With the development of assisted autonomous driving, AI technology plays an increasingly important role in key nodes in the vehicle cloud, and is expected to gradually realize the whole process of AI. According to EO Intelligent, the application of AI technology in the field of vehicle cloud is deepening day by day, and its data training and simulation optimization functions are constantly expanding to a wider range of links. Initially used in simulation testing of autonomous driving, AI technology has now been gradually extended to large-scale model training and OTA upgrades for intelligent cockpits. OTA stands for Over the Air Technology, which refers to over-the-air download technology, which can be upgraded remotely.
EO Intelligent believes that AI technology will play a central role in the update and iteration of applications and algorithms in various fields of automobiles in the future. According to EO Intelligent, AI technology in the field of vehicle cloud can be divided into three stages: exploration and application stage, preliminary application stage, and deepening application stage. 2006-2026 is the exploration and application stage, mainly for cloud computing + AI models in applications. The cloud computing platform embeds MATLAB/SPSS/Simulink and other software including AI algorithms to realize AI-based data processing, data annotation, and model training. It is mainly used in autonomous driving simulation training. The period from 2017 to 2026 is the initial application stage of AI in the vehicle cloud, mainly cloud native + AI model and cloud native + AI model + intelligent computing center. The test platform uses AI technology to improve efficiency and reduce costs in the above steps by using cloud-native data processing, hard example mining, data annotation, model training, scenario construction, and simulation testing. It is mainly used in autonomous driving model training and simulation testing, intelligent cockpit large model training and OTA upgrade. 2027 and beyond will be the stage of deepening the application of AI in the vehicle cloud, mainly AI cloud native. All aspects of software technology update and iteration are based on native development, and AI technology participates to realize the whole process of software iteration AI. It is mainly used in the application of all fields of AIEV and the update and iteration of algorithms.
Figure: The update and iteration of AI technology applications and algorithms in various automotive domains
According to the analysis of HUAWEI cloud, HUAWEI cloud data covers vehicle-end, road-end, and cloud services from the platform, ensuring that the full-link data from vehicles to infrastructure to cloud is satisfied. The core advantage of the platform is that it is based on AI technology, which is not only used to transform and process raw data through data cleaning, storage, analysis and mining, but also for data training and inference to improve the accuracy of processing. This series of processes not only ensures the quality of data, but also digs deep into the potential value and patterns of data, thereby significantly improving the development and utilization of data. In addition, Huawei's autonomous driving development platform provides end-to-end solutions, and AI technology empowers the autonomous driving simulation and evaluation system, leveraging the large computing power, large storage, and high speed of the cloud platform to meet the requirements of large-scale parallel simulation. With the support of AI technology, Huawei's autonomous driving development platform can help enterprises quickly build autonomous driving moats, deeply explore the value of data with low data storage and computing costs, and efficiently use autonomous driving scenario datasets. At the same time, it can help enterprises optimize resource scheduling, manage all aspects of data closed-loop, and shorten the iteration cycle of autonomous driving technology.
Huawei's cloud edge cloud data platform uses data as an asset and integrates AI training and inference models, such as the three-electric analysis model, driving behavior analysis model, vehicle residual value evaluation model, and comprehensive operation evaluation model, to realize the intelligence of vehicle cloud services and significantly improve the analysis efficiency with the support of AI technology. At the same time, HUAWEI CLOUD provides autonomous driving development cloud services, and AI technology-powered simulation scenario library + autonomous driving evaluation system to improve the large-scale training of autonomous driving algorithms and accelerate algorithm verification and mass production.
Relateds:
Artificial Intelligent Usher an Era of AIEV (1)
Artificial Intelligent Usher an Era of AIEV (2)
Artificial Intelligent Usher an Era of AIEV (3)
