From the show arena to the competition arena, on the eve of mass production of autonomous driving, Juefei Technology's "data engine" seizes the commanding heights of positioning and decision-making

Although we still have many problems to overcome in terms of technology and regulations, autonomous driving will become the mainstream trend of smart transportation. With the emergence of a large number of new roles of "technology suppliers" in the traditional industry chain, the continuous iteration and upgrading of technology and demand will greatly promote the construction and maturity of the new ecology of autonomous driving technology. The golden age of autonomous driving cars is coming to us in great strides.

1. On the eve of mass production, the wise will have different opinions

In recent years, autonomous driving has become one of the hottest trends in the automotive field. Despite the capital winter and the downturn in the auto market, autonomous vehicles still accounted for the largest share of global AI investment in the past year. The recently released "Stanford 2019 Global AI Report" shows that global private AI investment exceeded US$70 billion in 2019, and driverless car technology startups received US$7.7 billion in funding, accounting for 9.9% of the total private AI investment.

As the most favored track in the field of artificial intelligence, autonomous driving is becoming a hot spot for capital from all sides due to its vast space. From Waymo's announcement at the beginning of the year to build a factory to mass produce L4, to the autonomous driving alliance between Mercedes-Benz and BMW, from Didi's training of autonomous driving through online car-hailing data, to Meituan's development of unmanned delivery vehicles...all signs indicate that as technologies such as the Internet of Things, big data, and AI are accelerated in application scenarios such as smart logistics, it is expected that capital and start-up companies will further increase their investment in autonomous driving, delivery technology, and solutions. Under the strong promotion and guidance of capital to the survival of the fittest in the market, autonomous driving technology continues to innovate, and the industry will also undergo new structural upgrades and reorganizations. Landing has become a goal that the industry unanimously pursues.

If 2019 is taken as the dividing line, the industry was a "show field" for technology before, and will become an "arena" for practical application after that. Industry experts believe that within three to five years, the path of integrating algorithms, software, data, and stand-alone computing power will become the mainstream trend for the commercialization of autonomous driving.

2. The scenery here is unique

Since Google launched its driverless car project, humans have been on the road of autonomous driving for a full decade. At present, the development of autonomous driving technology has been divided into two camps: the ADAS and single-vehicle intelligent technology camp represented by automobile manufacturers, and the artificial intelligence and networking technology camp represented by Internet companies.

American companies, represented by autonomous driving giant Waymo and travel giant Uber, are based on single-vehicle intelligence and aim at L4-L5 driverless cars that are completely oriented to future travel. They have made vehicle-side intelligence the core of technological breakthroughs through comprehensive innovations in perception, decision-making, and control. However, this disruptive change is not only too costly, but also has difficulties in actual mass production due to the large gap with the traditional automotive industry.

Connected intelligence emphasizes the gradual empowerment of vehicles with intelligence without major changes to the vehicle's electronic architecture, bringing L2-L3 assisted driving to existing mass-produced vehicles. Compared with single-vehicle intelligence, this approach is easier to implement because of its controllable costs and is also supported by the industry chain's technology.

In the past 2019, intelligent connected cars seemed to be in full swing. According to professional organizations, the scale of China's intelligent connected car market will reach more than 100 billion yuan in 2020.

However, regardless of whether it is a "disruptive" or "incremental" development path, the underlying structure of the automotive industry has not undergone any qualitative changes. This has resulted in the existing development of autonomous driving still being subject to a series of restrictions, and new ideas and new methods are needed to improve it.

At the just-concluded 6th Annual Forum of China Electric Vehicle 100 Forum, Li Keqiang, professor of the School of Vehicle and Transportation Engineering of Tsinghua University and director of the State Key Laboratory of Automotive Safety and Energy Conservation, said that the future development is neither single-vehicle intelligence nor completely dependent on networked intelligence, but the integration of the two, that is, the "vehicle-road-cloud" integrated system, which will become the future development direction. In January 2020, at the 6th Annual Forum of China Electric Vehicle 100 Forum, Miao Wei, Minister of Industry and Information Technology, also reiterated that "In China, we are determined to promote 5G LTE-V2X in road reconstruction."

The upcoming 5G will provide new impetus in this direction. 5G networks have the characteristics of high transmission rate, low latency, and high reliability, and are regarded as the perfect match for Internet of Vehicles and autonomous driving.

At the same time, the integration of intelligence and networking will inevitably require local attributes. The transportation infrastructure, information infrastructure and human transportation habits of different countries are all different. Due to the rapid development of 5G in China, China's new generation of autonomous driving must be an integrated Chinese solution.

As a major transportation country in the world, China is in the historical transition stage from a "transportation power" to a "transportation powerhouse". From a national strategic perspective, the government's policies and support for intelligent driving are quite strong, far exceeding other countries in the world.

In this way, China has two prerequisites for the development of autonomous driving: first, government support; second, the rapid construction of 5G base stations and technological leadership. It is well known that the implementation and commercialization of 5G in the Chinese market will be far ahead of other countries. It is expected that by 2020, all cities above the prefecture level in the country will provide 5G commercial services.

These two unique advantages will become an important weight for China's smart car industry to occupy the commanding heights. In fact, China's autonomous driving in the 5G environment has already started.

On November 15, 2019, China Mobile held the "5G+Unlimited Future Possibilities" Global Partner Conference. At the same time, at this conference, the "5G Precision Positioning Alliance" led by China Mobile was formally established. Juefei Technology, as a representative of the new force of autonomous driving technology and a partner of China Mobile in the field of high-precision positioning, was invited to join the alliance and become a member of the board of directors of China Mobile's 5G Precision Positioning Alliance.

The picture shows China Mobile's 5G+ Beidou high-precision positioning application white paper

It is reported that Juefei Technology is an AI technology innovation company focusing on the autonomous driving industry, dedicated to building the "data engine" of autonomous driving. Through high-precision data services and high-precision positioning services, it provides the industry with reliable, accurate and safe intelligent driving solutions. In this cooperation with China Mobile, Juefei Technology will integrate the technology and application of 5G differential positioning into China Mobile's high-precision positioning service in the autonomous driving industry. This service will have the advantages of full-scene, real-time and high-precision, and will become a benchmark demonstration of 5G precise positioning, and will also become an important part of China Mobile's 5G strategy.

  Juefei Technology cooperates with China Mobile to demonstrate 5G high-precision positioning

Regarding the future development of the autonomous driving industry, Juefei Technology believes that within a certain period of time, 5G will become an important weight for China's smart car industry to occupy the commanding heights. With the rapid implementation and commercialization of 5G in the Chinese market, the pressure on the computing power of the vehicle-side processor will be greatly reduced, so that the cost of intelligentization on the vehicle side will be greatly reduced. At the same time, because of China's leading position in 5G technology and strong support from government policies, it leads the world in promoting the commercialization of 5G. This is of great significance to promoting the upgrading of China's autonomous driving industry.

3. Look up at the stars but keep your feet on the ground

Prediction and the future, implementation and blood-making are not contradictory. In the process of building the "data engine" of intelligent driving, Juefei Technology has always adhered to the technical route of "strong data and heavy integration".

High-precision maps are the foundation and perception container of intelligent driving. Juefei Technology has the ability to build its own high-precision, high-quality, and high-freshness map data service platform. Its high-precision maps use the industry-leading binocular vision ranging technology and AI deep learning image processing technology to provide sub-meter-level accuracy and weekly-updated road element data. At the same time, it uses the self-collected rasterization + crowd-source update technical solution to ensure accuracy while achieving efficient updates through data alliances.

At the same time, Juefei Technology has taken an innovative approach in the application of high-precision maps. Juefei Technology has worked with a Tier 1 manufacturer to provide services for heavy trucks, applying map data to the cruise control system. By acquiring and predicting the speed limit, slope, curvature, and heading of the current road in real time, the ECU will give precise throttle control instructions after local calculation, thus achieving fuel and energy saving in mountainous freight driving.

In the direction of high-precision positioning, Juefei Technology's core technology "Super Sense" has become the best practice in industrial decision-making applications. "Super Sense" uses the camera + GPS + IMU solution, based on visual road feature recognition and positioning technology, combined with edge computing, to achieve a low-cost, sub-meter-level positioning solution, perform current lane recognition, sign positioning, beyond-visual-range lane change recognition, and complete ADAS Map semantic data prior collection and lane-level positioning services. At the same time, it can be provided to commercial vehicle/passenger vehicle customers in the form of software and services. During vehicle driving, it provides users with high-precision road geometry data such as road slope and curvature, rich traffic sign information, and gives drivers strong reminders and vehicle decision-making basis, truly allowing autonomous driving to achieve the optimal path decision from point A to point B, reducing the cost of hardware sensors, integrating and making decisions with the real world and the virtual world, and enabling autonomous driving vehicles to have stronger perception and planning capabilities from a global perspective.

  Juefei Technology's "Super Sense": Following Cars in Low-Speed ​​Traffic Jams

  Juefei Technology's "Super Sense": Driving and merging under viaducts

  Juefei Technology's "Super Sense": Dealing with Chinese-style queue cutting

At the same time, Juefei Technology pays more attention to technological innovation. At present, it has reached a cooperation with Yutong Bus, the leader of China's commercial vehicle industry. The two parties will conduct joint research and development and strong cooperation in autonomous driving perception fusion technology. It is reported that the AR interactive system provided by Juefei Technology for Yutong autonomous driving buses will be put into mass production in the near future.

Facing the arrival of V2X, Juefei Technology has made strict arrangements in the direction of high-precision positioning and integrated adaptation, further promoting the "data engine" of intelligent driving into the field of smart transportation and travel.

4. The Golden Age Has Arrived

Faced with the once-in-a-century transformation of the automotive industry, a battle for the autonomous driving industry is underway among Chinese cities. As of June 2019, 18 cities in my country have issued autonomous driving road test license plates, totaling nearly 200, and a total of 50 companies have obtained the license plates.

On September 22, 2019, the National Intelligent Connected Vehicle (Wuhan) Test Demonstration Zone was officially unveiled, and the Wuhan Transportation Bureau issued the world's first commercial license for autonomous driving to companies such as Baidu. This means that related companies can conduct manned tests on public roads and conduct commercial operations, thereby promoting the application of intelligent driving technology in real scenarios.

At the same time, from the national policy level, the importance of autonomous driving has been raised again. On September 19, 2019, the CPC Central Committee and the State Council issued the Outline for Building a Strong Transportation Country, which emphasized the need to strengthen the research and development of intelligent networked vehicles (smart cars, autonomous driving, and vehicle-road collaboration) to form an independent and controllable complete industrial chain.

On the one hand, there is strong support from governments at all levels, and on the other hand, companies are investing a lot of manpower and material resources to step up research and development. It seems that it is only a matter of time before we can have fully autonomous driving. However, autonomous vehicles are ultimately a means of transportation for people. In the process of promoting the industrialization of autonomous vehicles, they have to deal with the procurement of parts and components, as well as the development of the entire vehicle production chain, which requires a mature industrial environment. In addition, as an emerging industry, autonomous vehicles also face social challenges such as law, safety, and human ethics. This tests the patience and wisdom of the government and consumers.

Although we still have many problems to overcome in terms of technology and regulations, autonomous driving will become the mainstream trend of smart transportation and this trend will not change. With the emergence of a large number of new "technology suppliers" such as Juefei Technology in the traditional industry chain, it will greatly promote the construction and maturity of the new ecosystem of autonomous driving technology.

The golden age of self-driving cars is approaching us in great strides.

As a winner of Toutiao's Qingyun Plan and Baijiahao's Bai+ Plan, the 2019 Baidu Digital Author of the Year, the Baijiahao's Most Popular Author in the Technology Field, the 2019 Sogou Technology and Culture Author, and the 2021 Baijiahao Quarterly Influential Creator, he has won many awards, including the 2013 Sohu Best Industry Media Person, the 2015 China New Media Entrepreneurship Competition Beijing Third Place, the 2015 Guangmang Experience Award, the 2015 China New Media Entrepreneurship Competition Finals Third Place, and the 2018 Baidu Dynamic Annual Powerful Celebrity.