomous vehicle cloud intelligent unit; Figure 7 is a schematic diagram of the composition structure of an AI large model framework; Figure 8 is a flowchart of the perception subsystem for an enterprise-oriented autonomous vehicle cloud intelligent unit; Figure 9 is a flowchart of an enterprise-oriented autonomous vehicle cloud intelligent unit. Figure 10 shows the flowchart of the prediction subsystem of an autonomous vehicle cloud intelligent unit for enterprises; Figure 11 shows the flowchart of the control subsystem of an autonomous vehicle cloud intelligent unit for enterprises; Figure 12 shows the flowchart of the integration module of an autonomous vehicle cloud intelligent unit for enterprises; Figure 13 shows the flowchart of the basic autonomous driving mobility service module and the advanced autonomous driving mobility service module of an autonomous vehicle cloud intelligent unit for enterprises; Figure 14 shows the flowchart of the autonomous driving application value-added service module of an autonomous vehicle cloud intelligent unit for enterprises; Figure 15 shows the event model workflow of an autonomous vehicle cloud intelligent unit for enterprises; Figure 16 shows the flowchart of the operation method of the advanced version of the autonomous vehicle cloud intelligent unit for enterprises.

[0028] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0029] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0030] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0031] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0032] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0033] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0034] First, the reference numerals in the accompanying drawings are explained as follows: 101: An enterprise-oriented autonomous vehicle cloud intelligent unit; 102: Service and management subsystem; 103: Communication subsystem; 104: Perception subsystem; 105: Prediction subsystem; 106: Planning and decision-making subsystem; 107: Control subsystem; 108: Integration and distribution subsystem; 109: Support subsystem; 201: An enterprise-oriented autonomous vehicle cloud intelligent unit; 202: Data source; 203: Event data; 204: Non-event data; 205: Data communication transmission; 301: An enterprise-oriented autonomous vehicle cloud intelligent unit; 302: Providing autonomous driving services; 303: Autonomous driving service user; 401: An enterprise-oriented autonomous vehicle cloud intelligent unit; 402: Hardware device; 403: Storage device; 404: Computing device; 405: Communication device; 406: Support device; 501: An enterprise-oriented autonomous vehicle cloud intelligent unit; 502: Central cloud; 503: Fog cloud; 504: Edge cloud; 505: Data communication transmission; 506: Data communication transmission; 507: Data communication transmission; 508: Data communication transmission; 509: Data communication transmission; 510: Data communication transmission; 601: Application of an enterprise-oriented autonomous vehicle cloud intelligent unit; 602: Service application; 603: Management application; 701: AI large model framework; 702: AI platform; 703: AI modeling framework; 704: Providing computing power; 705: Providing models; 706: AI computing; 707: AI Model; 1501: Event Model Workflow; 1502: Data Source; 1503: Event Data; 1504: Functional and Requirement Changes; 1505: Collaborative Technical Solutions; 1506: Vehicle Trajectory; 1507: Data Input; 1508: Calculation and Analysis; 1509: Calculation and Analysis; 1510: Result Output; Figure 1 illustrates the system components of an enterprise-oriented autonomous vehicle cloud intelligent unit 101. An enterprise-oriented autonomous vehicle cloud intelligent unit 101 includes a service and management subsystem 102 and a communication subsystem 103, as well as one or more of the following subsystems: perception subsystem 104, prediction subsystem 105, planning and decision-making subsystem 106, control subsystem 107, integration and distribution subsystem 108, and support subsystem 109.

[0035] Figure 2 illustrates the data sources of an enterprise-oriented autonomous vehicle cloud intelligent unit 201. The data source 202 of the enterprise-oriented autonomous vehicle cloud intelligent unit 201 includes event data 203 and non-event data 204. Event data 203 includes one or more of the following: construction zone data, meteorological data, traffic control equipment data, event data, activity data, and hazard data. Non-event data 204 includes one or more of the following: traffic flow status data, vehicle dynamics data, travel demand data, and system computing and storage resource data. Data from data source 202 can be transmitted to the enterprise-oriented autonomous vehicle cloud intelligent unit 201 via data communication 205.

[0036] Figure 3 illustrates a user of an enterprise-oriented autonomous vehicle cloud intelligent unit 301. This enterprise-oriented autonomous vehicle cloud intelligent unit 301 provides autonomous driving services 302 to users 303. Specifically, users 303 include vehicles, roads, travelers, automotive companies, suppliers, and technology companies.

[0037] Figure 4 illustrates a hardware device 402 of an enterprise-oriented autonomous vehicle cloud intelligent unit 401. The hardware device 402 of the enterprise-oriented autonomous vehicle cloud intelligent unit 401 includes a storage device 403, a computing device 404, a communication device 405, and a support device 406. Specifically, the storage device 403 includes local storage devices and cloud storage devices; the computing device 404 includes one or more of a central processing unit (CPU), a graphics processing unit (GPU), random access memory (RAM), and read-only memory (ROM); the communication device 405 includes wireless communication devices and wired communication devices; and the support device 406 includes power supply facilities.

[0038] Figure 5(a) illustrates the cloud platform hierarchy of an enterprise-oriented autonomous vehicle cloud intelligent unit 501. The cloud platform architecture of the enterprise-oriented autonomous vehicle cloud intelligent unit 501 consists of a central cloud 502, a fog cloud 503, and an edge cloud 504. Data is transmitted from the central cloud 502 to the fog cloud 503 via communication 505, from the fog cloud 503 to the central cloud 502 via communication 506, from the fog cloud 503 to the edge cloud 504 via communication 507, and from the edge cloud 504 to the fog cloud 503 via communication 508.

[0039] Figure 5(b) illustrates an exemplary cloud platform hierarchy for an enterprise-oriented autonomous vehicle cloud intelligence unit 501. In some embodiments, data can be transmitted directly from the central cloud 502 to the edge cloud 504 via communication 509. In some embodiments, data can also be transmitted directly from the edge cloud 504 to the central cloud 502 via communication 510.

[0040] Figure 6 illustrates a schematic diagram of an enterprise-oriented autonomous vehicle cloud intelligent unit application 601. This application 601 includes a service application 602 and a management application 603. Specifically, the service application 602 includes basic autonomous driving mobility services, advanced autonomous driving mobility services, autonomous driving application value-added services, remote services, operation services, and / or maintenance services. Specifically, the management service 603 includes vehicle management and incident response.

[0041] Figure 7 illustrates the structural composition of the AI ​​large model framework 701. The AI ​​large model framework 701 includes an AI platform 702 and / or an AI modeling framework 703. The AI ​​platform 702 provides computing power support 704 for AI computing 706, and the AI ​​modeling framework 703 provides model support 705 for the AI ​​model 707.

[0042] Figure 8 illustrates a perception method for the perception subsystem of an enterprise-oriented autonomous vehicle cloud intelligent unit. The perception subsystem acquires data from sensing devices and systems, processing positioning and multi-dimensional perception information. If the computing power and speed of the perception subsystem do not meet minimum requirements, cloud computing and AI large-scale model frameworks will provide computing power support. The E-VCIU perception subsystem transmits sensing information to other E-VCIU subsystems. Specifically, these other E-VCIU subsystems include one or more of the following: service and management subsystem, communication subsystem, prediction subsystem, planning and decision-making subsystem, control subsystem, integration and distribution subsystem, and support subsystem. In an exemplary implementation, the perception method of the perception subsystem of the autonomous vehicle cloud intelligent unit for enterprises begins by having the E-VCIU perception subsystem acquire data from perception devices and systems and determine whether the computing power and speed meet the minimum requirements. If the determination result is negative, the cloud computing and AI large model framework provides computing power support to the E-VCIU perception subsystem. If the determination result is positive, the E-VCIU perception subsystem generates positioning and multi-dimensional perception information, and then sends the positioning and multi-dimensional perception information to other subsystems and / or components, ending the process.

[0043] Figure 9 illustrates a prediction method for a prediction subsystem of an enterprise-oriented autonomous vehicle cloud intelligent unit. The prediction subsystem acquires perception information and generates multi-dimensional prediction information. If the computing power and speed of the prediction subsystem do not meet the minimum requirements, cloud computing and AI large model frameworks will provide computing power support. The E-VCIU prediction subsystem sends prediction information to other subsystems. In an exemplary embodiment, the prediction method begins with the E-VCIU prediction subsystem receiving perception information; it then determines whether the computing power and speed of the E-VCIU prediction subsystem meet the minimum requirements. If the result of this determination is negative, the cloud computing and AI model framework provides computing power support to the E-VCIU prediction subsystem. After obtaining the corresponding computing power support, the E-VCIU prediction subsystem generates multi-level prediction information. If the result of the above determination is positive, the E-VCIU prediction subsystem directly generates multi-level prediction information and sends the prediction information to other subsystems and / or components, ending the process.

[0044] Figure 10 illustrates a planning and decision-making method for the planning and decision-making subsystem of an enterprise-oriented autonomous vehicle cloud intelligent unit. The planning and decision-making subsystem acquires perception and prediction information, generates behavioral decisions and path planning instructions, and calculates the vehicle's position and motion parameters for the next time period. The E-VCIU planning and decision-making subsystem sends planning and decision-making information to other subsystems. If the computing power and speed of the planning and decision-making subsystem do not meet minimum requirements, cloud computing and AI large-scale model frameworks will provide computing power support. If safety and efficiency thresholds (such as speed, spacing, etc.) are not met, the planning and decision-making results will be adjusted. In one exemplary implementation, the planning and decision-making method begins by having the E-VCIU planning and decision-making subsystem receive perception and prediction information. It then determines whether the E-VCIU planning and decision-making subsystem's computing power and speed meet minimum requirements. If the determination is negative, cloud computing and AI large-scale model frameworks provide computing power support to the E-VCIU planning and decision-making subsystem. After meeting the minimum requirements for computing power and speed, the E-VCIU planning and decision-making subsystem generates behavioral decision and path planning instructions, and calculates the vehicle's position and dynamic state at the next moment. Next, it determines whether driving safety and efficiency meet requirements. If the determination is negative, the planning and decision-making are adjusted. After adjustment, the E-VCIU planning and decision-making subsystem calculates the vehicle's position and dynamic state at the next moment again and re-executes the above determination. If the determination is positive, the E-VCIU planning and decision-making subsystem sends the planning and decision-making information to other subsystems and / or components, and the process ends.

[0045] Figure 11 illustrates a control method for a cloud-based intelligent unit control subsystem of an enterprise-oriented autonomous vehicle. The control subsystem receives perception, prediction, planning, and decision-making information, generates detailed, real-time vehicle control commands, and calculates the vehicle's position and motion parameters for the next time period. If the control subsystem's computing power and speed do not meet minimum requirements, cloud computing and AI large-scale model frameworks provide computing support. If safety and efficiency thresholds (such as speed and spacing) meet requirements, control commands are sent to the vehicle via the communication subsystem. If the control commands are not acknowledged, new commands are sent to the vehicle. In one exemplary embodiment, the control method begins by having the E-VCIU control subsystem receive perception, prediction, planning, and decision-making information. The subsystem then determines whether its computing power and speed meet minimum requirements. If not, cloud computing and an AI large-scale model framework provide computing power support to the E-VCIU control subsystem. Once the minimum requirements for computing power and speed are met, the E-VCIU control subsystem generates detailed and timely vehicle-specific control commands. The E-VCIU control subsystem then compiles the next... The system first measures the vehicle's position and dynamic state at a given moment. Then, it assesses whether the driving safety and efficiency requirements are met. If the assessment result is negative, the control command is adjusted. After adjustment, the E-VCIU control subsystem calculates the vehicle's position and dynamic state at the next moment and re-executes the above assessment. If the assessment result is positive, a detailed and timely vehicle-specific control command is sent to the vehicle. Then, it assesses whether the vehicle confirms the control command. If the assessment result is negative, a new control command is sent to the vehicle. If the assessment result is positive, the control method process ends.

[0046] Figure 12 illustrates the operation of the integration module and the allocation module. The integration module receives resource information from vehicles, roads, the cloud, and / or the central hub through the communication subsystem and interface module. This resource information includes the functions, data, and computing power of the subsystems and / or components. The integration module performs data fusion based on the received different types of resource information and stores it through the storage module. The allocation module reallocates the functions, data, and computing power resources of different subsystems and / or components based on their needs.

[0047] Figure 13 illustrates the operation methods of the basic autonomous driving mobility service module and the advanced autonomous driving mobility service module. The vehicle sends an autonomous driving mobility service request to the service and management subsystem. The service and management subsystem determines whether the request belongs to an advanced autonomous driving service request or a basic autonomous driving service request. If it is an advanced autonomous driving service request, the advanced mobility service module generates an advanced service policy; if it is a basic autonomous driving service request, the basic mobility service module generates a basic service policy and sends it to the vehicle. In an exemplary embodiment, the operation method begins with the vehicle sending an autonomous driving mobility service request to the service and management subsystem; subsequently, the service and management subsystem analyzes the vehicle request and determines whether it meets the requirements of advanced autonomous driving services; if the determination result is yes, the advanced autonomous driving mobility service module generates an advanced service policy; if the determination result is no, the basic autonomous driving mobility service module generates a basic service policy; then, the generated service policy is sent back to the vehicle, and the process ends.

[0048] Figure 14 illustrates the operation method of the autonomous driving application value-added service module. Enterprise users send application value-added service requests to the autonomous driving application value-added service module. The module analyzes the service request and determines whether data analysis is required. If data analysis is required, the analysis results are sent to the user; if no data analysis is required, the data is sent directly to the user based on the request. In an exemplary implementation, the operation method of this autonomous driving application value-added service module begins with the enterprise user sending an autonomous driving application value-added service request to the business management subsystem. The module then analyzes the service request and determines whether further data analysis is needed. If the determination is yes, data analysis based on autonomous driving data is performed, and the analysis results are fed back to the enterprise user, ending the process. If the determination is no, the autonomous driving data is sent directly back to the enterprise user, ending the process.

[0049] Figure 15 illustrates the workflow 1501 of the event model. The event model receives perception information from data sources 1502, including historical and / or real-time monitoring data from cloud platforms, traffic management centers, and meteorological departments, through the communication subsystem, and extracts event data as input 1507. Event data 1503 includes construction zone data, weather data, traffic control data, event data, and hazardous event data. The impact of this event data on perception, prediction, planning and decision-making, and control functions and requirements is analyzed 1504 using data-driven machine learning methods 1508. The event model acquires functional, resource, and status information of users and E-VCIU subsystems and / or components, and generates collaborative technical solutions 1505 based on cloud computing, including collaborative perception, collaborative prediction, collaborative planning and decision-making, and collaborative control technologies 1509. Based on the collaborative technical solutions, the event model generates vehicle trajectories 1506 through the cloud computing module and AIaaS 1510.

[0050] Figure 16 illustrates the operation method of the advanced version of the enterprise-oriented autonomous vehicle cloud intelligent unit. First, the vehicle sends a request to the advanced version of the E-VCIU, which determines whether the service request is satisfied. If not, the advanced version of the E-VCIU operation method is executed. This method determines whether the service request is a basic service request. If it is a basic service request, a basic mobility service strategy is generated through the basic autonomous driving mobility service module based on the collected data, and the plan is returned to the control subsystem. If not, an advanced mobility service plan is generated through the advanced autonomous driving mobility service module based on the collected data, and the control subsystem generates control commands and sends them to the vehicle. The vehicle is controlled by selecting certain control commands. In one exemplary embodiment, the operation method begins by having the vehicle send an autonomous driving service request to the advanced version of E-VCIU. It then determines whether any service requests are unmet; if the determination is negative, the process ends. If the determination is positive, the advanced version of E-VCIU operation method is executed to determine whether the service request is a basic service request. If the determination is positive, the basic autonomous driving mobility service module generates a basic mobility service strategy based on collected data; if the determination is negative, the advanced autonomous driving mobility service module generates an advanced mobility service strategy based on collected data. After either the basic autonomous driving mobility service module or the advanced autonomous driving mobility service module generates the basic mobility service strategy based on collected data, the mobility service strategy is sent to the control subsystem. The advanced version of E-VCIU control subsystem then generates control commands and sends them to the vehicle, ending the process.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A cloud-based intelligent unit for enterprise-oriented autonomous vehicles, characterized in that: The enterprise-oriented autonomous vehicle cloud intelligent unit E-VCIU supports autonomous vehicles to achieve autonomous driving functions and provides users with multi-temporal and spatial autonomous driving services in different autonomous driving scenarios based on traffic fences, through single-source or multi-source input data. The system includes a service and management subsystem, a communication subsystem, a control subsystem, and one or more of the following subsystems: (1) perception subsystem; (2) prediction subsystem; (3) planning and decision-making subsystem; (4) integration and allocation subsystem; (5) support subsystem; The autonomous driving function includes one or more of the following functions: (1) perception function; (2) prediction function; (3) planning and decision-making function; (4) control function; (5) service function; (6) management function; (7) operation function; (8) maintenance function.

2. The enterprise-oriented autonomous vehicle cloud intelligent unit according to claim 1, characterized in that: It is configured to provide users with improved autonomous driving functions / services through autonomous vehicles and intelligent connected vehicles; wherein the collaborative operation is configured as a collaborative operation method that enables the enterprise-oriented autonomous vehicle cloud intelligent unit E-VCIU to share functions, resources and information flow with autonomous vehicles and intelligent connected vehicles through a communication subsystem.

3. The cooperative operation method according to claim 2, characterized in that: Generate a combination of functions, resources, and information flows for perception, prediction, planning and decision-making, and control based on the E-VCIU cloud computing and integrated distribution subsystem to complete driving tasks; It is used to provide basic autonomous driving mobility services, advanced autonomous driving mobility services, value-added services for autonomous driving applications, remote services, operation services and maintenance services to meet users' service needs.

4. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The traffic fence is configured to include the demand and supply characteristics of the autonomous driving system and is used to classify autonomous driving scenarios: (1) demand characteristics, which are autonomous driving requests generated by autonomous driving service users; (2) supply characteristics, which are the functions and resources provided by the various subsystems of E-VCIU, including one or more of perception, computing, storage and communication.

5. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The single-source or multi-source input data includes event data and / or non-event data obtained from one or more sources, including the cloud platform and the traffic control center: (1) Event data, including one or more of construction area data, weather data, traffic control equipment data, event data, special activity data, activity data and hazard data; (2) Non-event data, including one or more of traffic flow status data, vehicle dynamics data, travel demand data and system computing and storage resource data.

6. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The traffic fence and autonomous driving service are configured to be classified by spatial and temporal dimensions: (1) The temporal dimension includes the following three levels: a. Micro level: between 1 millisecond and 10 milliseconds, including 1 millisecond and 10 milliseconds; b. Mesoscopic level: 10 milliseconds to 1000 milliseconds; c. Macroscopic level: greater than 1000 milliseconds; (2) The spatial dimension includes the following three levels: a. Microscopic level: between 0.1 cm and 1 cm, including 0.1 cm and 1 cm; b. Mesoscopic level: 1 cm to 100 cm; c. Macroscopic level: greater than 100 cm.

7. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The E-VCIU receives user requests for perception, prediction, planning and decision-making, and control functions. By analyzing the characteristics of the requests and the characteristics of the supply, it provides a combination of autonomous driving functions that meet the requirements. The E-VCIU provides long-tail scenario solutions for autonomous vehicles through centralized and / or distributed computing.

8. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The communication subsystem is configured to establish communication channels and transmit data between various subsystems, components, and users within the E-VCIU through multiple wired and / or wireless communication modes.

9. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The perception subsystem is configured to use cloud platform sensors and / or traffic control center sensors to acquire single-source or multi-source input data through the communication subsystem, providing positioning, identification and multi-dimensional perception functions.

10. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The prediction subsystem is configured to provide autonomous vehicles with prediction functions at different spatiotemporal levels by utilizing perception information and single-source and / or multi-source input data.

11. The enterprise-oriented autonomous vehicle cloud intelligent unit according to claim 1, characterized in that: The planning and decision-making subsystem is configured to generate planning and decision-making information for autonomous vehicles using perception information, prediction information, and single-source and / or multi-source input data.

12. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The control subsystem is configured to receive perception, prediction, planning and decision-making information, and generate refined and timely vehicle-specific control commands for autonomous vehicles based on single-source and / or multi-source input data; the refined and timely vehicle-specific control commands include one or more of the following control commands: (1) longitudinal and lateral position; (2) vehicle speed; (3) vehicle acceleration; (4) vehicle steering angle; (5) vehicle spacing; (6) lane designation; (7) vehicle following and lane changing commands; (8) vehicle path guidance.

13. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The integration and allocation subsystem includes an integration module and an allocation module: (1) An integration module, used to integrate and optimize cloud platform resources including one or more of sensing resources, computing resources and storage resources; the integration module is configured to improve the overall system's intelligence level by realizing the collaborative operation between different subsystems and components; (2) An allocation module, used to centrally deploy the functions, resources and information flow of each subsystem and / or component of E-VCIU in terms of perception, prediction, planning, decision and control according to user needs; the allocation module is configured to realize collaborative perception, collaborative prediction, collaborative planning and decision and collaborative control in different autonomous driving scenarios.

14. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 1, characterized in that: The service and management subsystem includes one or more of the following modules: (1) a cloud computing module, used to provide cloud-based data analysis and computing functions at different spatiotemporal scales for each subsystem and / or component of E-VCIU, the cloud computing module including one or more of central cloud, fog cloud and edge cloud; (2) an information service module, used to provide users with one or more of the following services: storage as a service (STaaS), control as a service (CCaaS), computing as a service (CaaS), information as a service (IaaS), perception as a service (SEaaS), operation and maintenance as a service (OMaaS) and artificial intelligence as a service (AIaaS); (3) a basic autonomous driving mobility service module, used to provide basic autonomous driving mobility services and remote services to autonomous vehicles at different spatiotemporal scales according to the personalized travel needs of autonomous vehicles, traffic flow status and traffic management regulations; (4) an advanced autonomous driving mobility service module, used to provide advanced autonomous driving mobility services and remote services to autonomous vehicles at different spatiotemporal scales according to travel needs, traffic flow status and traffic management regulations, so as to meet the personalized needs of autonomous vehicles to the greatest extent; (5) The autonomous driving application value-added service module is used to provide data analysis services for autonomous driving applications and provide autonomous driving travel data to autonomous driving application suppliers to provide further services based on different types of autonomous driving data; (6) the vehicle management module is used to provide management of autonomous vehicles at different spatiotemporal levels; (7) the event response module is used to respond to traffic events and provide autonomous driving solutions based on event analysis to achieve control and optimization of autonomous vehicles at different spatiotemporal levels. (8) An operation module for performing one or more of the daily operation of the E-VCIU and online software updates; the daily operation of the E-VCIU includes monitoring and recording the working status of one or more subsystems and / or components of the E-VCIU and the cloud platform; (9) Maintenance module, used for daily maintenance of the system, the daily maintenance of the system includes one or more of the following: diagnosis and troubleshooting of system emergencies, remote maintenance of E-VCIU subsystems and / or components; wherein the remote services include one or more of the following: remote monitoring service, remote driving service, remote rescue service: (1) Remote monitoring service, used to realize real-time monitoring and monitoring information fusion of autonomous driving operation status through one or more of cloud platform and / or traffic control center, and provide fused data for remote services; the real-time operation monitoring includes one or more of the following: a. vehicle status monitoring; b. sensor data monitoring; c. environmental perception monitoring; d. planning and decision monitoring; e. communication and connectivity monitoring; f. fault diagnosis and early warning; (2) Remote driving service, used to meet the remote driving needs of autonomous vehicles in different scenarios; the collected data is processed and analyzed through cloud computing module, and detailed and timely vehicle-specific control instructions are generated for autonomous vehicles at the micro level; The remote driving service sends vehicle-specific control commands to the autonomous vehicle through the communication subsystem to ensure its real-time execution; the remote driving service can be provided in a request-response mode or a mandatory provision mode according to the urgency of the scenario; the scenario includes emergency scenarios and non-emergency scenarios: a. Emergency scenarios include severe weather conditions, sudden obstacles, unexpected traffic situations, and external emergency events when the autonomous vehicle cannot be safely ensured by manual operation; b. Non-emergency scenarios, including situations where the driver can operate safely under the premise of ensuring safety, or situations where the autonomous vehicle itself can ensure safe operation. (3) Remote rescue service, used to allocate and schedule rescue resources when the autonomous vehicle malfunctions and / or an unexpected event requiring human intervention occurs, thereby improving the safety and stability of the autonomous vehicle; the allocation and scheduling of rescue resources includes one or more of the following: a. Software fault repair; b. Allocation of rescue personnel; c. Provision of rescue equipment.

15. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 14, characterized in that: The AIaaS is configured to: (1) provide one or more of the following: autonomous driving database, AI model training and download, and AI large model framework, to provide AI computing services based on cloud computing modules; (2) provide end-to-end full-stack models and / or hybrid sequence models, including one or more of the following: data fusion, perception, prediction, planning, decision-making, control, integration and allocation models based on autonomous driving database data information, and send the trained customized vehicle control commands to the vehicle to complete the autonomous driving task; (3) provide a combined autonomous driving solution for each subsystem and / or component of E-VCIU to meet the needs of autonomous driving services. The AI ​​large model framework is configured to include: (1) an AI platform that provides computational resources for training end-to-end full-stack models and / or hybrid sequence models to generate model parameters; and (2) an AI modeling framework for: a. providing AI models, data integration, modeling platforms, simulation and / or optimization and evaluation; b. providing AI models for autonomous driving functions and services of cloud platforms and / or traffic control centers; and c. providing model parameters required for deploying end-to-end full-stack models and / or hybrid sequence models.

16. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 14, characterized in that: The basic autonomous driving mobility service module and the advanced autonomous driving mobility service module operate as follows: (1) The autonomous vehicle sends a request to the E-VCIU, which is received by the communication subsystem and forwarded to the service and management subsystem; (2) The service and management subsystem analyzes the vehicle's needs and determines whether the request meets the requirements of the advanced service; (3) The basic autonomous driving mobility service module or the advanced autonomous driving mobility service module generates a service strategy; (4) The communication subsystem sends the service strategy to the control subsystem; (5) The control subsystem generates vehicle control commands and sends them to the autonomous vehicle.

17. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 14, characterized in that: The autonomous driving application value-added service module adopts the following operation method: (1) The autonomous driving application supplier sends a request to the E-VCIU, the communication subsystem receives the request and sends it to the service and management subsystem; (2) The autonomous driving application value-added service module provides corresponding services based on the generated service strategy through the basic autonomous driving travel service module and / or the advanced autonomous driving travel service module.

18. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 14, characterized in that: The event response module is supported by a technical solution characterized by an event model. The event model is used to realize trajectory prediction, behavior decision-making and control of autonomous vehicles at the micro level. The event model is configured to: (1) access one or more event data through the communication subsystem, use the event data as input, convert macro event data into micro event data, and optimize the management and control of autonomous vehicles; (2) customize the management and control scheme for long-tail scenarios of autonomous vehicles to improve the user's ability to solve long-tail scenario problems; The event model adopts the following operation method: (1) The event model receives the perception information of the cloud platform and the historical and / or real-time monitoring data of the traffic management center and meteorological department through the communication subsystem, and extracts the event data as input; (2) The event model analyzes the changes in functions and requirements brought about by event data to perception, prediction, planning decision-making and control; (3) The event model obtains the functional, resource and status information of users, E-VCIU subsystems and / or components, and generates collaborative technology solutions based on cloud computing, including one or more of collaborative perception, collaborative prediction, collaborative planning and decision-making and collaborative control; (4) The event model generates vehicle trajectories with the help of AI models through cloud computing modules.

19. The cloud-based intelligent unit for enterprise-oriented autonomous vehicles according to claim 14, characterized in that: The operation module and maintenance module are configured to provide full lifecycle operation of E-VCIU and full lifecycle maintenance of E-VCIU users and software and / or hardware devices, wherein: (1) the hardware devices of E-VCIU include storage devices, computing devices, communication devices and / or support devices; (2) the full lifecycle of autonomous driving includes the design phase, manufacturing and testing phase, operation phase and upgrade phase.

20. The enterprise-oriented autonomous vehicle cloud intelligent unit according to claim 1, characterized in that: The support subsystem is configured to provide information interfaces, information security, power supply, multi-engine, and storage support functions to the E-VCIU's subsystems and / or components; the support subsystem includes one or more of the following modules: (1) an interface module for interfacing with the E-VCIU and its internal and / or external information, the internal information including information interfacing through a human-machine interface (HMI) and / or adapter plugs, and the external information including one or more of weather, construction, maintenance, accident, and rescue information; (2) an information security module for providing privacy protection for one or more types of private information, ensuring the security of the E-VCIU's subsystems and / or components and their communication; (3) a multi-engine module for deploying multiple engines for mirroring to achieve information synchronization, backup, processing, and improved computing power, providing authorization to one or more users and subsystems and / or components of the E-VCIU to achieve specific levels of autonomous driving functions and improve system reliability and resilience; (4) a storage module for: a. storing user and / or E-VCIU configuration file information; b. Store and / or retrieve real-time sensor data, real-time prediction data, real-time planning and decision data, real-time control data, operation and maintenance data, service management log data and / or operation log data of each subsystem and / or component of E-VCIU; c. Store and manage one or more of the following: traffic crashes, network congestion patterns, weather events, and / or road construction.

Images