Huawei's first public autonomous driving test ride: Easily handles complex roads

On April 15, the BAIC New Energy Alpha S HI version of the Arcfox equipped with Huawei's autonomous driving technology was publicly tested in Shanghai. This was also the first test ride demonstration of Huawei's autonomous driving technology.

Car blogger @42号垃圾箱 posted a video on Weibo showing the test results of Huawei's autonomous driving technology. From the video, it can be seen that the on-site test vehicle drove smoothly and performed well in terms of starting and stopping at traffic lights, unprotected left turns, lane changes, and avoiding cars and pedestrians.

What surprised Xiao Lei was that Huawei's autonomous driving was very experienced in dealing with emergencies. For example, it would give way to vehicles based on priority at traffic lights, and could successfully pass through sections of road with cluttered electric bikes. It would actively keep a distance when there was a delivery electric bike on the right side of the vehicle.

Huawei's rotating chairman Xu Zhijun said: "The R&D team told me that Huawei's autonomous driving technology can achieve 1,000 kilometers of unattended autonomous driving in urban areas, which is much better than Tesla."

Currently, autonomous driving technologies are divided into the LiDAR route and the pure machine vision route. The LiDAR route uses laser pulses to measure distance and works well both day and night, but the cost of LiDAR is still relatively high. The pure machine vision route uses ordinary cameras to image and achieve distance measurement through the visual difference between two sets of cameras. This technology is greatly affected by ambient light, but its advantage is that the implementation cost is low.

Tesla can be said to be the representative of the pure machine vision route. Nowadays, almost all Tesla cars are equipped with cameras to realize Autopilot automatic assisted driving. Tesla CEO Musk once blasted "those who use lidar are fools" at the "Autonomy Day" product launch conference. He believes that automatic driving can be completed by relying on pure visual algorithms based on cameras, without the need for expensive lidar. Relying on data collection from all Tesla cars currently running on the road, Tesla has achieved a leading level among its peers in terms of technology implementation and data collection volume.

Unlike Tesla, Huawei uses both LiDAR and camera technologies in hardware, and the integration of the two technologies provides better visual feedback for autonomous driving. At the same time, Huawei has also solved the problem of high prices for LiDAR, reducing the cost from thousands of dollars to hundreds of dollars, and has already achieved mass production. Huawei has found a new way to achieve breakthroughs in the field of autonomous driving in the face of US sanctions on chip technology. This tenacious spirit of innovation is truly admirable.