The 2024 intelligent internet connection new energy automobile parts industry ecological conference was held in Chongqing, China's "first automobile city" on the 29th. "Global Intelligent Internet connected cars have entered a state of accelerated competition." Li Keqiang, an academician of the Chinese Academy of Engineering, said at the meeting and analyzed five major problems faced by the large-scale industrialization of smart internet vehicles.

The first is the lack of understanding of the "essential limitations" of bicycle intelligence. AI a large model based on bicycle autonomous driving lacks correct solutions to performance bottlenecks, long tail problems, and data limitations. For example, in terms of data, the information density is low, the data quality is poor, and the collection cost is high; In terms of computing power, the training cost is high and the computing power of the vehicle is insufficient.

The second problem is that the product definition is inaccurate and it is difficult to form a business closed loop. There are two reasons for this. One is that the current research and development and demonstration of vehicle-road cloud integration are still in the initial stage. The products are mainly based on bicycle intelligence, and the data of vehicle enterprises have not been accessed or received. On the other hand, currently, most vehicle-road cloud integration systems are still chimney-type architectures, without hierarchical decoupling and cross-domain sharing.

The third problem is that some key technologies still need to be broken through and technology closed loop needs to be established. Facing the characteristics of cross-border integration of intelligent internet-connected vehicles, new development methods, processes, tools, etc. have not yet been formed, and cannot support the complete and efficient development, testing and application of intelligent internet-connected vehicles, perception, decision-making, control, common key technologies and core capabilities such as networking still need further breakthroughs.

The fourth problem is the low level of competition in the industry, which is difficult to make profits. The market is facing unprecedented fierce competition, shifting from product competition to industrial chain Competition. The focus of intelligence lies in intelligent cockpit and luxury configuration. Some innovations of intelligent driving are stagnant, technological progress is slow, and investment is lacking.

The fifth problem is that laws and regulations need to be improved and the ecological system needs to be constructed urgently. There is insufficient support for autonomous driving policies and regulations, and the right-of-way policies or regulations supporting all-weather and all-time automatic driving commercial applications have yet to be broken. Autonomous driving still focuses on demonstration operation, and normalized operation still needs to be promoted, the ecological system of policies, regulations and ethical cognition still needs to be continuously improved.

While raising questions, Li Keqiang stressed that we should firmly develop confidence, the development trend of intelligent internet connection vehicles is irreversible, maintain strategic determination, pay attention to new components, and continuously promote the technological progress of intelligent internet connection; fully understand the current problems that need to be solved and the causes of the problems, form high-quality solutions and systematically promote them.

Chongqing is one of the important automobile manufacturing bases in China. In the first half of 2024, the city produced 1.214 million cars, becoming the "first city of cars" in the country ". Among them, the output of new energy vehicles is 391000, with a year-on-year growth of 1.5 times, ranking first among the top ten provinces and cities in automobile production. At present, the total number of new energy parts enterprises in the city is more than 540.

Li Keqiang put forward four suggestions for the high-quality development of Chongqing intelligent internet connection vehicles. The first is to practice the correct intelligent driving scheme, develop the integrated technology of vehicle and road cloud, effectively improve the level of automatic driving technology, form a business model and application ecology, and revitalize urban infrastructure and data assets. The second is to seize the window period of intelligent networking and strengthen the research and development of key parts technology. Third, it is based on the industrialization of five basic platforms (new "parts") to promote the high-quality development of intelligent internet-connected vehicles. Fourth, accelerate the transformation and upgrading of auto parts enterprises and promote the extension of advantageous industries. (Finished)