Li Rui, Manager of Oriental New Energy Vehicle Hybrid Fund: The future of autonomous driving is bright and still needs to be down-to-earth

Reporter Wang Siwen

2021 is the first year of automotive intelligence, the progressive route L3 is initially introduced, the pattern of the crowd is undecided, and the current situation and future of the mass production of high-end intelligent driving technology are what? On October 15, the "2021 Auto Capital Forum" jointly sponsored by Securities Daily and Beijing Economic and Technological Development Zone and co-organized by West China Securities was successfully held in Beijing. Li Rui, manager of the Oriental New Energy Vehicle Hybrid Fund, said at the forum that the introduction path of autonomous driving in the future will be: from low scenes to high scenes, from closed scenes to open scenes, from carriers to manned, from commercial to human use. "The future of autonomous driving is bright, but we need to be down-to-earth together."

Li Rui mentioned the two rounds of market conditions on the autonomous driving track. He believes that the opportunity for the first round of autonomous driving comes from the industry's expectations. "There were a lot of events at that time, and the biggest background was the development of artificial intelligence technology, especially the widespread application of deep learning technology. It can be said that the application of deep learning technology is the first round of boom, which is the most important reason for the market, and the capital market has very high expectations. ”

"Looking back, the expectations at that time were a bit overly optimistic, and the most important reason was that the long-tail problem became the biggest obstacle to the landing of automatic driving, which was manifested in the redundancy of the system and the perfection of the test." Li Rui said.

Li Rui said frankly that the difference between the second round of market conditions and the first round is that the industrialization of automatic driving has really begun to land, rather than simple expectations. The reason why I say this is because of The development of China in recent years, many technologies, including perception lidar, high-computing chips, and the introduction of data and algorithm technologies, have made great progress, and many specific scenarios have gradually been commercialized. Whether it is a new force or a traditional car factory, there are many low-end autonomous driving passenger cars on the market and hot sales, and many related companies have also begun to go public. ”

Li Rui believes that the introduction path of autonomous driving in the future will be: from low scenes to high scenes, from closed scenes to open scenes, from carriers to people, from commercial to human use. "The future of autonomous driving is bright, but we need to be down-to-earth together."

When it comes to the core technology of autonomous driving, every aspect of perception, decision-making, and execution is crucial. Li Rui analysis believes that the most concerned at present is still in the decision-making process. "First of all, AI chips account for the largest proportion in hardware, and now more and more domestic companies, including Huawei, Horizon, etc., more and more high-quality chip companies have emerged, which will gradually become the choice of domestic alternatives for OEMs."

Li Rui further said, "For the decision-making link, the largest proportion of software is data, algorithm, because not only involves functional safety, but also involves data security, many times in fact, the state is also supporting domestic companies; for the perception link, now the solid-state lidar program has emerged, the cost of mechanical lidar has dropped to less than 1,000 US dollars; for the implementation link, the original chassis execution agency is monopolized overseas, and the progress of domestic localization substitution is relatively slow. Now OEMs need to integrate the chassis control mechanism more and more, and the integration process provides an opportunity for the development of the domestic chassis actuator. ”

In terms of application, Li Rui believes that landing should be a fusion of technology and scenarios, but the technical requirements in different scenarios are not the same. For example, the automatic driving of specific scenarios should be landed, the battery interference in the mine scene is relatively strong, the signal is weak, the power consumption is relatively weak, the requirements for safety in the grass-roots scene are relatively high, the air in the port scene is relatively humid or it will rain, and the technical requirements of different scenarios will be different.

In Li Rui's view, whether it is overseas or domestic, from the perspective of companies above the L4 leapfrog type, in fact, we have not seen mass production on the car. Mining vehicle routes may be equipped with more sensing equipment, and algorithms can be designed more advanced. Overall, Li Rui is more optimistic about the progressive route, but this also requires the company to have research and development capabilities comparable to L4.

For the technological evolution of the main engine factory, Li Rui believes that the head main engine factory has a strong power self-research, the integration of the traditional automobile manufacturing industry chain has entered a mature stage, and there is little room for continued cost reduction, if the electric intelligence track, the whole vehicle will relatively improve the value of bicycles and the utilization rate of bicycles. On the other hand, grasping this data and system is also capturing customers, especially for the head vehicle company is very motivated.

Specific to what is the relationship between the whole vehicle and the new force company, Li Rui believes that it is a competitive relationship, but the specific situation should be analyzed in detail. "For example, head OEMs and head new power companies may compete more strongly. Other head OEMs and other head new power companies may cooperate, but may also be acquired. If the head technology company has the strength of capital, it may build a car, and the non-head may have cooperation, and the dominant relationship in the process of cooperation, including the benefit distribution mechanism, needs to be paid attention to. ”

As for the new forces, Li Rui believes that it may be more likely to compete with the head OEMs, because their own software capabilities are stronger and their organizational structure is more flexible.

(Editing by Shangguan Monroe)