In recent years, China's car networking industry has developed rapidly, key technological innovations have been accelerating, and competition in the field of autonomous driving is intensifying. However, the application of autonomous driving technology is still in the early stages of development, and there are still many problems to be solved. Wu Junjie, executive director of the Intelligent Driving Research Institute, said in an interview with the media recently that autopilot is not just a matter that a car needs to solve, but something that the entire industry chain and the whole society must face.
Auto-driving hot
Auto-driving cars from research and development to the road, people are gradually turning the sci-fi plots in the movie into real life scenes in human life. The State Council issued the "New Generation Artificial Intelligence Development Plan" last year, and proposed a new generation of artificial labor for China in 2030. The guiding ideology of intelligent development, strategic goals, key tasks and safeguards. Among them, intelligent networked vehicles with automatic driving technology as the core are regarded as one of the key applications of artificial intelligence in China.
According to the China Automotive Technology and Research Center, the relevant institutions predict that the number of China's intelligent networked car users will exceed 40 million in 2020, and the market scale will exceed 400 billion yuan. Under the background of the development of intelligent networked vehicles, various car companies And technology companies are making a lot of efforts in the field of autonomous driving. For example, Baidu launched the 'Apollo Plan' for open autopilot platforms. Google and ride sharing service provider Lyft recently announced that they will jointly develop new autopilot technology. Xiaomi also announced A strategic cooperation agreement with BAIC Group to create a smart driving system. In addition, Hezhong New Energy recently announced the launch of the 'Archimed Project' version 1.0, which covers product planning, vehicle technology, services and other aspects. It also plans for autonomous driving technology.
Wu Junjie said that intelligent networked vehicles are equipped with advanced in-vehicle sensors, controllers, actuators, etc., and incorporate modern communication and network technologies to realize intelligent information exchange and sharing between vehicles, people, vehicles, roads, and clouds. In the end, it will be a new generation of cars that will replace the people, and it will become a new outlet for the automotive industry. Wu Junjie was the senior director of automatic driving at Delphi (China) Technology R&D Center. He has been working on automotive safety electronics for 14 years. Product development, joined Hezhong New Energy in June this year, served as Executive Dean of the Intelligent Driving Research Institute.
Eureka 01 for automated driving demonstrations
It is understood that in the field of autonomous driving, Hezhong New Energy has also laid out a clear technical roadmap, from 'carrying drivers' to 'drivers' and finally becoming a reliable and credible driver in a specific situation. The exclusive driver'. Wu Junjie said, 'The driverless function of the PIVOT intelligent system mounted on the Eureka 01 concept car has already realized the low-speed automatic driving, the car automatically finds people and the independent valet parking, driving On the way to achieve obstacle avoidance and other functions. '
According to the intelligent driving technology plan released by Hezhong New Energy: The first step is to launch products with adaptive cruise, lane keeping assist and traffic congestion assistance function in 2019; the second step will be gradually introduced into unmanned parking in 2021. Traffic jam cruise driving, highway cruise assist function; the third step, in 2023, will continue to launch unmanned, highway cruising driving function. Through these three steps, to achieve the vehicle from the 'driver', 'generation Driving the driver's role in the 'exclusive driver'.
Wu Junjie said that the intelligent networked car with autopilot technology as the core is an important part of the future auto market and will become the new engine of the Chinese auto industry. According to China's "medium and long-term development plan for the automotive industry", by 2025, the height And fully self-driving cars began to enter the market.
Technology needs innovation, and more cooperation between industries is needed.
To achieve automatic driving, technical support is indispensable. At present, Chinese car companies are advancing early warning, lane departure warning, and panoramic parking assistance to promote products with international advanced level. From the technical point of view of basic support, China is LTE-V2X and the communication technology fields such as 5G-V2X have mastered certain international discourse rights and formulated V2X data exchange interaction standards. In addition, technologies based on Beidou's high-precision positioning and maps have begun to take shape, so, intelligence The network test and evaluation system has been basically constructed.
In addition, major car companies have begun to develop intelligent driving technology, the main functions include sensory positioning, intelligent decision-making, system control and related voice systems, remote control and so on.
Wu Junjie said that the key to automotive intelligence is to build a new architecture with sensing, computing, communication, decision-making functions, and design an intelligent computing platform that integrates data fusion, high-speed computing, intelligent decision-making, and collaborative control capabilities to complete vehicle driving and information. High-speed computing processing of multi-source massive heterogeneous data in the process of interaction, providing real-time response for decision-making and control, realizing auto-driving, networking services and other functions of the car.
Through sensing, decision making and control, vehicles can be driven automatically (pictures from the network)
Taking Hezhong New Energy as an example, it has built a mature and complete R&D system in intelligent driving, including the integration of multi-sensory positioning, the algorithm of multi-scenario optimization, the interaction of multiple environmental information, and the platform for multi-dimensional computing needs. The domain controller is the intelligent driving brain (IDB), with the intelligent cockpit as the core human-machine interface, providing algorithm solutions with advanced technologies such as artificial intelligence, deep learning, blockchain, big data, etc., cloud services as a value-added service tentacles, fully defined Smart travel new experience, and look forward to this as a fulcrum, inciting the future of smart travel in the big market.
Wu Junjie introduced the unmanned function of Hezhong PIVOT intelligent system at the strategy conference held recently.
Wu Junjie said that although smart cars are becoming 'winds', many of them have similar technologies and lack core originals. 'Xinzhong New Energy will adhere to independent research and development in four aspects: sensory integration, algorithm development, internet communication development and computing platforms. In other areas, it will be promoted in cooperation with third parties. '
Wu Junjie also said that independent research and development does not mean closing the door to build a car. Especially in the case of China's relatively weak automotive electronics related industry, insufficient technology accumulation, and lack of effective cooperation mechanisms between related companies, development still faces arduous problems. Challenge. This requires a unified understanding of the whole industry, cross-border integration, active innovation, and accelerate the electronic information and automotive industry 'changing overtaking'.
Actively embrace autopilot technology, safety attributes are still the first
'The complexity of autonomous driving determines that its development cannot be achieved overnight. The application of autonomous driving technology is still in the initial stage of development. For autonomous vehicles, the safety attribute is still the first. ' Wu Junjie said that we are both To actively embrace the autonomous driving technology, we must also strengthen the basic skills of building a car to make the car more secure and reliable.
Today's headline arithmetic center previously released "Running to 2025 - Automated Driving Potential Consumer Report", which surveyed nearly 6,000 users. The results showed that 80% of respondents said they dared to use autopilot cars. For drivers who drive cars, safety is their biggest concern.
Wu Junjie said that quality and safety is something we don't care about, no matter what stage the technology is in. For example, in terms of network security, if the mobile phone OTA (software online upgrade) is cracked, it may be caused by information leakage. And so on, but if the car's OTA is cracked, it will directly threaten the safety of the occupants.
Automated driving development speed policy and regulations need to be improved
Autopilot cars are currently accelerating into technology outbreaks, and the lack of policies and regulations, infrastructure, high-resolution maps, and technical safety standards may directly affect their development.
Wu Junjie said that in the commercialization of high-definition maps, the relevant national ministries and commissions have not yet come up with detailed timetables. There are no uniform standards for other technical aspects. Everyone is waiting, they are crossing the river by feeling the stones. At the same time, in the closed environment of enterprises. Under the test and operation, the enterprise can meet its own needs; if it is tested on an open road, it must comply with the relevant laws and regulations. On the one hand, in terms of policies and regulations, it is necessary to introduce laws and regulations that are compatible with self-driving cars, and regulate market behavior; On the one hand, it is necessary to focus on the standard construction of automatic driving technology and supporting measures.
It is reported that in April this year, China issued the autopilot road test specification, which was officially implemented on May 1 this year, providing a standard for technical testing of autonomous driving on real roads. Currently, Beijing, Shanghai, Chongqing, Shenzhen and other cities. Automatic driving test management regulations have been issued.
Wu Junjie said that autonomous driving is a complex systematic project that will revolutionize the automotive industry and the entire industry. Today's autonomous driving technology is in the process of transition from maturity to maturity, any key technology. The breakthrough will play a decisive role in the acceleration of autonomous driving. But automatic driving is not just a matter that a car needs to solve, but something that the entire industry chain and the whole society have to face.