Cross-language vehicle cloud collaborative algorithm cloud integrated deployment method

Abstract

The invention relates to a cross-language vehicle cloud collaborative algorithm cloud integrated deployment method, which comprises the following steps of: packaging an SDK (Software Development Kit) by algorithms, respectively packaging the algorithms written by Python, C or C + + and Java programming languages into the SDK, and packaging a third-party library or a configuration file on which the algorithms depend into the SDK; algorithm SDK integration, algorithm SDK deployment, algorithm SDK calling and algorithm result issuing: the cloud control platform further processes the algorithm result, converts the result into a corresponding instruction or information according to a communication protocol between vehicles and cloud, and issues the instruction or information to the vehicle-mounted terminal through a communication network; algorithm reusability and integration efficiency are improved: by packaging algorithms written by different programming languages into a unified SDK form, differences between languages and environments are shielded, so that the algorithms can be conveniently reused in different projects, and the workload of repeated development is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent connected vehicle technology, and in particular to a cloud-based integrated deployment method for a cross-language vehicle-cloud collaborative algorithm. Background Technology

[0002] As vehicles become increasingly intelligent and connected, on-vehicle hardware resources, such as the computing power and storage capacity of computing chips, are relatively limited and struggle to support the increasingly complex demands of algorithm execution. As a result, algorithms are increasingly shifting from on-vehicle deployment to cloud deployment. For example, by deploying dynamics and fuel economy algorithms to the cloud, real-time interaction between vehicle and cloud data can optimize vehicle stability and reduce energy consumption. However, these algorithms are typically developed by different teams using programming languages ​​such as Python, C or C++, and Java, based on their respective technical strengths and needs. The differences in their operating environments and data structures present significant challenges in integrating and deploying algorithms written in different programming languages ​​onto the same cloud control platform. For instance, Python algorithms rely on the Python interpreter and specific library environments, C or C++ algorithms require appropriate compilation and linking configurations, and Java algorithms require Java Virtual Machine support. Therefore, a technical solution is urgently needed to address the challenges of integrating and deploying cross-language vehicle-cloud collaborative algorithms on a cloud control platform. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by providing a cloud-based integrated deployment method for cross-language vehicle-cloud collaborative algorithms.

[0004] This invention is achieved using the following technical solution:

[0005] A method for cloud-based integrated deployment of a cross-language vehicle-cloud collaborative algorithm includes:

[0006] Algorithm encapsulation SDK: Algorithms written in Python, C or C++, and Java programming languages ​​are encapsulated into SDKs respectively, and the third-party libraries or configuration files that the algorithms depend on are also packaged into the SDK;

[0007] Algorithm SDK integration involves placing the packaged SDK on a specified path on the cloud control platform server. The cloud control platform then writes platform code to integrate the SDK based on the type and structure of the algorithm SDK's input and output parameters.

[0008] After the cloud control platform completes the code writing for the integrated SDK, it deploys the integrated code to the cloud control platform via Docker to achieve resource isolation and version management of the algorithm SDK runtime environment.

[0009] When the cloud control platform receives a calculation request, it processes the data uploaded by the vehicle terminal according to the type and structure of the algorithm input parameters, and then inputs it into the algorithm SDK for calculation to obtain the algorithm output result.

[0010] Algorithm result distribution: The cloud control platform further processes the algorithm results and, according to the communication protocol between the vehicle and the cloud, converts the results into corresponding instructions or information, which are then distributed to the vehicle terminal through the communication network.

[0011] Preferably, the algorithm written in Python is packaged into an executable file under Linux;

[0012] Algorithms written in C or C++ programming languages ​​are packaged into .so files;

[0013] Algorithms written in the Java programming language are packaged into .Jar files.

[0014] Preferably, the algorithm SDK deployment includes: writing a Dockerfile, using Docker commands to build the integrated code into a Docker image, using commands to assign a specific version label to the image during image building to achieve algorithm version management, then creating a Docker container based on the Docker image, and finally running the Docker container to complete the deployment of the algorithm SDK.

[0015] Preferably, the communication network uses wireless communication technology for data transmission between the vehicle terminal and the cloud control platform.

[0016] Preferably, the vehicle-mounted terminal collects various types of vehicle data in real time, performs preliminary processing and encoding on the data, and uploads it to the cloud control platform via a communication network.

[0017] Preferably, the various data of the vehicle include engine speed, vehicle speed, gear, and gradient.

[0018] Compared with the prior art, the present invention has the following beneficial technical effects:

[0019] Improving algorithm reusability and integration efficiency: By encapsulating algorithms written in different programming languages ​​into a unified SDK, language and environment differences are shielded, enabling algorithms to be easily reused in different projects and significantly reducing the workload of repetitive development. At the same time, the standardized encapsulation method also facilitates algorithm integration. Actual testing shows that cross-language algorithm integration efficiency is improved by approximately 45%, significantly reducing the difficulty and cost of platform development.

[0020] Achieving efficient vehicle-cloud collaboration: A complete vehicle-cloud collaborative algorithm data interaction link has been constructed, realizing real-time data collection from the vehicle, efficient processing of cloud algorithms, and rapid feedback of results, forming a closed-loop data management system. This enables vehicles to make timely adjustments and optimizations based on the analysis results from the cloud.

[0021] Facilitates algorithm updates and maintenance: Since the algorithm is deployed via Docker in the form of an SDK, when the algorithm needs to be updated or maintained, only the corresponding SDK needs to be updated, without the need for large-scale modifications and deployments to the entire platform. According to statistics, the time required for algorithm update and deployment is reduced by about 50% compared to the original, reducing the platform's maintenance costs and risks.

[0022] Enhance the platform's scalability and resource utilization: New algorithm SDKs can be added at any time according to business needs, enabling the platform to quickly adapt to new business scenarios and functional requirements;

[0023] Furthermore, by choosing different programming languages ​​to write algorithms and encapsulating them accordingly, the advantages of each programming language can be fully utilized. Meanwhile, in terms of resource utilization, Docker's resource isolation and scheduling mechanism increases server resource utilization from 35% to 60%, effectively improving platform resource utilization. Attached Figure Description

[0024] Figure 1 This is a logic diagram of the present invention;

[0025] Figure 2 This is a data interaction link diagram for the present invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0027] cloud control platform

[0028] The cloud control platform is a distributed system built on technologies such as cloud computing and big data. As the core hub of intelligent connected vehicles, its core functions include: real-time aggregation of vehicle network data, supporting high-concurrency access of millions of vehicle terminal data; real-time processing and in-depth analysis of collected vehicle operating status and sensor data using distributed computing and stream processing technologies; and precise issuance of control commands to downstream vehicle terminals through standardized interfaces based on the analysis results.

[0029] Vehicle-cloud collaboration

[0030] Vehicle-cloud collaboration refers to the efficient collaboration of data, computing, and control between in-vehicle terminals and cloud control platforms through a stable, high-speed communication network. The in-vehicle terminal collects vehicle operating data in real time and uploads it to the cloud. The cloud control platform utilizes its powerful computing resources to analyze and process the data, execute complex algorithms, and promptly feed the results back to the in-vehicle terminal to guide vehicle decision-making and control. Through vehicle-cloud collaboration, applications such as vehicle performance optimization, intelligent driving assistance, and remote fault diagnosis can be realized, improving the vehicle's intelligence level and safety.

[0031] SDK

[0032] SDK stands for Software Development Kit. By encapsulating algorithms into toolkits, providing easy-to-use interfaces and documentation, other developers can easily call its methods. SDKs can simplify the development process, improve code reusability and maintainability, while hiding complex implementation details.

[0033] PyInstaller

[0034] PyInstaller is a tool for packaging Python scripts into standalone executable files. It can package Python code, its dependent libraries, and runtime environments together to generate files that can run on the target platform without a Python interpreter. In this invention, it is used to encapsulate Python algorithms into executable files for Linux.

[0035] GCC

[0036] GCC is an abbreviation for GNU Compiler Collection, a suite of programming language compilers that supports multiple languages, including C and C++. In this invention, it is used to compile and package C or C++ algorithms into .so files, facilitating integration with cloud control platforms.

[0037] Maven

[0038] Maven is a project management and build automation tool, primarily used for Java projects. It can manage project dependencies and automatically download necessary library files based on project configuration. It can also package Java projects into .jar files, and in this invention, it is used to package Java algorithms into integrable .jar files.

[0039] Docker

[0040] Docker is an open-source application container engine that packages applications and their dependencies into a portable container, enabling applications to run in the same way across different environments. In this invention, it is used to deploy the algorithm SDK on a cloud control platform, ensuring consistency and isolation of the deployment environment.

[0041] Vehicle-to-Cloud Communication Protocol

[0042] The vehicle-to-cloud communication protocol refers to the rules and conventions followed when transmitting data between the vehicle terminal and the cloud control platform, including data format, transmission method, error handling, etc., to ensure that data can be exchanged between the two accurately and in an orderly manner.

[0043] like Figure 1-2 As shown, a cloud-based integrated deployment method for a cross-language vehicle-cloud collaborative algorithm includes:

[0044] Algorithm encapsulation SDK: Algorithms written in Python, C or C++, and Java programming languages ​​are encapsulated into SDKs respectively, and the third-party libraries or configuration files that the algorithms depend on are also packaged into the SDK;

[0045] Algorithm SDK integration involves placing the packaged SDK on a specified path on the cloud control platform server. The cloud control platform then writes platform code to integrate the SDK based on the type and structure of the algorithm SDK's input and output parameters.

[0046] After the cloud control platform completes the code writing for the integrated SDK, it deploys the integrated code to the cloud control platform via Docker to achieve resource isolation and version management of the algorithm SDK runtime environment.

[0047] When the cloud control platform receives a calculation request, it processes the data uploaded by the vehicle terminal according to the type and structure of the algorithm input parameters, and then inputs it into the algorithm SDK for calculation to obtain the algorithm output result.

[0048] Algorithm result distribution: The cloud control platform further processes the algorithm results and, according to the communication protocol between the vehicle and the cloud, converts the results into corresponding instructions or information, which are then distributed to the vehicle terminal through the communication network.

[0049] After receiving the data uploaded by the vehicle terminal through the communication network, the cloud control platform processes the data into the type and format required by the deployed algorithm SDK. Then, when the algorithm SDK is called, the processed data is input into the algorithm SDK. After the algorithm calculates the result, it is sent to the vehicle terminal.

[0050] When the deployed cloud-based fuel economy algorithm is invoked, the cloud control platform inputs the processed vehicle speed, engine speed, gear, and gradient data into the algorithm. After the algorithm calculates, it obtains the vehicle speed result planned by the cloud-based fuel economy algorithm. Finally, the result is sent to the vehicle terminal, and the vehicle travels at the speed planned by the algorithm to achieve the purpose of saving energy consumption.

[0051] Algorithms written in the Python programming language are packaged into executable files under Linux;

[0052] Algorithms written in C or C++ programming languages ​​are packaged into .so files;

[0053] Algorithms written in the Java programming language are packaged into .Jar files.

[0054] The algorithm SDK deployment process involves writing a Dockerfile, using Docker commands to build the integrated code into a Docker image, assigning explicit version tags to the image during image building to achieve algorithm version management, creating a Docker container based on the Docker image, and finally running the Docker container to complete the deployment of the algorithm SDK.

[0055] For executable files on Linux that encapsulate Python algorithms, the cloud control platform can call them using the ProcessBuilder method in Java. For .so files that encapsulate C or C++ algorithms, the cloud control platform can call the algorithms in the SDK based on the Java Native Interface (JNI). For jar files that encapsulate Java algorithms, they can be placed in the Java library path of the cloud control platform, and the methods in the jar file can be called directly using Java.

[0056] The communication network uses wireless communication technology for data transmission between the vehicle terminal and the cloud control platform.

[0057] The vehicle-mounted terminal collects various types of vehicle data in real time, performs preliminary processing and encoding on the data, and uploads it to the cloud control platform via the communication network.

[0058] Vehicle data includes engine speed, vehicle speed, gear, and gradient.

Claims

1. A cloud-based integrated deployment method for a cross-language vehicle-cloud collaborative algorithm, characterized in that, include: Algorithm encapsulation SDK: Algorithms written in Python, C or C++, and Java programming languages ​​are encapsulated into SDKs respectively, and the third-party libraries or configuration files that the algorithms depend on are also packaged into the SDK; Algorithm SDK integration involves placing the packaged SDK on a specified path on the cloud control platform server. The cloud control platform then writes platform code to integrate the SDK based on the type and structure of the algorithm SDK's input and output parameters. After the cloud control platform completes the code writing for the integrated SDK, it deploys the integrated code to the cloud control platform via Docker to achieve resource isolation and version management of the algorithm SDK runtime environment. When the cloud control platform receives a calculation request, it processes the data uploaded by the vehicle terminal according to the type and structure of the algorithm input parameters, and then inputs it into the algorithm SDK for calculation to obtain the algorithm output result. Algorithm result distribution: The cloud control platform further processes the algorithm results and, according to the communication protocol between the vehicle and the cloud, converts the results into corresponding instructions or information, which are then distributed to the vehicle terminal through the communication network.

2. The cloud-based integrated deployment method for cross-language vehicle-cloud collaborative algorithms according to claim 1, characterized in that, Algorithms written in the Python programming language are packaged into executable files under Linux; Algorithms written in C or C++ programming languages ​​are packaged into .so files; Algorithms written in the Java programming language are packaged into .Jar files.

3. The cloud-based integrated deployment method for cross-language vehicle-cloud collaborative algorithms according to claim 2, characterized in that, The algorithm SDK deployment includes writing a Dockerfile, using Docker commands to build the integrated code into a Docker image, using commands to assign explicit version tags to the image during image building to achieve algorithm version management, then creating a Docker container based on the Docker image, and finally running the Docker container to complete the deployment of the algorithm SDK.

4. The cloud-based integrated deployment method for cross-language vehicle-cloud collaborative algorithms according to claim 1, characterized in that, The communication network uses wireless communication technology for data transmission between the vehicle terminal and the cloud control platform.

5. The cloud-based integrated deployment method for cross-language vehicle-cloud collaborative algorithms according to claim 4, characterized in that, The vehicle-mounted terminal collects various types of vehicle data in real time, performs preliminary processing and encoding on the data, and uploads it to the cloud control platform via a communication network.

6. The cloud-based integrated deployment method for cross-language vehicle-cloud collaborative algorithms according to claim 5, characterized in that, The vehicle's various data include engine speed, vehicle speed, gear, and gradient.

Images