In 2021, the autonomous driving circle is mixed with fire and ice.
Fire, from the introduction of a number of open policies, shortened the commercialization process of autonomous driving in the mainland; from the mass production of lidar on the car, providing the possibility of large-scale mass production of autonomous driving products; from more intensive investment and listing news, adding fuel to the sustainable development of autonomous driving.
Ice, starting from the frequent personnel changes of autonomous driving executives, bringing a hint of haze to the development of the industry; from the acquisition/closure of a number of autonomous driving companies, the intensification of competition in the industry; from the public's doubts about the maturity of autonomous driving technology after a number of safety accidents.
In any case, however, we are finally pleased to see that the autonomous driving industry is growing much faster than before, encouraging more aspirants to join.
There are a thousand Hamlets in the eyes of a thousand people, and for the wonderful autonomous driving 2021, we hope to interpret it in another way: from the perspective of autonomous driving companies close to the industry, from the perspective of different people, we can observe the past year, and at the same time, we can extend it and look forward to 2022, which has been prelude.
To this end, New Intelligent Driving invited 16 executives from 15 companies in the field of autonomous driving to share the insights of front-line personnel in the industry.
Zhang Kai, Chairman of Zhixing
Gu Weihao, CEO of Mimo Zhixing
For the development of the automatic driving industry in 2022, Zhang Kai gave a ten-point prediction:
First, in terms of data model, data intelligence will become the positive or negative hand of automatic driving mass production.
Second, in terms of algorithms, Transformer and CNN are deeply integrated, which will become the glue for the integration of autonomous driving algorithms.
Third, in terms of AI technology development, the large computing platform will be officially mass-produced in 2022, providing support for the landing of Transformer technology and ONESTAGE CNN technology.
Fourth, in terms of mass production landing, 2022 will be the first year of mass production of urban smart pilot auxiliary driving.
Fifth, in the field of unmanned logistics at the end, its commercialization will show continuity.
Sixth, in terms of innovative products, Robotruck will officially open the exploration of the road of mass production.
Seventh, in terms of market development trends, the intelligent car market based on intelligent driving technology will gather at the head of the automobile industry, and the head effect is very obvious.
Eighth, in terms of hardware, with the improvement of the scale of mass production, the AI perception technology composed of lidar and machine vision will achieve deep integration of lidar applications with large computing platforms, which greatly enhances the accuracy and certainty of automatic driving perception capabilities.
Ninth, safety will remain a top priority for autonomous driving systems.
Tenth, in terms of talent flow, 2022 will be the most intense year for AI autonomous driving talents.
Gu Weihao believes that the most important action of The Millima Wisdom Walk in the past year is to release the data intelligence system MANA, which will be the core driving force for the evolution of the Mo Mo Zhi Xing ability and the key to the commercialization of the closed loop.
In addition, Chi Heng also made achievements in products and capital in 2021.
In terms of products, the auxiliary automatic driving system of Millima Zhixing has been successively installed in tens of thousands of vehicles of 5 models of Great Wall Motor, and its terminal unmanned logistics vehicles have been off the production line of 1,000 vehicles, and have reached cooperation with Meituan and Ali Damo Academy; in terms of capital, Millima Zhixing completed the Pre-A round of financing of hundreds of millions of yuan in early 2021, and completed the A round of financing of nearly 1 billion yuan at the end of the year.
Chairman of CIIC Wang Jin
At the beginning of the establishment of CIIC, it decided to all in the car road coordination. In the past year, the most important progress of CIIC has been to identify vehicle-road synergy as the company's goal for the next five years. The core of this is to focus on China, standardization, scale, and promote the dual promotion of "car-aware road" and "car that understands the road".
Last year, CIIC established a R&D system and talent structure based on vehicle-road collaboration at the end of the car and the end of the road. More importantly, CIIC, China Telecom and Suzhou Joint Venture established "Tianyi Transportation Technology Co., Ltd.", which has become one of the important standardized road end promotion links.
For the automatic driving industry, Wang Jin also pays attention to some phenomena.
He believes that data value and data security will be more recognized and fully valued in 2021. Today's data accumulation and application, as well as data security issues, may play a decisive role in the technical route, capital replenishment direction and resource allocation in the coming years.
In addition, completely independent and controllable intellectual property rights and systems will also become the consensus of the industry in 2021.
Wang Jin believes that the traffic revolution is one of the greatest changes of this era, and autonomous driving will definitely change the way humans travel. In 2022, the autonomous driving industry will continue to develop at a very fast pace, and will even become the industry where China's best talents in various fields gather.
Chairman of Freitek Zhang Lin
Thanks to the intelligence of the car, Zhang Lin obviously feels that the number of new vehicles and penetration rate of assisted driving are in a continuous upward channel, and more and more consumers are beginning to use various driving assistance functions carried by vehicles more frequently. At present, industry data shows that the market penetration rate of new passenger cars for L2 driver assistance has reached 20% and is growing.
In Zhang Lin's view, behind this trend there is a more noteworthy phenomenon - the deep integration of automotive software and hardware has brought fundamental changes to the entire automotive technology architecture, and with the application of the SOA architecture driven by atomic services, domain controllers have gradually developed as software carriers in the direction of high-end autonomous driving computing platforms.
In this case, integrating software and algorithms into the car and achieving mass production requires a great test of hard verification in engineering research and development, software integration, testing, supply chain management and other aspects for autonomous driving enterprises. Moreover, in view of the demand for large-scale mass production of high-end automatic driving in the future, the integration of technologies and products such as highways, urban roads, and automatic parking in typical scenarios not only requires in-depth scene mining and continuous function iteration and optimization, but also inseparable from the sincere cooperation between scientists, engineers, and partners.
In the past year, Forretech has continued to focus on the commercialization of ADAS and technical iteration, achieved the sustained large-scale mass production growth of ADAS and launched high-end autonomous driving solutions with domain control as the carrier, which are the two important forces that promote the continuous advancement of Fredtek, and are also the core of The current strategy of Fredtech. Focusing on the established strategy, we will fully invest in the transformation of software-defined vehicles with complete solutions for platform-based R&D, engineering services and partner ecosystems.
Looking forward to 2022, Zhang Lin believes that the trend of transformation and upgrading in the direction of software leadership and relying on "product + service" to win will become more and more obvious. At the same time, the boundaries of the automotive industry will continue to expand, cross-border integration will become the norm, and cooperation between technology companies, system solution providers and OEMs will be closer.
Zhang Tianlei, CEO of Mainline Technology
In 2021, the most important action of mainline technology is to carry out the normalization test of L4 self-driving trucks on the open highway as the first batch of enterprises.
This means that: on the one hand, based on high-speed open road testing, Mainline Technology can accelerate the iteration of autonomous driving algorithms through more real and rich test data, thereby promoting mass production. On the other hand, based on 6 open road test sections of highways with a total mileage of 143 kilometers, Mainline Technology will soon carry out demonstration operations of autonomous freight business with partners.
If the keywords are selected for 2021, Mainline Technology believes that "return" is the most appropriate.
Capital returns to value investment, enabling more investors in logistics, automobiles and other industries to lay out the autonomous driving track; enterprises return to technology and products, and truly develop and lay out from the perspective of creating value for customers. For example, Mainline Technology has deployed more than 100 driverless collection cards in Tianjin Port and Ningbo Zhoushan Port, and has the ability to quickly promote and apply.
For 2022, Zhang Tianlei believes that the industry will accelerate integration. With the release of more favorable policies and the leap from the research and development of autonomous driving technology to mass production of products, the commercialization of autonomous driving will continue to accelerate. Even in the next few years, the mass production of autonomous driving will remain the main theme of the same.
YuanRong Qixing CEO Zhou Guang
Yuanrong is perhaps one of the most optimistic companies about the prospects for commercialization of autonomous driving today. Zhou Guang believes that L4 self-driving cars are about to enter the market.
He expects that Yuanrong Qixing will cooperate with OEMs to develop a pre-installation mass production plan for vehicle specifications from 2022 to 2023; in 2024, L4 self-driving cars will be mass-produced and enter the market; in 2026, autonomous vehicles will form a large-scale application.
This judgment may be based on its $300 million strategic financing from Alibaba in 2021 and the launch of autonomous driving solutions that cost less than $10,000.
Alibaba's strategic financing provides more funds and resources for Yuanrong's "123N" strategy – that is, to build an autonomous driving system, focus on the two major businesses of travel and same-city freight, and produce autonomous driving products that meet the three requirements of safer, more reliable and smarter. The launch of the L4 level autonomous driving solution, which costs less than $10,000, will reduce the investment in the test operation fleet and will also provide it with the opportunity to mass-produce L4 autonomous driving.
For the development of autonomous driving in 2022, Zhou Guang gave three major predictions:
In terms of industry, 2022 will be a year of closer cooperation within the industry, especially the cooperation between car manufacturers and autonomous driving companies, and the cooperation between travel companies and autonomous driving companies.
In terms of technology, more and more companies will use solid-state lidar to achieve L4 autonomous driving and reduce the cost of autonomous driving solutions.
In terms of policy, all localities will also become more and more open, and it is expected to formulate corresponding management systems for all-unmanned autonomous driving tests and automatic driving charges.
MAXIEYE CEO Zhou Shengyan
In the era and industrial background of the epidemic and the "lack of core", Zhou Shengyan obviously felt that the development opportunities of independent brand technology and products have increased, and the development momentum is stronger. One of the phenomena is that domestic suppliers are gradually cooperating with independent brand companies to explore new ecological models.
On the other hand, the industry continues to call for the landing of technology scale business, making the capital market pay more attention to the "mass production landing" and the core internal driving force and growth force of the enterprise.
In 2021, scale is the key word in the autonomous driving industry, which is intuitively reflected in the rapid increase in the penetration rate of the domestic market of ADAS, and the number of new vehicles and the expected configuration scale is increasing. In addition, the establishment and improvement of the independent technical capabilities of the whole industry chain, the diversified transformation of cooperation models, and the gradual downward cost are complementary to the trend of scale.
For the development of the industry in 2022, he gave five predictions:
First, as more L2 and above ADAS systems are mass-produced, and the rate of intelligent driving mounting continues to increase, more problems may be exposed.
Second, the industry will shift from the pursuit of product landing to the pursuit of multi-dimensional product experience.
Third, effective scene data will be a key factor in bridging the performance gap for next-generation products.
Fourth, the trend of central all-in-one computing will be more pronounced.
Fifth, the industrial chain will jointly define products in the roundtable cooperation model.
MINIEYE CEO Liu Guoqing
Liu Guoqing deeply felt that in 2021, there will be "localization and substitution" in the automatic driving industry.
In 2021, due to the strength of autonomous driving companies, new and old car manufacturing strength, and the multi-faceted layout of cross-border players in various lines, the industry has entered a highly competitive state. During the period, the policy called for localization and the global "lack of core tide" for a long time, and domestic manufacturers continued to rise. He expects that in the next three years, more than half of the share of the intelligent driving field will shift from foreign suppliers to domestic suppliers.
However, he also believes that mass production delivery will be a major test for the autonomous driving industry in 2022.
The higher the level of autonomous driving, the longer the development process and delivery cycle, the more chains involved. On the track where domestic and foreign players are competing to catch up, domestic suppliers will compete not only in core technology, but also in how to link up with the advantages of upstream and downstream of the industrial chain and ecological partners, and achieve large-scale and stable delivery of more high-quality products and services through a stable supply chain system and a complete strategic ecological closed loop.
At present, MINIEYE has formed a product matrix including L0-L2 class ADAS products, L2+ and L2++ level intelligent pilot auxiliary products, intelligent cockpit perception and interaction products iCabin, which can provide mature and mass-produced intelligent car full-stack solutions for different customers.
Rong Li, General Manager of Zhijia Technology China
In 2021, autonomous driving will develop rapidly due to favorable factors such as policies, capital, and industrial chain hardware, and new players will become more and more hot. While the industry has received unprecedented attention, the enterprise advantages of the first release bureau have gradually emerged.
After several years of development, some trends have reached an industry consensus.
First, the industry is generally more optimistic about autonomous commercial vehicles, especially trunk logistics trucks. If 2021 is the first year of mass production, then in 2022, autonomous driving companies will pay more attention to the mass production of technology. Commercial vehicles represented by trunk logistics will be optimistic about the industry due to the advantages of semi-closed landing scenes and large market scale.
Second, some of the driverless companies that once insisted on the leapfrog route began to take the incremental route.
Unmanned driving is the ultimate goal of the development of autonomous driving, and L4 autonomous driving technology is the technical basis for achieving this goal. In the development process of the autonomous driving industry, some enterprises can achieve mass production through technology dimensionality reduction, which can not only generate economic value and social value, but also use data closed-loop to promote the optimization of automatic driving technology algorithms. At present, more companies have discovered the advantages of this route and embarked on the gradual route of "mass production products and L4 technology research and development" in parallel.
Rongli believes that the effectiveness of the data closed loop will be revealed in 2022, and some enterprises that laid out the data closed loop earlier will usher in harvest in 2022, and the front-loading mass production and delivery of intelligent vehicles equipped with automatic driving systems have also become the key differentiating advantages of the leading industry for leading enterprises.
Therefore, the practical application with cost reduction and efficiency enhancement function (different from the trial operation that now takes testing as the main form and does not consider the cost scheme) will be realized by a small number of head enterprises. For example, according to the actual operation data, Zhijia Technology and Zhitu Technology help FAW Jiefang's front-loading mass-produced smart heavy trucks, and the Zhijia PlusDrive automatic driving system equipped with it can save more than 10% of fuel consumption.
Zhou Xin, CPO of Yishi Technology
Commercialization runs through the autonomous driving industry in 20221.
In general, each autonomous driving enterprise can be divided into four different exploration paths in terms of commercial exploration:
First, many companies that originally focused on the Robotaxi track have begun to get involved in the autonomous driving segmentation scenario and make up for the single market strategy by laying out more scenarios.
Second, autonomous driving companies such as Yishi Technology, which laid out "all-scenario" at the beginning of the company's establishment, began to implement autonomous driving technology in multiple scenarios in 2021 to accelerate the commercialization process.
Third, affected by the epidemic, the demand for autonomous driving in subdivision scenarios such as factories and airports has become more obvious, and the application boundaries of autonomous driving in subdivision scenarios have been continuously broadened. For example, in cooperation with Hong Kong International Airport, Yishi Technology has landed unmanned logistics vehicles and unmanned patrol vehicles, and unmanned buses at the airport are about to be put into operation.
Fourth, the cooperation between autonomous driving companies and car companies has been deepening, and the To G model has continued to land in various places.
Zhou Xin believes that commercialization will become the main trend of the autonomous driving industry in 2022, and enterprises will explore more scenarios and models.
It should be noted that as the commercialization process continues to accelerate, autonomous driving technology will need to be responsible for the daily operation of enterprises, and the commercial understanding of different market segments and the ability to deeply adapt to upstream and downstream customers will also face challenges, which requires more than one million kilometers of commercial operation experience as a support. In the future, autonomous driving companies with rich operational experience and commercial mileage accumulation will have greater and greater advantages in subdivision scenarios.
Momenta CFO Zhang Peng
Zhang Peng believes that the autonomous driving industry and Momenta have experienced a "thick accumulation" of 2021.
From the perspective of the entire industry, the automobile industry is experiencing major changes unprecedented in a hundred years, and the future will be a period of rapid development of automobile intelligence, which is inseparable from the continuous accumulation of related technologies in previous years. From the company's point of view, Momenta has been building a "flywheel" and "two legs" since its establishment in 2016, and the mass-produced self-driving Mpilot will be installed in cooperative models this year for more end users; robotaxi products based on fully unmanned MSD will be launched in more cities. The development of the two legs is also closely related to industry trends.
In the past year, all parties have attached great importance to intelligent vehicles, which has opened the prelude to automobile intelligence. Now that the process of electrification of the automotive industry is in the ascendant, after 2022, autonomous driving will be the first big opportunity in the next era.
He also said that the second half of autonomous driving is different from the 0-1 of the first half, the development of the industry will tend to 1-N, and the biggest test of this process will be the ability to scale.
Pony Zhixing COO Wang Haojun
Fast is the deepest feeling of Xiaoma Zhixing for the autonomous driving industry in 2021, and it is also its summary of its own development in the past year.
Wang Haojun believes that the rapid development of the automatic driving industry is first reflected in the more mature upstream and downstream industrial chains, the significant reduction of costs, and the increasingly rich choice of hardware, thus opening the process of productization of autonomous driving technology, and the technology is moving more and more towards vehicle regulation and mass production.
In addition, policies and regulations have taken a key turn around the application and commercialization of the land, and a multi-level policy and regulation system has been formed from the state to the local level. More importantly, the demand and acceptance of autonomous driving technology by the public and the automobile market has also increased significantly, and autonomous driving has begun to truly enter our lives and serve people's travel.
The above factors have jointly promoted the rapid development of the industry, of which the opening of policies is particularly prominent in 2021.
With the support of the opening policy, Xiaoma Zhixing has opened unmanned tests of autonomous driving in Guangzhou and Beijing in 2021, and opened the pilot of the commercialization of autonomous driving travel in Beijing in November, and Robotaxi's business has officially entered the commercialization mode verification stage from the technical normalization test.
Yu Qian, CEO of Light Boat Zhihang
In 2021, the software and hardware technology of automatic driving will continue to upgrade, improving the safety awareness of automatic driving, while a number of favorable policies have been introduced one after another, and the number of investment and financing events has increased, promoting the large-scale application of automatic driving.
Yu Qian believes that before the large-scale application of autonomous vehicles, it is first necessary to achieve a safety level that is 1-2 orders of magnitude higher than that of manual driving, because safety is the eternal premise of automatic driving.
In addition, Yu Qian also mentioned his views on bicycle intelligence and vehicle-road coordination.
He believes that vehicle-road collaboration needs to invest huge costs to transform infrastructure and technology iteration, and the completion of this process is long, and it may be more than ten years before vehicle-to-road collaboration can have an impact on the demand for bicycle intelligence. Therefore, at this stage, the vehicle-road collaboration can be regarded as the safety redundancy of bicycle intelligence.
In addition, Light Boat Zhihang has never left the field of Robotaxi, and at present, Robobus and Robotaxi apply the same set of technical solutions. It has also previously launched a 5G self-driving ride-hailing bus, breaking the boundaries between Robobus and Robotaxi.
Yiqing Innovation Chairman Liu Ming
In 2021, the autonomous driving industry is thriving and blossoming everywhere. Liu Ming believes that this degree benefits from the mass production of lidar and the optimization of automatic driving technology, and the reduction of vehicle body costs accelerates the rapid development of the automatic driving industry.
In his view, the commercialization of unmanned vehicles cannot be transformed by existing models, and requires positive design, and now the arrival of the "new four modernizations" provides a possibility for car transformation to adapt to driverless technology models. Therefore, Liu Ming believes that the premise of the commercial realization of unmanned vehicles needs to be based on electric vehicles and combine car manufacturing experience and technical advantages when the cost of lidar is low.
For the development of the autonomous driving industry in 2022, Liu Ming judged that the high-frequency, low-cost, functional low-speed unmanned driving scenario will take the lead in ushering in the outbreak, and the development of the industry will also focus on product functions to serve the society.
This judgment may be related to yiqing innovation's achievements last year. In 2021, Yiqing's innovative unmanned system will be applied on a large scale in port terminals and hundreds of parks, and the "Kwafu" series of unmanned vehicles will be landed in eight scenarios, and more than 100,000 kilometers of new safe operation will be added.
Dong Fangliang, Executive Director of Wenyuan Zhixing
Helong lianheng is the keyword of the 2021 autonomous driving industry identified by Dong Fangliang of Wenyuan Zhixing.
Since Wenyuan Zhixing first realized the automatic driving iron triangle model in 2018, by 2021, there will be more and more cooperation cases in the industry, that is, the three-way combination of autonomous driving companies, car companies, and travel operation companies, including many strategic in-depth cooperation, and enterprises with win-win cooperation thinking and upstream and downstream resources will be more favored by the industry, so as to achieve a win-win situation for each side and a win-win situation for all parties.
In his view, with the landing and promotion of this multi-party cooperation model, "full unmanned driving" has changed from a technical term to a living tool, which highlights the power of scientific and technological innovation while practicing the social responsibility of scientific and technological enterprises.
Unmanned "anti-epidemic" is a realistic example. For example, during the outbreak of the epidemic in Guangzhou, Wenyuan Zhixing participated in the task of delivering daily life and emergency materials to residents in the sealing and control area, with a total of more than 500 departures for 20 consecutive days, with a total transport weight of more than 100 tons.
In Dong Fangliang's view, in 2022, scale and unmanned driving will still be the direction of the industry's efforts to develop, and then it will promote the faster arrival of driverless passenger operations.
Last year, Wenyuan Zhixing has achieved three major product layouts such as Robotaxi, Robobus and Robovan last year. Including Wenyuan Zhixing, Volkswagen will see more and more different driverless cars driving on the streets of cities in 2022, covering transportation and urban governance fields such as travel and freight transport.
Leifeng networkLeifeng networkLeifeng network