System for sharing idle computing power of intelligent automobile and intelligent automobile

Abstract

The invention discloses a system for sharing idle computing power of an intelligent automobile, which comprises a user terminal serving as a computing power user, the intelligent automobile for providing computing power service, a cloud end for scheduling the computing power of the intelligent automobile, a computing power sensing module, a computing power measuring module, a computing power arrangement module and a vehicle-mounted computing power execution module, and the computing power arrangement module schedules the computing power of the intelligent automobile according to the processing results of the computing power sensing module and the computing power quantity module so as to provide computing power service for the user terminal. The embodiment of the invention provides an efficient and flexible intelligent vehicle idle computing power sharing system, so that full utilization of vehicle idle computing resources is realized. The system has wide applicability and expandability, can meet the calculation requirements of different users, brings extra benefits to vehicle owners, and makes contributions to social calculation resource optimization.

Description

Technical Field

[0001] The present invention relates to the field of smart car technology, and in particular to a system and a smart car for realizing sharing of idle computing power of smart cars. Background Art

[0002] The development of cloud computing has enabled computing resources to be provided as services, allowing users to rent the computing resources they need on demand. In this model, cloud service providers (such as Amazon AWS, Microsoft Azure, Alibaba Cloud, etc.) build large data centers to centrally manage and allocate computing, storage, and network resources. For the user side, they do not need to invest a lot of money to build and maintain their own computing infrastructure, but only need to pay for the actual resources used to obtain the corresponding computing services.

[0003] However, with the rapid development of the Internet of Things, big data, and artificial intelligence, the demand for computing power is growing exponentially. The centralized cloud computing model has problems such as high bandwidth pressure, high network latency, and increased data security risks when facing a large number of distributed terminal devices. In addition, the construction and operation costs of data centers are high, the energy consumption is huge, and it has also caused a certain impact on the environment.

[0004] Based on the above reasons, exploring more efficient and flexible computing power leasing and sharing models has become the focus of the industry. To this end, edge computing technology has emerged.

[0005] As an extension of cloud computing, edge computing moves computing and data processing capabilities to the edge of the network, that is, closer to the source of data generation. By deploying computing resources at the edge and scheduling edge computing resources through centralized scheduling algorithms, the bandwidth requirements and latency of data transmission can be effectively reduced, and the response speed and service quality can be improved.

[0006] However, due to factors such as limited resources of edge devices, dynamic network changes, and diversified task requirements, traditional centralized scheduling algorithms are difficult to meet the computing power scheduling needs of more diversified distributed edge computing devices. Summary of the invention

[0007] The purpose of the present invention is to provide a system and a smart car for realizing the sharing of idle computing power of smart cars, so as to realize the sharing of computing power for diversified distributed smart cars and solve the problems existing in the prior art.

[0008] The objective of the present invention is achieved through the following technical solutions:

[0009] A system for sharing idle computing power of smart cars, comprising: a user terminal as a computing power user, a smart car for providing computing power resource services, and a cloud for scheduling the computing power of the smart car; and the system also includes:

[0010] A computing power sensing module, used to monitor whether the computing power of the smart car is available and the available time period information;

[0011] A computing power measurement module; used to obtain the amount of available computing power of the smart car, including a unified measurement of heterogeneous computing power of different smart cars;

[0012] A computing power scheduling module, used to schedule the computing power of the smart car to provide computing power services for the user terminal according to the processing results of the computing power perception module and the computing power measurement module;

[0013] The vehicle-mounted computing power execution module is used to execute the computing power service task when the smart car receives the notification of executing the computing power service task given by the computing power orchestration module.

[0014] Preferably, the computing power perception module includes:

[0015] A corresponding personalized model is established by using personalized vehicle usage data of a smart car in a machine learning manner; the personalized vehicle usage data includes one or more of historical usage data of the smart car, daily usage habits, and data on the impact of external environment data on its usage;

[0016] Determine the idle time period information of the smart car according to the personalized model, and determine the resource computing power information that can be shared and used according to the smart car power and computing device hardware parameters obtained by monitoring;

[0017] The idle time period information and the resource computing power information that can be shared are uploaded and used as basic data of the computing power orchestration module.

[0018] Preferably, the computing power perception module further includes:

[0019] Real-time monitoring of the smart car's operating status, battery power, hardware temperature, and network connection parameter information; and dynamic perception of changes in the smart car's usage status to update the available computing power in real time;

[0020] The idle time period information and the resource computing power information that can be shared are determined based on the real-time monitoring and dynamic perception results, and uploaded as the basic data of the computing power orchestration module.

[0021] Preferably, the computing power measurement module includes:

[0022] quantifying different types of computing resources of the smart car in a standardized manner to perform heterogeneous computing power modeling;

[0023] And, based on the output evaluation of the computing power perception module, the availability of the computing power resources of the smart car and the computing power resource performance level that it can provide are determined.

[0024] Preferably, the process of performing heterogeneous computing power modeling includes using floating-point computing capability, memory bandwidth and computing unit quantity indicators to conduct comprehensive evaluation to achieve heterogeneous computing power modeling.

[0025] Preferably, the computing power orchestration module includes a submodule for decomposing the computing power service tasks of the user terminal, a distribution submodule, and a management submodule, wherein:

[0026] The decomposition submodule is used to decompose the computing power service task that needs to be decomposed into several subtasks;

[0027] The allocation submodule is used to allocate the computing service task or several subtasks of the decomposed computing service task to the smart car providing computing service according to a predetermined rule; the predetermined rule refers to one or more of the available time, utilization rate, energy consumption and battery status of the available computing resources of the smart car;

[0028] The management submodule is used to support dynamic migration or fault-tolerant processing of the computing service task or the subtask when the start / stop state of the smart car changes and / or the network is interrupted.

[0029] Preferably, the system further includes a computing power routing module, and the processing performed by the computing power routing module includes:

[0030] According to the prediction result of the network status, an adaptive routing strategy is used to transmit the task data of providing computing power services. The adaptive routing strategy refers to selecting the optimal transmission path according to the real-time network bandwidth, delay, signal strength and stability, and the task data is also transmitted using encryption and authentication mechanisms;

[0031] It also uses historical network data and environmental information to predict future network conditions and adjusts adaptive routing strategies based on the prediction results; and / or monitors changes in network and smart car status during the execution of computing service tasks, and adjusts adaptive routing strategies based on the monitoring results.

[0032] Preferably, the vehicle-mounted computing power execution module includes a computing power service task computing execution submodule, a resource management submodule and a computing security processing submodule, wherein:

[0033] The computing execution submodule is used to execute the computing tasks included in the computing power service tasks allocated by the cloud;

[0034] The resource management submodule is used to manage the computing resources of the smart car and control the provision of computing resources for the computing service task while giving priority to the basic functions and driving requirements of the vehicle;

[0035] The computing security processing submodule is used to implement end-to-end encryption of data, data access control management, and realize the anti-tampering function of each module during the execution of computing service tasks.

[0036] Preferably, the system also includes a computing power trading module, which is used to establish a computing power service pricing trading mechanism between user terminals, smart car users and the cloud. The trading mechanism includes the user terminal paying a computing power service fee to the cloud platform, the cloud paying a computing power resource usage fee to the smart car user, and the cloud obtaining a corresponding intermediary service fee. The cloud is also responsible for establishing trading models and rules.

[0037] An intelligent car for realizing idle computing power sharing, used as an intelligent car for providing computing power services in the above system, comprising:

[0038] A computing power sensing module, used to monitor whether the computing power of the smart car is available and the available time period information;

[0039] A computing power measurement module; used to obtain the amount of available computing power of the smart car, including a unified measurement of heterogeneous computing power of different smart cars;

[0040] The vehicle-mounted computing power execution module is used to execute the computing power service task when the smart car receives a notification to execute the computing power service task.

[0041] Compared with the existing technology, the present invention realizes the efficient perception, measurement, routing, orchestration and transaction of intelligent vehicle computing resources. Specifically, it proposes a targeted and innovative solution for the particularity of intelligent vehicle scenarios, and provides new ideas and technical support for the field of edge computing and resource sharing. The corresponding technical solution can bring about the following beneficial effects, including but not limited to:

[0042] It can accurately predict vehicle idle time and improve the utilization rate of computing resources by deeply learning the user's car-using behavior. It can also uniformly measure the heterogeneous computing resources of smart cars to solve the problem of computing power evaluation between the same hardware architecture.

[0043] In addition, the corresponding adaptive routing algorithm can be adopted according to the complex wireless network environment to effectively ensure the efficient and secure transmission of task data; and an optimized task scheduling and scheduling model is proposed, so as to realize the dynamic computing resources of smart cars and the continuity of task execution;

[0044] Based on the above system architecture, it is also possible to combine the characteristics of smart cars to implement flexible computing service pricing and incentive mechanisms, thereby promoting the development of a computing power sharing ecosystem.

[0045] In summary, the implementation of the present invention can effectively utilize the computing resources of smart cars, provide nearby edge computing services, and reduce the load and network latency of cloud computing; enable car owners to gain benefits by sharing the idle computing power of vehicles, thereby achieving efficient utilization of resources and maximizing value; and by making full use of existing computing resources, reduce the demand for new data centers, and reduce energy consumption and carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0047] Figure 1 A schematic diagram of a system architecture provided by an embodiment of the present invention;

[0048] Figure 2 A flow chart of vehicle computing power perception provided by an embodiment of the present invention;

[0049] Figure 3 A flowchart of computing power routing and task scheduling provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in combination with the specific content of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, which does not constitute a limitation of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of the present invention.

[0051] First, the terms that may be used in this article are explained as follows:

[0052] The term “and / or” means that either or both of them can be realized at the same time. For example, X and / or Y means both “X” or “Y” and “X and Y”.

[0053] The terms "include", "comprises", "contains", "has" or other descriptions with similar semantics should be interpreted as non-exclusive inclusion. For example, including certain technical feature elements (such as raw materials, components, ingredients, carriers, dosage forms, materials, dimensions, parts, components, mechanisms, devices, steps, procedures, methods, reaction conditions, processing conditions, parameters, algorithms, signals, data, products or products, etc.) should be interpreted as including not only certain technical feature elements explicitly listed, but also other technical feature elements known in the art that are not explicitly listed.

[0054] The term "consisting of..." means excluding any technical feature elements not explicitly listed. If this term is used in a claim, it will make the claim closed, so that it does not contain technical feature elements other than the technical feature elements explicitly listed, except for the conventional impurities related to them. If this term only appears in a clause of a claim, it only limits the elements explicitly listed in the clause, and the elements recorded in other clauses are not excluded from the overall claim.

[0055] Unless otherwise specified or limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this article can be understood according to specific circumstances.

[0056] When concentration, temperature, pressure, size or other parameters are expressed in the form of a numerical range, the numerical range should be understood to specifically disclose all ranges formed by the pairing of any upper limit, lower limit, and preferred value in the numerical range, regardless of whether the range is explicitly stated; for example, if a numerical range of "2 to 8" is stated, the numerical range should be interpreted as including ranges such as "2 to 7", "2 to 6", "5 to 7", "3 to 4 and 6 to 7", "3 to 5 and 7", "2 and 5 to 7", etc. Unless otherwise specified, the numerical ranges stated herein include both their end values ​​and all integers and fractions within the numerical range.

[0057] The orientation or position relationship indicated by terms such as "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", etc. are based on the orientation or position relationship shown in the drawings and are only for the convenience and simplification of description, and do not explicitly or implicitly indicate that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation of this document.

[0058] In the specific implementation process of the present invention, it was found through investigation that in the edge computing environment, computing resources present distributed and heterogeneous characteristics, which may include routers, gateways, smart terminal devices, and even vehicles and drones with computing capabilities. In the edge computing environment, the computing scheduling of edge computing resources is one of the core technologies of edge computing, which involves how to efficiently allocate and manage computing tasks on distributed edge nodes. Specifically, due to factors such as limited resources, dynamic changes in the network, and diversified task requirements of edge devices with computing capabilities such as smart vehicles, the corresponding computing scheduling needs to comprehensively consider factors such as resource availability, task priority, network conditions, energy consumption, and formulate optimization strategies.

[0059] Furthermore, in the field of smart cars, vehicles have certain underutilized computing power during driving. If an effective computing scheduling mechanism is used, the idle computing power of smart cars can be shared to serve the computing needs of the surrounding areas and maximize the utilization of computing resources. Based on this, a system for sharing the idle computing power of smart cars is proposed in the present invention.

[0060] That is, the purpose of the present invention is to provide a technical framework and system design for sharing the idle computing power of smart cars, aiming to solve the current problem of sharply increasing computing power demand and insufficient computing power, and then efficiently utilize the computing resources of smart cars during idle time to provide services for users with computing needs. Specifically, by sharing the computing power of smart cars during idle time, distributed computing power resource scheduling and management can be realized, the utilization efficiency of computing resources can be improved, and computing costs can be reduced.

[0061] The overall system architecture adopted by the specific implementation scheme provided by the present invention can be but is not limited to referring to Figure 1 As shown, the system for realizing the sharing of idle computing power of smart cars provided by the present invention may include, but is not limited to, three parts: a vehicle-side terminal node, a cloud management platform, and a user terminal, wherein:

[0062] The vehicle-end node is a smart car used to provide computing power services. It can include computing power perception and computing power measurement modules, which are deployed inside the smart car and are responsible for monitoring the vehicle status and computing power resources, accepting tasks and executing them.

[0063] A cloud management platform, i.e., a cloud for scheduling the computing power of the smart car, which may include computing power routing, computing power orchestration and transaction platforms, responsible for global resource management, task scheduling and transaction processing;

[0064] User terminals, i.e. computing power users, can submit computing tasks (computing power service tasks) through client applications, view the execution status and results of computing power service tasks, and conduct corresponding transaction settlements.

[0065] Specifically, the above technical framework may include, but is not limited to, four main processing modules, namely: a computing power perception module, a computing power measurement module, a computing power orchestration module, and an on-board computing power execution module. Optionally, a computing power routing and computing power transaction module may also be included.

[0066] It should be noted that the above-mentioned computing power perception module and computing power measurement module can be set in the vehicle end node (i.e., smart car), or can be set in an independent device (such as a cloud device or other independently set device that can communicate with the cloud, etc.); the corresponding computing power orchestration module can be, but is not limited to, set in a cloud device or other independently set device that can communicate with the cloud; the vehicle-mounted computing power execution module can usually be set in a smart car as a vehicle end node; the computing power routing module can be set in a smart car, the cloud and / or other independently set devices that can communicate with the cloud; the corresponding computing power transaction module can be, but is not limited to, set in a cloud device or other independently set devices that can communicate with the cloud; that is, Figure 1 The specific arrangement and design structure of the system given in the figure is only a schematic diagram for example, and does not constitute the only limitation on the implementation mode of the present invention.

[0067] The following will refer to Figure 1 As shown in the figure, the specific implementation functions of each module are described in detail:

[0068] (1) Computing power perception module

[0069] That is, the smart car computing power perception module is used to monitor whether the computing power of the smart car is available and the available time period information; its core is to perceive the available computing power resources of the smart car in real time and accurately, especially for the user's personalized car use behavior, and predict the idle time of the vehicle; specifically, it includes:

[0070] Establish a personalized prediction model. Specifically, a machine learning method can be used to use the personalized vehicle usage data of smart cars to establish a corresponding personalized model; that is, use machine learning and data mining technology to establish a personalized model of user vehicle usage behavior. The personalized vehicle usage data includes one or more of the historical usage data of the smart car, daily usage habits, and the impact of external environmental data on its usage. That is, by collecting the historical usage data of the vehicle, the user's daily travel habits, schedules, and environmental factors (such as weather, traffic conditions), etc., predict when the vehicle is idle;

[0071] Real-time status monitoring function, used to monitor the operating status, battery power, hardware temperature and network connection parameter information of smart cars in real time; specifically, it can integrate the vehicle's internal sensors and diagnostic systems to monitor the vehicle's operating status, battery power, hardware temperature and network connection in real time to ensure that smart cars can provide shareable computing resources while meeting their own operating and safety requirements;

[0072] A dynamic perception mechanism is a mechanism that dynamically perceives changes in the usage status of smart cars in order to update the available computing power in real time. Specifically, since the vehicle may be activated at any time, a corresponding dynamic perception mechanism needs to be designed so that the computing power availability information can be updated in time when the vehicle status changes, so as to promptly notify the subsequent related computing power orchestration modules to perform the corresponding computing power resource scheduling and orchestration processing.

[0073] That is to say, based on the above-mentioned personalized model, and the corresponding real-time monitoring and dynamic perception results, the computing power resources that the smart car can provide can be determined and uploaded as the basic data of the computing power orchestration module.

[0074] Specifically, Figure 2 As shown, the corresponding computing power perception processing process may include:

[0075] (11) Collecting vehicle (i.e., smart car) idle time and battery timing information, and using a timing prediction model to predict high-probability idle time information, where the idle time period is the time period that may provide shared computing resources;

[0076] And, the computing power that can be used is estimated based on the power and hardware, that is, the resource computing power that can be shared;

[0077] The above-mentioned process of collecting vehicle idle time and battery timing information and shared resource computing power can be obtained through real-time monitoring or regular monitoring;

[0078] (12) Upload the corresponding idle time information and available computing power information as the basic data information for cloud computing power orchestration.

[0079] (2) Computing power measurement module

[0080] That is, the intelligent vehicle heterogeneous computing power measurement module is used to obtain the available computing power of the intelligent vehicle, including the unified measurement of the heterogeneous computing power of different intelligent vehicles; it is responsible for evaluating and quantifying the heterogeneous computing power resources of intelligent vehicles to provide a basis for subsequent task scheduling and resource allocation. Its core technical content is as follows:

[0081] Heterogeneous computing power modeling is to quantify different types of computing resources of the smart car in a standardized way to perform heterogeneous computing power modeling. Smart cars are usually equipped with a variety of computing resources, such as CPU, GPU, ASIC and FPGA, etc. In order to facilitate the measurement of corresponding computing resources, it is necessary to establish a unified computing power measurement standard so as to quantify different types of computing resources in a standardized way. For example, indicators such as floating point computing power (FLOPS), memory bandwidth and number of computing units can be used for comprehensive evaluation to achieve heterogeneous computing power modeling.

[0082] Resource availability evaluation and performance prediction, that is, determining the availability of the computing power resources of the smart car and the level of computing power resource performance it can provide based on the output evaluation of the computing power perception module; that is, taking into account the battery capacity and energy consumption of the vehicle, it is necessary to dynamically evaluate the availability of computing power resources to ensure that the normal operation and safety performance of the vehicle are not affected during the sharing of computing resources. At the same time, for different types of computing tasks, estimate the performance level that the vehicle's computing power resources can provide; for example, for AI reasoning tasks, evaluate the actual reasoning speed of the GPU; for data processing tasks, evaluate the processing power of the CPU, and so on.

[0083] (3) Computing power orchestration module

[0084] That is, the computing power scheduling module is used to schedule the computing power of the smart car to provide computing power services for the user terminal according to the processing results of the computing power perception module and the computing power measurement module, and realize the computing power scheduling of multiple computing power service tasks for the user terminal. It mainly includes the decomposition, allocation and management of tasks to ensure the efficient execution of tasks on the vehicle computing power resources. Specifically, the computing power scheduling module may include a submodule for decomposing the computing power service tasks of the user terminal, a submodule for allocating the submodule and a submodule for managing the computing power service tasks of the user terminal, wherein:

[0085] The decomposition submodule is used to decompose the computing power service task that needs to be decomposed into several subtasks; for example, a large computing power service task that cannot meet the computing power resource requirements of a certain vehicle computing power resource can be decomposed into subtasks that can be executed on a single vehicle computing power resource;

[0086] The allocation submodule is used to allocate the computing service task or several subtasks of the decomposed computing service task to the smart car providing computing service according to a predetermined rule, for example, mapping the task to a suitable vehicle according to the computing power characteristics and available time of the vehicle; wherein the predetermined rule refers to one or more of the available time, utilization rate, energy consumption and battery status of the available computing resources of the smart car; specifically, the allocation submodule may adopt the following task scheduling strategy and resource utilization optimization and restriction strategy during the allocation process, namely:

[0087] The task scheduling strategy can design a scheduling algorithm that takes into account task switching costs and deployment costs, so as to prioritize short-term, lightweight tasks for short-term available computing resources; and can schedule complex tasks for long-term available resources;

[0088] The resource utilization optimization and restriction strategy takes into account the utilization rate and energy consumption of vehicle computing resources, optimizes the execution order of tasks and resource allocation, maximizes resource utilization efficiency, and minimizes energy consumption; at the same time, the computing power sharing of smart cars needs to consider the impact of battery power and energy consumption on vehicle endurance; the above task scheduling strategy will comprehensively evaluate the battery status of the vehicle to avoid adverse effects on the normal use of the car by the owner;

[0089] The management submodule is used to support the dynamic migration or fault-tolerant processing of the computing service task or the subtask when the start / stop state of the smart car changes and / or the network is interrupted; that is, to realize task migration and fault tolerance, that is, when the vehicle state changes (such as the vehicle starts driving) or the network is interrupted, it supports dynamic migration and fault-tolerant processing of tasks to ensure the continuity and integrity of the tasks.

[0090] (4) Computing Execution Module

[0091] That is, the vehicle-mounted computing power execution module is used to execute computing power service tasks on the smart car that provides computing power services, that is, when the smart car receives the notification of executing computing power service tasks given by the computing power orchestration module, it executes the computing power service tasks; it may include a computing execution submodule, a resource management submodule and a computing security processing submodule of the computing power service tasks, that is, the vehicle-mounted computing power execution module needs to actually execute computing tasks on the allocated idle vehicles, including computing execution, resource management and computing security, wherein:

[0092] The computing execution submodule is used to execute the computing tasks included in the computing service tasks assigned by the cloud; that is, the vehicle-mounted computing execution module is responsible for receiving computing tasks from the cloud management platform. After receiving the corresponding computing tasks, it is necessary to use the idle time and computing resources of the vehicle to execute the assigned computing tasks; accordingly, the vehicle-mounted computing execution module needs to have the functions of task initiation, execution and termination, as well as monitoring of the execution progress;

[0093] The resource management submodule is used to manage the computing resources of the smart car, and control the provision of computing resources for the computing service task while giving priority to the basic functions and driving requirements of the vehicle; that is, the resource management submodule needs to monitor and manage the computing resources of the vehicle in real time (such as CPU, GPU usage and memory occupancy, etc.), and dynamically adjust the allocation of computing resources according to task requirements; in addition, it is also necessary to consider energy-saving management of resources, especially to optimize resource usage when the battery is low, so as to ensure the basic functions and driving requirements of the vehicle;

[0094] The computing security processing submodule is used to implement end-to-end encryption of data during the execution of computing service tasks, data access control management, and the tamper-proof function of each module; that is, the computing security processing submodule ensures data security and model security, prevents data leakage and various processing functions implemented by each module (such as established models, preset billing rules, etc.) from being tampered with; implements end-to-end encryption, data access control and other security measures. Computational integrity verification requires software and hardware to verify whether the execution results of the computing tasks have not been tampered with, ensuring the correctness and reliability of the results.

[0095] In the above system, optionally, the following two processing modules may also be included:

[0096] (5) Computing power routing module

[0097] That is, the user-intelligent vehicle computing power routing module is designed to achieve efficient transmission of task data and results in a complex network environment; specifically, the computing power routing module can be used to transmit task data that provides computing power services using an adaptive routing strategy based on the prediction results of the network status. The adaptive routing strategy refers to selecting the optimal transmission path based on the real-time network bandwidth, delay, signal strength and stability, and the task data is also transmitted using encryption and authentication mechanisms; its core technical content may include:

[0098] Adaptive routing strategy: smart cars are connected to the cloud through wireless networks (such as 5G and V2X). Network conditions may change at any time, so it is necessary to design corresponding adaptive routing strategies to select the optimal transmission path based on real-time network bandwidth, latency, signal strength, and stability.

[0099] Network status prediction: To facilitate the execution of adaptive routing strategies, it is also necessary to use historical network data and environmental information to predict future network conditions and adjust routing strategies in advance to avoid task interruptions caused by network fluctuations.

[0100] The secure transmission strategy is to use encryption and authentication mechanisms to ensure the security of data after the routing strategy selects the optimal transmission path to prevent data from being stolen or tampered with.

[0101] Specifically, Figure 3 As shown, the corresponding computing power routing and the corresponding task scheduling process may include:

[0102] (51) The cloud platform receives computing service tasks submitted by users and determines the corresponding task requirements, such as computing power, storage, and network bandwidth requirements;

[0103] (52) Evaluate vehicles that can complete the task and meet the computing power requirements based on the predicted available time and computing power of currently available vehicles;

[0104] (53) After the corresponding vehicle that can meet the requirements of completing the computing service task is determined, a decision is made to determine the vehicle and transmission route for specifically executing the computing service task based on the network status and task requirements;

[0105] (54) dispatching the computing service task to the determined vehicle node that specifically executes the computing service task;

[0106] (55) Monitor changes in network and vehicle status during the execution of computing service tasks, and adjust the corresponding scheduling strategy in a timely manner according to demand to ensure that the corresponding computing service tasks are completed reliably.

[0107] (6) Computing power trading module

[0108] The computing power transaction module is used to establish a computing power service pricing transaction mechanism between user terminals, smart car users and the cloud. The transaction mechanism includes the user terminal paying the computing power service fee to the cloud platform, the cloud paying the computing power resource usage fee to the smart car user, and the cloud obtaining the corresponding intermediate service fee. The cloud is also responsible for establishing the transaction model and rules; that is, realizing the computing power transaction between the user terminal, the cloud platform and the smart car owner;

[0109] The corresponding computing power trading module not only needs to establish a transaction mechanism among users, platforms and car owners, but also needs to ensure that users pay reasonable fees to obtain computing power services, car owners can obtain reasonable benefits by sharing the idle computing power of vehicles, and the platform can charge reasonable fees by providing services. That is, the computing power trading module needs to fully consider the particularity of smart car scenarios and design a flexible, fair and efficient transaction model; to this end, the computing power trading module specifically needs to establish corresponding transaction mechanisms, pricing models, incentive mechanisms and legal and privacy protections, among which:

[0110] The transaction mechanism refers to the user terminal submitting a computing task (i.e., computing power service task) on the cloud platform. The cloud platform calculates the fee to be paid based on the resource requirements, urgency and estimated execution time of the task. After the user terminal confirms the fee, the payment can be completed through a secure payment method to obtain the required computing power service. The owner of the smart car registers on the platform and agrees to share the idle computing power of his smart car. The cloud platform can calculate the income that the owner should obtain based on the computing power resource type, performance indicators, available time and energy consumption cost provided by the owner. After the task is completed, the owner will obtain the corresponding income share, which will be directly transferred to his account. As an intermediary, the platform is responsible for services such as task scheduling, resource matching, security and transaction settlement. In order to maintain operations and sustainable development, the platform charges a certain percentage of handling fees from the fees paid by users.

[0111] The setting of the pricing model needs to take into account the dynamics of smart car computing resources and the complexity of the network environment. For this purpose, the platform can adopt a dynamic pricing model that refers to multiple pricing factors; the corresponding pricing factors may include the supply and demand relationship of computing power, the complexity of computing power service tasks, computing power quality (such as the computing performance and stability of smart car computing equipment), network conditions and energy consumption, etc.; the cloud platform needs to accurately calculate the car owner's energy consumption costs, equipment depreciation and potential maintenance costs to ensure that the car owner's income can cover the costs and obtain a reasonable profit; at the same time, the fees paid by the user terminal should also be competitive in the market and meet the user's expectations.

[0112] The incentive mechanism mentioned above refers to the fact that for the owners of smart cars, in order to encourage them to actively participate in computing power sharing, the cloud platform can establish various incentive measures; for example, for owners who provide high-quality computing power in a long-term and stable manner, additional income rewards can be given; a car owner rating system can be established to increase the income ratio according to the contribution; value-added services such as vehicle maintenance and maintenance discounts can be provided; for the users where the user terminals are located, the cloud platform can provide coupons, discounts, points and other preferential measures for users who frequently use computing power services, so as to effectively improve user stickiness and satisfaction; through the corresponding incentive mechanism, the owners of smart cars and the users of user terminals can be well enabled to actively participate in the cloud platform that provides computing power sharing, ensuring that all three parties benefit.

[0113] The legal and privacy protection mentioned above means that the cloud platform needs to formulate clear service agreements and privacy policies to ensure that the rights and interests of car owners, users of user terminals and cloud platform parties are protected, taking into account the privacy and legal risks of car owners, users of user terminals and cloud platform parties.

[0114] In the above system provided by the present invention, in the process of implementation on a smart car, a smart car for realizing idle computing power sharing is specifically provided. The smart car is a smart car for providing computing power services in the above system, and may include the following processing modules in the process of implementation:

[0115] A computing power sensing module, used to monitor whether the computing power of the smart car is available and the available time period information;

[0116] A computing power measurement module; used to obtain the amount of available computing power of the smart car, including a unified measurement of heterogeneous computing power of different smart cars;

[0117] The vehicle-mounted computing power execution module is used to execute the computing power service task when the smart car receives a notification to execute the computing power service task.

[0118] Specifically, the vehicle-mounted computing power execution module may further include a computing power service task computing execution submodule, a resource management submodule, and a computing security processing submodule, wherein:

[0119] The computing execution submodule is used to execute the computing tasks included in the computing power service tasks allocated by the cloud;

[0120] The resource management submodule is used to manage the computing resources of the smart car and control the provision of computing resources for the computing service task while giving priority to the basic functions and driving requirements of the vehicle;

[0121] The computing security processing submodule is used to implement end-to-end encryption of data, data access control management, and realize the anti-tampering function of each module during the execution of computing service tasks.

[0122] It should be noted that the specific implementation methods of the above modules have been described in detail in the description of the above system, so they will not be repeated here.

[0123] The technical framework system for sharing idle computing power of smart cars constructed by the present invention and the corresponding smart cars realize efficient perception, measurement, routing, orchestration and trading of vehicle computing resources. Moreover, in view of the particularity of smart car scenarios, innovative solutions are proposed, providing new ideas and technical support for the field of edge computing and resource sharing. That is, the technical solution provided by the present invention has the following technical advantages and can produce the following positive technical effects:

[0124] (1) Through personalized computing power perception, deep learning of the user behavior of smart cars is carried out to accurately predict the idle time of vehicles, thereby effectively improving the utilization rate of computing power resources;

[0125] (2) It provides a unified measurement solution for heterogeneous computing resources in smart cars, effectively solving the technical problem of computing power evaluation between different hardware architectures;

[0126] (3) The adaptive computing power routing strategy enables the technical solution to be applicable to the adaptive routing algorithm in complex wireless network environments, ensuring efficient and secure transmission of task data;

[0127] (4) Based on the dynamic nature of intelligent vehicle computing resources and the continuity of task execution, an optimized task scheduling and dispatching model is provided;

[0128] (5) An exclusive computing power trading mechanism has been set up, which can combine the characteristics of smart cars to achieve flexible pricing and incentive mechanisms, and promote the development of the computing power sharing ecosystem;

[0129] Through the above technical advantages, the present invention can produce corresponding beneficial technical effects including:

[0130] The computing power resources of smart cars can be used to provide nearby edge computing services, reduce the load of cloud computing and network latency, and realize edge computing. Moreover, car owners can gain profits by sharing the idle computing power of their vehicles, realize efficient use of resources and maximize value, and promote the development of the resource sharing economy. In addition, by making full use of existing computing resources, the demand for new data centers can be reduced, energy consumption and carbon emissions can be reduced, and corresponding environmentally friendly computing services can be realized.

[0131] The technical framework and system design of the present invention will be described in detail below in conjunction with specific application embodiments, but the protection scope of the present invention is not limited thereto.

[0132] The embodiment of the present invention provides an idle computing power sharing system based on electric smart cars, that is, a system for realizing idle computing power sharing of smart cars, still referring to Figure 1 As shown, it may include:

[0133] The computing power perception module is used to monitor the vehicle (smart car) operating status, battery power, hardware temperature and other information in real time, and use sensor data and user driving habits to predict the vehicle's idle time period;

[0134] A computing power measurement module is used to evaluate the available computing power of the CPU, GPU and other specialized computing chips (such as ASICs for autonomous driving) in the vehicle and quantify it into standard computing resource indicators (such as FLOPS);

[0135] The computing power sensing module and computing power measurement module can be set in a smart car, or can be independently set in a device that communicates with a cloud management platform (referred to as the cloud) or set on a device belonging to the network-side cloud management platform;

[0136] The computing power routing module is used to dynamically allocate the corresponding computing power service tasks according to the computing power service tasks requested by the user and the computing power availability of the vehicle, and then use the adaptive routing algorithm to select the transmission path with the best network condition from the user terminal to the vehicle (smart car) for information transmission, so as to complete the execution process of the corresponding computing power service tasks through the best transmission path;

[0137] The computing power orchestration module is used to decompose and schedule the computing tasks (i.e., computing power service tasks) submitted by users. During the decomposition and scheduling process, the vehicle's battery power and expected idle time need to be considered to formulate the optimal task execution plan.

[0138] The computing power trading platform (or computing power trading module) is used to handle the payment process of computing power service tasks initiated by users, as well as the profit settlement process of computing power resources shared by smart car owners, thereby providing transparent transaction records and capital flow process.

[0139] The computing power routing module, computing power orchestration module and computing power trading platform can be set in the settings belonging to the cloud management platform, or can also be set in an independent device and establish communication with the cloud.

[0140] In the above system, the user terminal side usually includes multiple computing power users (such as computing power user 1, computing power user 2, computing power user 3, etc.), and the computing power users are users who can initiate computing power service task requests; the vehicle-side terminal point of the system (that is, the smart car side) can usually include multiple smart cars (such as smart car 1, smart car 2, smart car 3, etc.), and the smart cars can provide computing power resources with computing capabilities, that is, they can share their idle computing power resources with user terminal side applications through the cloud.

[0141] In the process of providing computing power sharing services, the above system may specifically include the following processing steps:

[0142] Computing power perception and prediction process

[0143] The vehicle-mounted node (including the computing power perception module) collects the vehicle's historical driving data and user schedules, and predicts future idle time periods through machine learning models;

[0144] For example, you can apply Figure 2 In the processing process shown, based on the user's usual parking state from 9:00 to 18:00 on weekdays, the system predicts that the vehicle can provide computing power sharing services during this time period;

[0145] Hashrate Measurement and Reporting

[0146] Evaluate the vehicle's available computing resources through the computing power measurement module;

[0147] For example, you can apply Figure 2 From the processing process shown, it is learned that the vehicle's GPU currently has 70% idle resources, the CPU has 50% idle resources, the battery power is 80%, and it is estimated that it can provide 4 hours of computing power service continuously; in this way, the corresponding computing power information can be uploaded to the cloud management platform for its task scheduling.

[0148] Task submission and scheduling

[0149] The processing process is as follows Figure 3 As shown, the user submits an AI task (i.e., computing service task) through the user terminal to require GPU computing power and sets the latest completion time; after receiving the AI ​​task, the cloud management platform searches for available vehicle computing resources according to its task requirements, selects a qualified vehicle (such as the self-measurement and the vehicle exemplified in the report above), and assigns the task to the vehicle;

[0150] Task execution and monitoring

[0151] After receiving the computing service task, the vehicle-mounted node (including the vehicle-mounted computing execution module) starts to execute the corresponding task. During the execution of the task, the hardware status can be monitored in real time to ensure that its temperature and battery power are within a safe range. If the vehicle status changes (such as the owner starts the vehicle and prepares to drive), the system needs to feedback the task status to the cloud management platform.

[0152] Task migration and fault tolerance;

[0153] When a change in the vehicle status is detected, for example, the vehicle needs to drive immediately, the computing service task being executed will be suspended, and the cloud management platform will migrate the unfinished computing service task to other available vehicles or edge servers to ensure the continuity and reliability of the computing service task execution;

[0154] Result return and transaction settlement

[0155] After the computing power service task is completed, the execution result will be returned to the user terminal through a secure network channel; and the cloud management platform will calculate the fees that the user on the user terminal needs to pay and the benefits that the owner of the smart car that provides computing power resource sharing should obtain based on parameters such as the actual computing time and resource consumption during the execution of the computing power service task, and complete the corresponding transaction settlement based on the calculation results.

[0156] The embodiment of the present invention can also provide a multi-vehicle collaborative computing power sharing service, and the corresponding processing process may include:

[0157] Task decomposition process

[0158] When a user submits a computing service task that requires a large amount of computing resources through a user terminal, such as complex scientific calculations or large-scale machine learning model training, the cloud management platform needs to decompose the task into multiple computing service task subtasks, and multiple subtasks can be executed in parallel;

[0159] Resource Scheduling Process

[0160] That is, according to the computing power availability and network conditions of each vehicle, the above subtasks are assigned to different vehicles for execution;

[0161] Task execution process

[0162] Each vehicle independently performs the assigned subtasks. When the system takes network latency and bandwidth into consideration, subtasks that require frequent communication can be assigned to vehicles that are physically close to each other as much as possible.

[0163] Summary of multiple subtask execution results

[0164] After multiple subtasks are completed, the corresponding execution results need to be uploaded to the cloud management platform, and the cloud platform will summarize and integrate the corresponding results to generate the final task output to the user terminal side.

[0165] The above-mentioned embodiments of the present invention provide a system for sharing idle computing power of smart cars, which can be applied to various types of smart vehicles with certain computing capabilities, including electric vehicles, hybrid vehicles, and traditional fuel vehicles, etc. The corresponding system can be widely used in edge computing, distributed computing, Internet of Things data processing, artificial intelligence reasoning and training, and other fields.

[0166] In summary, the above-mentioned system for realizing the sharing of idle computing power of smart cars can specifically include the process of car owner registration, user task submission, task execution, result acquisition and profit settlement in actual use, among which:

[0167] The owner registration process refers to the owner registering an account on the cloud management platform, downloading and installing the vehicle-mounted node software; and setting sharing preferences, including the type of computing resources available, sharing time period, battery power threshold, etc.

[0168] The user task submission process refers to the user submitting a computing task (computing service task) through the application of the user terminal and setting the task parameters and budget of the computing service task;

[0169] The task execution process is that the system automatically completes the scheduling, allocation and execution of tasks, and users do not need to pay attention to the underlying details;

[0170] The process of obtaining the results is that after the computing power service task is completed, the user can view and download the corresponding calculation results on his user terminal side;

[0171] The process of profit settlement includes that car owners can view the profits obtained from their shared computing power on the platform, and can support cash withdrawal or use for platform service consumption, etc.

[0172] It can be seen from the description of the above embodiments that the embodiments of the present invention provide an efficient and flexible intelligent vehicle idle computing power sharing system, thereby achieving full utilization of idle computing resources of vehicles. The system has wide applicability and scalability, can meet the computing needs of different users, bring additional benefits to car owners, and also contribute to the optimization of computing resources in society.

[0173] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed in the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims. The information disclosed in the background technology section of this article is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as an admission or in any form that the information constitutes prior art known to those skilled in the art.

Claims

1. A system for sharing idle computing power of smart cars, characterized in that: include: A user terminal as a computing power user, a smart car for providing computing power resource services, and a cloud for scheduling the computing power of the smart car: and the system also includes: A computing power sensing module, used to monitor whether the computing power of the smart car is available and the available time period information; A computing power measurement module; used to obtain the amount of available computing power of the smart car, including a unified measurement of heterogeneous computing power of different smart cars; A computing power scheduling module, used to schedule the computing power of the smart car to provide computing power services for the user terminal according to the processing results of the computing power perception module and the computing power measurement module; The vehicle-mounted computing power execution module is used to execute the computing power service task when the smart car receives the notification of executing the computing power service task given by the computing power orchestration module.

2. The system according to claim 1, characterized in that The computing power perception module includes: A corresponding personalized model is established by using personalized vehicle usage data of a smart car in a machine learning manner; the personalized vehicle usage data includes one or more of historical usage data of the smart car, daily usage habits, and data on the impact of external environment data on its usage; Determine the idle time period information of the smart car according to the personalized model, and determine the resource computing power information that can be shared and used according to the smart car power and computing device hardware parameters obtained by monitoring; The idle time period information and the resource computing power information that can be shared are uploaded and used as basic data of the computing power orchestration module.

3. The system according to claim 2, characterized in that The computing power perception module also includes: Real-time monitoring of the smart car's operating status, battery power, hardware temperature, and network connection parameter information; and dynamic perception of changes in the smart car's usage status to update the available computing power in real time; The idle time period information and the resource computing power information that can be shared are determined based on the real-time monitoring and dynamic perception results, and uploaded as the basic data of the computing power orchestration module.

4. The system according to claim 1, characterized in that The computing power measurement module includes: quantifying different types of computing resources of the smart car in a standardized manner to perform heterogeneous computing power modeling; And, based on the output evaluation of the computing power perception module, the availability of the computing power resources of the smart car and the computing power resource performance level that it can provide are determined.

5. The system according to claim 4, characterized in that The process of heterogeneous computing power modeling includes using floating-point computing power, memory bandwidth and computing unit quantity indicators to conduct comprehensive evaluation to achieve heterogeneous computing power modeling.

6. The system according to claim 1, characterized in that The computing power orchestration module includes a submodule for decomposing the computing power service tasks of the user terminal, a distribution submodule, and a management submodule, wherein: The decomposition submodule is used to decompose the computing power service task that needs to be decomposed into several subtasks; The allocation submodule is used to allocate the computing service task or several subtasks of the decomposed computing service task to the smart car providing computing service according to a predetermined rule; the predetermined rule refers to one or more of the available time, utilization rate, energy consumption and battery status of the available computing resources of the smart car; The management submodule is used to support dynamic migration or fault-tolerant processing of the computing service task or the subtask when the start / stop state of the smart car changes and / or the network is interrupted.

7. The system according to any one of claims 1 to 6, characterized in that: The system also includes a computing power routing module, and the processing performed by the computing power routing module includes: According to the prediction result of the network status, an adaptive routing strategy is used to transmit the task data of providing computing power services. The adaptive routing strategy refers to selecting the optimal transmission path according to the real-time network bandwidth, delay, signal strength and stability, and the task data is also transmitted using encryption and authentication mechanisms; It also uses historical network data and environmental information to predict future network conditions and adjusts adaptive routing strategies based on the prediction results; and / or monitors changes in network and smart car status during the execution of computing service tasks, and adjusts adaptive routing strategies based on the monitoring results.

8. The system according to any one of claims 1 to 6, characterized in that: The vehicle-mounted computing power execution module includes a computing power service task computing execution submodule, a resource management submodule and a computing security processing submodule, wherein: The computing execution submodule is used to execute the computing tasks included in the computing power service tasks allocated by the cloud; The resource management submodule is used to manage the computing resources of the smart car and control the provision of computing resources for the computing service task while giving priority to the basic functions and driving requirements of the vehicle; The computing security processing submodule is used to implement end-to-end encryption of data, data access control management, and realize the anti-tampering function of each module during the execution of computing service tasks.

9. The system according to any one of claims 1 to 6, characterized in that: The system also includes a computing power trading module, which is used to establish a computing power service pricing trading mechanism between user terminals, smart car users and the cloud. The trading mechanism includes the user terminal paying a computing power service fee to the cloud platform, the cloud paying a computing power resource usage fee to the smart car user, and the cloud obtaining a corresponding intermediary service fee. The cloud is also responsible for establishing trading models and rules.

10. A smart car that realizes idle computing power sharing, characterized in that: An intelligent vehicle for providing computing power services in the system according to any one of claims 1 to 9, comprising: A computing power sensing module, used to monitor whether the computing power of the smart car is available and the available time period information; A computing power measurement module; used to obtain the amount of available computing power of the smart car, including a unified measurement of heterogeneous computing power of different smart cars; The vehicle-mounted computing power execution module is used to execute the computing power service task when the smart car receives a notification to execute the computing power service task.

Images