Three perspectives, 16 voices, looking at the future trend of intelligent driving, the first list of guests announced | 2019 Global Autonomous Driving Summit

Autonomous driving has evolved from a scientific research issue to an engineering issue.

In the 0 to 1 stage, autonomous driving needs to solve the problem of creating something from nothing; in the 1-10000 stage, autonomous driving needs to face the problem of how to solve the scale problem, and there are further requirements for the scalability of the solution and the stability of the system. The industry is focusing on engineering issues, and autonomous driving companies are about to usher in a protracted battle of "mass production and delivery". How to bid farewell to the sample vehicles produced in the "handmade workshop" style and form a large-scale, standardized, and modular production process is a problem that needs to be solved on the eve of mass production.

In addition, autonomous driving is a very complex systematic project, which involves multiple issues such as the software and hardware level of the car, road facilities, laws and regulations, urban planning, etc. New variables are also constantly emerging: the upcoming 5G will enable the further development of autonomous driving with its advantages of low latency and high reliability, and the arrival of vehicle-road collaboration will also accelerate the implementation of autonomous driving.

At this point, Leiphone.com's New Intelligent Driving, a vertical media in the field of intelligent driving, will hold the "2019 Global Intelligent Driving Summit" in Xiangcheng, Suzhou on October 26 and October 27. It will be guided by the "Jiangsu Provincial Department of Industry and Information Technology" and the "Suzhou Municipal People's Government", hosted by the "Suzhou Xiangcheng District People's Government" , and undertaken by Leiphone.com New Intelligent Driving.

For more conference content, click Read original text for details.

This article contains unique insights from confirmed guest speakers at the summit, allowing more people to understand the smart driving challenges and solutions faced by leaders.

5G communications and vehicle-road collaboration are becoming an indispensable part of smart transportation.

Vehicle-road collaboration can significantly reduce the sensor cost of single-vehicle intelligence; provide a comprehensive perspective and improve single-vehicle safety; and shorten the time it takes for autonomous driving to be put on the road.

Professor Mao Guoqiang is developing a smart beacon that can be deployed in large numbers on highways and their surrounding areas at a low cost and integrated with multiple sensors. It can combine the deployed lidar, cameras and other facilities to achieve multi-source heterogeneous data fusion, and realize multiple important functions of smart highways such as auxiliary vehicle precise positioning, pedestrian collision warning, road condition warning and road health status monitoring, thereby improving the informationization and intelligence level of the transportation system to improve safety and efficiency, and provide basic road perception and communication infrastructure support for intelligent connected vehicles in the future, so that vehicle-road collaboration can achieve beyond-line-of-sight perception and accurate and reliable perception and positioning in complex environments.

Moreover, transforming roads into smart roads will make them safer and more efficient in the short term, while providing fine-grained real-time traffic information for traffic management; in the long run, it will provide much-needed road infrastructure support for the booming revolution in the future. On October 26, Professor Mao Guoqiang, Professor of Wireless Networks at the University of Technology Sydney, Senior Principal Researcher at the Australian Federal Government Scientific and Industrial Research Institute, and IEEE Fellow, will share his latest thoughts.

Liu Ming, director of the Intelligent Autonomous Driving Center of the Hong Kong University of Science and Technology, believes that what autonomous driving players have to surpass is the height of Mount Everest. The key point in this process is which unmanned driving path is the path that can actually reach the summit.

He believes that from the perspective of solving practical problems, he thinks that autonomous driving freight logistics may be faster. However, autonomous driving has passed the stage of watching demos, so engineering is a problem that lies ahead of every autonomous driving company. Most autonomous driving companies play the role of technology solution providers or algorithm companies, and do not have the ability to manufacture vehicles themselves. Therefore, in order to achieve mass production and delivery of autonomous driving vehicles, in-depth cooperation with the OEM is required. On October 27, Liu Ming, director of the Autonomous Driving Center of the Hong Kong University of Science and Technology, will bring more insights into autonomous driving.

Automobiles have entered a new era of intelligence. The era of software-defined cars is coming. AI, ecological construction, and service capabilities are important manifestations of competitiveness. Car manufacturers and Internet companies are scrambling to seize the entrance to intelligent networking.

In the era of smart cars, BYD has launched the DiLink smart network system, taking a differentiated route. On October 27, Yang Dongsheng, director of BYD Auto Industry Co., Ltd.'s Product Planning and Automotive New Technology Research Institute, will bring BYD's latest thinking on smart cars.

The purpose of high-precision maps is to improve the intelligence level of vehicles to enhance their performance and safety.

High-precision maps are not the GPS navigation we are familiar with. They are new maps loaded with geocoding metadata and designed specifically for autonomous driving.

The most important purpose of high-precision maps is to enhance the confidence of autonomous driving systems and solve the computational problem of road parsing. Existing high-precision maps can be divided into multiple levels, such as "precise lane markings, boundaries, geometry and 3D markings for positioning."

NavInfo is a leading representative company in high-precision maps and has a deep understanding and knowledge of high-precision maps. On October 27, Chen Dan, vice president of NavInfo, will share NavInfo's latest thoughts on high-precision maps.

In the process of implementing autonomous driving, scenarios are the most important factor. The data accumulated by Didi in online car-hailing operation experience can be used to train autonomous driving vehicles, making them safer, smarter, and more adaptable to various travel scenarios, with natural advantages in implementation. On October 27, Meng Xing, COO of Didi's autonomous driving subsidiary, will share Didi's latest thoughts on autonomous driving.

In the wave of autonomous driving, Visteon is changing rapidly. Previously, Visteon proposed the "Lego" concept of autonomous driving and provided flexible ideas for the development of autonomous driving.

Visteon DriveCore autonomous driving platform provides autonomous driving solutions above Level 3, and has significant advantages in terms of mass production development and application prospects. In the future, cars will shift from function-oriented to user experience-oriented. After 2024, the application of AI technology in smart cars will be more in-depth, specifically manifested in the application of neuron-shaped processing units at the hardware level, and AI will truly participate in the entire decision-making layer. On October 27, Visteon Chief Architect Wang Kai will bring Visteon's latest thinking on autonomous driving.

In the past 20 years, the Internet has essentially realized the "internet of people", and the opportunities for the Internet in the next 20 years will definitely be in the "internet of things". With the perception and connection of the surrounding environment, cars are no longer just a means of transportation, but have become ecological terminals, requiring more partners to jointly build a software and service ecosystem.

So following this trend, Tencent has built a complete technical system for autonomous driving R&D and product iteration. In terms of services, it leverages its cloud platform and simulation industry advantages to provide full-chain service capabilities and high-precision map commercialization capabilities. At the software level, it has made breakthroughs in core technologies and system integration, and accumulated data and technology, and provides software products to partners in a flexible, componentized, and pluggable manner.

Of course, if we want to achieve safe, comfortable, environmentally friendly and efficient autonomous driving at the same time, the vehicle-side intelligence alone cannot meet the requirements. We must realize the three-dimensional integration of cloud, field and end. We must abstract everything into the digital world and connect it to the Internet to realize the interconnection and interaction between things and between people. On October 27, Su Kuifeng, general manager of Tencent Autonomous Driving, will bring Tencent's latest thinking on autonomous driving.

On September 26, the trial operation of Baidu's self-driving taxi fleet Robotaxi was officially launched in Changsha. The first batch of 45 "Hongqi EV" Robotaxi fleets jointly developed by Apollo and FAW Hongqi began trial operation on some open test sections in Changsha.

Four years after its establishment, Apollo has continued to accelerate its implementation and expand its ecological alliance. Liu Shuo, general manager of ecological cooperation of Baidu Intelligent Driving Business Group, will bring Baidu's latest thinking on autonomous driving.

The wave of artificial intelligence has driven the entrepreneurial trend of AI chips, and among AI chips, some people say that autonomous driving chips are the Mount Everest of AI chips. The road ahead in the field of autonomous driving is long and arduous, and the factors affecting computing power requirements, vehicle regulations, and industrialization have made AI chip companies face more challenges.

NVIDIA is an important representative player among them. On October 27, Dong Fangliang, head of NVIDIA's autonomous driving China, will bring NVIDIA's thoughts and strategies on autonomous driving chips.

The implementation of autonomous driving is closely related to the scenario. Among all vertical scenarios, logistics is the industry with the most commercial value.

Inceptio Technology is a representative enterprise in the logistics industry. Since its establishment, it has quickly established its vehicle engineering and autonomous driving teams across China and the United States, completed research and development in three months, and released the Inceptio No. 1 prototype. Inceptio Technology is moving towards mass production, and has in-depth cooperation with vehicle manufacturers, Tier 1 and other industry chain companies, and has the ability to implement mass production. On October 27, Ma Zheren, founder and CEO of Inceptio Technology, will share Inceptio Technology's thoughts and strategies on autonomous driving chips.

Landing is the keyword of autonomous driving. WeRide has always focused on the Robotaxi track. In early August this year, WeRide, Guangzhou Baiyun Taxi Group and Science City (Guangzhou) Investment Group jointly established the smart travel company WeRide Guangdong.

Han Xu, co-founder and CEO of WeRide, believes that the concept of mass production is not a simple matter of quantity. On the premise of ensuring safety, sufficient technology and confidence to remove the safety officer from the car is the sign of mass production. Mass production requires in-depth cooperation with the OEM to solve problems such as system stability. On October 27, Han Xu, co-founder and CEO of WeRide, will bring the implementation of WeRide.

AI chips are the biggest topic in the artificial intelligence industry, but among AI chip companies, there are not many companies that have invested in the field of autonomous driving. Autonomous driving requires the highest computing power, which is a huge challenge for startups.

Horizon Robotics, the unicorn of artificial intelligence chips, recently released China's first automotive-grade edge AI chip, the Journey II, which was unveiled at the World Artificial Intelligence Conference and the Frankfurt Motor Show. On October 27, Zhang Yufeng, vice president of Horizon Robotics and general manager of the intelligent driving product line, will share Horizon Robotics' thoughts and strategies on autonomous driving chips.