The theme of this forum is "refresh", with more than 60 keynote speeches and topic discussion 12. Hundreds of the most thoughtful leaders and elites in China automobile industry, such as Dongfeng Motor, SAIC Motor, BAIC Group, FAW Group, Great Wall Motor, Chery Automobile, Daimler, Weilai, Tucki, Baidu, Huawei, Horizon, etc., were invited to make suggestions for the annual ideological feast and the development of the industry.
What users are most concerned about is the product experience. Driver assistance products are changing drivers from controlling cars to supervising cars.
Unlike previous autopilot forums, which focused on the technology itself, China Blue Book Forum focused on "what kind of autopilot users need". ?
202 1 is called the first year of commercialization of autonomous driving, and the high-speed advanced assisted driving products of major car companies and head autonomous driving companies have been commercialized; After entering 2022, with the competition of autonomous driving technology from high-speed domain to urban domain, the research and development of advanced assisted driving technology around urban domain has also entered the countdown.
Although the industry is optimistic about the development of autonomous driving, the user perception is obviously insufficient. Specifically, L3 autopilot always stays in the concept stage, while marketing packages such as L2+, L2.5 and L2.99 around L2 assisted driving emerge one after another. Is this for the needs of corporate publicity, or can it really bring users a quality travel experience?
In this regard, Gu expressed his views. He believes that the main difference between L2-level assisted driving and L3-level automatic driving lies in the division of responsible subjects; For users, the latter is more concerned with the product experience. Gu said at the scene, "As a user of Zhixing's assisted driving products, I drive micro-mocha and Tank 300 to and from work every day. In this process, I am most concerned about product experience and improvement. "
"In the process of driving, my task is to supervise my car to drive automatically. I hope that I will get involved less and less, so that I can know what scenes I am suitable for and what scenes I am not suitable for. Some scenes can be relaxed and some scenes can be urgent. " Gu said that driver assistance products are changing the driver's role from "controller" to "supervisor".
How do users perceive the differentiated experience brought by different levels of autonomous driving ability? Gu Hao Wei mentioned that users don't care what you provide, and L has applicable scenarios. When users are familiar with it, they naturally know which way to drive, and drive slowly. With the continuous upgrading of vehicles, users' own feelings can also be upgraded and increased. At present, the core usage scenarios of user-assisted driving high frequency are car-following scene, intersection scene and lane-changing scene.
The key to the breakthrough of autonomous driving lies in data. Thanks to the advantages of mass production of Great Wall Motor, Gu said that Mimo Zhixing won the key to win the rapid iteration of autonomous driving, and thus launched the data intelligence system MANA, focusing on accumulating data with high efficiency and low cost to drive the continuous iteration of autonomous driving technology.
At present, Mimo Zhixing's assisted driving products have been mass-produced on six models of Great Wall Motor, and the cumulative assisted driving mileage of users has exceeded100000 km, and this data will also increase exponentially with the mass production capacity. "Autopilot needs a larger vehicle scale and data scale, in addition to a stronger ability to turn data into knowledge, so that autopilot can quickly realize the iteration from quantitative change to qualitative change, and thus grow to L3 or even L4."
The mass production sprint of NOH in Millimeter City started from the first principle and chose the re-perception route.
A successful product should not only pay attention to the technical level, but also consider the demands of users. Based on the in-depth thinking of the first principle, Gu put forward his own views on the development direction of autonomous driving.
In Gu Hao Wei's view, urban commuting has become the most important use scene of cars at present; Users' demands for advanced assisted driving products in urban areas are also more clear, that is, safe, simple and easy to use.
What demands should advanced driver assistance products in an urban area meet from users? Prior to this, many autonomous driving companies put the core of competition on perceptual hardware and high-precision maps, but Gu Hao Wei put forward the development trend of autonomous driving industry and expressed his own views. "High-precision maps and lidar first appeared because the camera sensing ability of autonomous driving technology was weak at that time, and the computing performance was not enough to support the sensing algorithm; Now, not only the camera pixels have been greatly improved, but also large-scale computing chips have begun to appear, and the perception ability has been greatly improved. Even the cost of expensive lidar has gradually begun to explore. "
Therefore, Mimo Zhixing put forward the technical scheme of "re-perception". It takes the data intelligent system MANA as the core, and takes massive data as the nourishment, so that the study time of MANA is more than 240,000 hours, and the virtual driving experience is equivalent to 20,000 years of driving by human drivers, so that NOH, a city at the end of the century, can explore and master driving skills to deal with various complex urban scenes, thus realizing the transformation from "novice" to "old driver" and allowing drivers and passengers to enjoy safe, comfortable and efficient travel.
Excellent user experience is not limited to safety, comfort and efficiency. In fact, there are still many details worthy of attention and improvement when users use autonomous driving products. In this regard, Gu pointed out a typical example. "When a high-speed NOH vehicle passes a big truck, the driver often instinctively takes over the vehicle out of fear of the big truck, even if the vehicle does not collide with the big truck at this time."
In this regard, Gu introduced the "intelligent evasion" function of NOH. "If there is a big car on the right, we set the vehicle to shift to the left in the left lane 10- 15 cm. At the same time, NOH will be marked on the interactive interface to inform the user of the evasive action of the vehicle and give the user psychological expectations. " Gu Hao Wei mentioned, "At first, this function will use simulated artificial voice to remind users, but when there are many big cars on the road, frequent voice prompts will bring more trouble to users. Later, we changed this interaction into a short voice in a low voice, which is easier for users to accept than the previous voice reminder. " It can be seen from this case that by paying attention to and adjusting the details, Mimo Zhixing has gradually made the assisted driving products more mature and is gradually realizing product differentiation.
Gu's clear and unique views also triggered discussions and heated discussions among many guests at the scene. The theme of this year's China Automotive Blue Book Forum is "Refresh", which shows the high-speed development of the industry, making innovations hand in hand and spanning the cycle.