Kunpeng Project
Domestic autonomous driving technology is leading, and the Tesla phenomenon is seen from perspective
The world's first in the field of unmanned driving in China, understanding Tesla is Tesla!
It is known that in the initial period of the development of UAVs, there were two main camps. One is a pure visioner, such as Tesla, which uses cameras to collect data and achieve autonomous driving through fully visual calculations.
In the second direction, Huawei is the leader, with Huaxia Company as the mainstay, using lidar to collect information, and then combining it with cameras to realize the automatic driving of drones.
At the beginning of Tesla's image computing rules, it was criticized for the information it could obtain in the dark, snow, snow, heavy rain, fog, and other low-light environments.
Because the cameras are two-dimensional and lack the ability to detect objects, even the most sophisticated algorithms cannot erase all these defects, so in their eyes, Tesla's so-called "unmanned" is not even comparable to the magical "lidar"...
But Tesla's cars made in China are equipped with a lidar, which seems to give people more awareness of the fact that they cost much more than cameras.
Therefore, it is understandable for Tesla to reduce production costs as much as possible, after all, Tesla's super computing power can be used to solve this problem.
Later, several local companies began to imitate Tesla and changed their name to Tesla, in other words, they replaced lidar with a pure image processing system.
First of all, through an intuitive way, you can save a lot of money, today's car companies, the higher the **, the more reluctant consumers are to buy, which is a theoretical empty talk.
Secondly, camera technology is still developing, and the network supplier chip can now work normally in low-light environments, so the problem of insufficient information has been gradually solved.
At the same time, with the continuous development of emerging technologies such as AI and OCC coating mesh, the three-dimensional image is restored to a three-dimensional image by using the two-dimensional image captured by the camera, and then the three-dimensional image is restored.
This shows that anyone who claims that Tesla bought the cameras just because of ** is ridiculous.
Musk believes that since people can use their eyes to drive cars, then cameras can definitely do it, but if they can't, then there is something wrong with the technology, so they can only stop technological progress and make improvements.
In retrospect, Musk's understanding of the nature of technology is indeed quite in place.