This article is from the WeChat public account "Yiou Network", author: Liu Huiying, editor: Hao Qiuhui, Pencil Road is published with permission.
In recent years, the automotive industry is in the midst of a dramatic upheaval, with leaps and bounds almost every year.
In the past 2021, new energy vehicles accounted for more than 10% of China's auto market, lidar was installed in mass production models for the first time, Robotaxi scenario-based landing operation was officially opened, and major car companies have released carbon neutrality goals...
Time into 2022, what new changes will the automotive travel industry usher in? Where will the winds of the industry blow? Who is the most favored pearl in the palm of the capital market
Autonomous driving shows the trend of "five modernizations"
In the past 2021, autonomous driving has become a veritable hot track.
According to Yiou Automobile's "2021 China Auto Travel Industry Investment and Financing Market Insight Research Report (including list)", the autonomous driving track ranks first in the total number of financing events in the nine major tracks of the automobile travel industry, with a total of 102 financing events, the market heat is the highest, the financing amount reaches 40.74 billion yuan, and the industry gradually develops to the deep water area.
The capital bubble of autonomous driving may burst, but technology will evolve over time to fake and remain true.
In 2022, the popularity of "automatic driving" will continue, but the center of heat will be boldly transferred from the stage of capital support and Demo support to technology on the car and mass production.
After industry visits and research, Yiou Automobile believes that autonomous driving will show the development of "five modernizations" in 2022: standardization of supporting regulations, accelerated autonomy of chips, refinement of scenes, normalization of mass production, and clarity of competition formats.
At the policy level, supporting regulations have developed in standardization. With the popularization and development of autonomous driving technology, policies and regulations urgently need to take the lead, provide directional guidance for the development of autonomous driving from a macro perspective, and provide policy support for the industry, improve the degree of legal norms, and accelerate the formulation of industry standards.
At the technical level, chips will accelerate the development of autonomy. Lack of core has become a major problem in the development of the industry, the automobile industry is overwhelmed, in order to get rid of the lack of core of the card neck as soon as possible, China's core-making enterprises will accelerate the independent research and development of chips, respond to market demand at a faster speed, and accelerate algorithm upgrades at the product application level to provide stability support for the development of automatic driving technology.
At the scene level, the scene operation will be more refined. The operation of autonomous driving in closed scenarios has entered the stage of development and popularization, and with the iterative optimization of technology, autonomous driving will be put into use in more subdivided scenarios to provide professional services in specific scenarios. In the case of different professions, we will incubate self-driving manufacturers with strong competitiveness.
At the commercial level, autonomous driving will officially enter the stage of mass production normalization. In 2021, new energy vehicles have achieved explosive growth, and automatic driving has increasingly become the standard technology of new energy vehicles, further large-scale mass production, promoting the road of autonomous vehicles, realizing the collection of massive data, and thus upgrading the technology.
At the level of the competitive landscape, the competitive situation of autonomous driving will become clearer and clearer, showing a typical Matthew effect. Autonomous driving enterprises will form the first echelon camp and join forces with main engine factories to achieve ecological co-construction and accelerate the popularization of autonomous driving technology in a group development manner.
Removal of the "Sword of Damocles"
In 2021, the collective "lack of core" in the automotive industry has choked the throat of Chinese car companies' production capacity. In the future, how to make China's autonomous driving achieve self-hematopoiesis and remove the sword of Damocles hanging over Chinese car companies, Chinese technology companies are gathering momentum.
Chen Mo, founder of Tucson Future, is a pessimistic activist. In his eyes, although self-driving software is close to commercial mass production in some areas, supply chain and hardware problems are far from mature. In 2022, from the sensors, the car itself, to the redundant system, are all problems that autonomous driving companies need to solve.
If autonomous driving technology is to land, software and hardware are indispensable.
If only software is provided without a domain controller, it is difficult to make a product; if only the hash rate is provided to the depot, then it is not an integrated solution for the depot. In the actual development process, tucson future and other autonomous driving manufacturers have chosen a similar "soft and hard power" technology development path.
Chen Mo said bluntly: "The key to measuring the mass production capacity lies in whether there is a domain controller at the vehicle regulation level. "Behind the domain controller, there is a car specification level chip.
In 2022, the Chinese market has become the "gladiatorial arena" of the world's top automotive smart chips. Autopilot "Ready Player One" Tesla, from the early "Mobileye chip + self-developed algorithm", "NVIDIA chip + self-developed algorithm", to the final choice of "self-developed chip (FSD) + self-developed algorithm", Tesla has achieved an "Apple-style" breakthrough.
Worldwide, underlying technology manufacturers such as NVIDIA and Google have occupied a first-mover advantage in the field of autonomous driving. In order not to be constrained by the underlying R&D ecology, chip manufacturers such as Horizon and Black Sesame Intelligence have actively focused on the field of vehicle chips in recent years.
In 2021, Black Sesame Intelligent launched int8 hash rate 106TOPS, INT4 hashing power up to 196TOPS high-computing power autonomous driving chips; and the AI computing power of a single vehicle-grade chip on Horizon is currently up to 128TOPS.
Yang Yuxin, CMO of Black Sesame Intelligence, told Yiou Automobile that the supply chain system of autonomous driving is not yet fully mature, and domestic and foreign enterprises are in the stage of technology competition. What domestic chip companies have to do is to focus on their own technical advantages, constantly innovate and iterate, and strive to achieve vehicle-level performance through technological breakthroughs.
Including the new car-making force "Wei Xiaoli", many companies that are deeply engaged in automatic driving are doing pre-research of high-performance chips and reserve work of high-computing technology.
Wang Shuliang, deputy general manager of Jimu Intelligence, pointed out that ecological security issues cannot be ignored. At present, the standardized software architecture and safety-related software and hardware development process of automobile autonomous driving and assisted driving has always been under the constraints of foreign organizations, and we follow the international norms that have been adhered to for many years, and at the same time cultivate our own basic software development capabilities, with the intention of investing more resources to replace foreign forces.
It should be admitted that autonomous driving is still in the early stage of development of the industry, and in the process of moving forward, technology, business and safety are the invisible shackles that the entire industry has to take into account.
Making money is the last word
"Make money, business, don't be humble." A classic line of Ge You in "Let the Bullets Fly" is probably the common voice of entrepreneurs.
Being able to form a smooth business closed loop is the basic premise for entrepreneurs to achieve sustainable development. After all, having no market is the norm, and too many companies with good products and good technologies eventually die from not realizing the commercialization closed loop.
Yiou Automobile believes that in 2022, many autonomous driving companies that have entered the middle and late stages of entrepreneurship will concentrate on the commercial landing level.
Although autonomous driving pilots have blossomed everywhere, from the perspective of real trial operation, automatic driving is still in the stage of "early adopter".
Zhang Tianlei, CEO of Mainline Technology, told Yiou Automobile that enterprises need to bring innovative applications to the market through lighthouse projects and flagship customers in real scenarios to confirm that enterprises can quickly deliver products and achieve stable mass production.
He told Yiou Auto that Robotaxi is the most technically difficult in the autonomous driving subdivision scenario, while the unmanned heavy truck has both rigidity, urgency and certainty of realization. Autonomous driving needs to take into account the market volume and application maturity, in the short term, perhaps the feasibility of unmanned heavy trucks is higher.
But in Wang Shuliang's view, whether it is to solve the last kilometer of unmanned delivery cars, or low-speed automatic driving in specific scenarios, if you want to form a scale in the future, you must go through the entire business model - either reduce costs or improve operational efficiency, of course, the safety and reliability of the system is the first issue to consider.
Zhu Lei, founder and CEO of Mushroom Car Association, believes that the essence of automatic driving is unmanned operation, that is, this automated and intelligent operation process improves safety and productivity, and creates the value of "cost reduction and efficiency increase".
Scene disputes are inevitable, and 2022 will be a year when subdivision scenarios are gradually commercialized. Guo Jishun, vice president of Joyson Electronics, believes that the key factor at the stage of autonomous driving mass production is how to obtain a large amount of structured scene information and solve the problem of edge scenes.
In his view, the current automatic driving that has entered the deep-water area of mass production development is no longer facing research and development problems, but testing problems. Whoever can obtain a large amount of accurate structured information in the shortest possible time, at the lowest possible cost, will be the winner in the field of intelligent driving.
epilogue
The field of autonomous driving has always been a coveted but extremely difficult "hard bone" for players.
From the small wave in 2018 to the big wave in 2021, the entrants in the field of automatic driving have entered the product landing stage from the start-up stage of barbaric growth technology after years of precipitation.
Ten years of grinding a sword, the frost blade has not been tried. Commercialization will be the touchstone for new things to move from technology to industry. In 2022, China's autonomous driving players have come to a moment before the decisive battle. In order to grab the ticket to the decisive battle, this year's autonomous driving industry must be short and fierce.