On February 21st, American stocks surged 70% in Arbe trading, closing up 47.13%, spurred by the news that Tesla, Inc. will carry 4D millimeter wave radar.
Tesla, Inc. uses 4D radar
Tesla, Inc. submitted a vehicle change application to European regulators, confirming that the latest self-driving hardware HW4.0 is about to be mass produced. It is reported that the performance of HW4.0 is estimated to be about 2-4 times that of HW3.0. In the upcoming HW4.0 hardware system, Tesla, Inc. not only equipped a high-resolution millimeter wave radar (or high-precision 4D millimeter wave radar), but also added a heater for the radar to prevent the impact on radar signals in ice and snow weather.
4D millimeter wave radar is not an unfamiliar new technology. In March 2020, Alphabet Inc-CL C's Waymo company released the fifth generation autopilot system perception scheme, upgrading millimeter wave radar to 4D imaging radar, making 4D millimeter wave radar technology applied to the vehicle side for the first time. Huawei entered the bureau in 2021, launched high-resolution 4D millimeter wave radar, and plans to achieve mass production in the second half of 2022, pushing 4D millimeter wave radar to a hot peak.
In October last year, foreign media electrek reported that Tesla, Inc. was adding a new 4D radar with twice the range of the previous radar, which mentioned Arbe Robotics, which has a radar called Phoenix. The word Phoenix was also mentioned in the report submitted by Tesla, Inc. in December. In addition, Green, a well-known hacker, mentioned the possible relationship between Phoenix radar and Tesla, Inc. earlier, and thought that Phoenix radar perfectly met Tesla, Inc. 's requirements.
Musk's "dislike" of radar is well known. He said in 2019 that people who use lidar are fools. In May 2021, Tesla, Inc. Model 3 and Y models in North America were the first to cancel millimeter wave radar. The last remaining ultrasonic radar was officially cancelled in October 2021.
Will there be more than one million 4D radars in 2023?
According to research by professional institutions, 4D imaging radar will be imported on a small scale from the beginning of 2023, and the fixed-point / carrying capacity is expected to exceed one million in 2024. Tesla, Inc. 's recovery of radar scheme also shows that the current pure vision scheme is still difficult to meet the requirements of security redundancy, and multi-sensor pre-fusion is still the general trend in the future. With the continuous improvement of high-order ADAS permeability, the sensor market such as lidar and 4D imaging radar will maintain rapid growth in the future.
The millimeter wave radar carrying rate of new cars of L2 + and above in the domestic market is expected to exceed 50% in 2025, according to data from the High Engineering Intelligent Automobile Research Institute. At the same time, 4D millimeter wave radar will be imported on a small scale from the beginning of 2022. It is expected that the number of 4D millimeter wave radars will exceed one million by 2023, and the proportion of all forward millimeter wave radars is expected to exceed 40% by 2025.
Multi-sensor fusion is a realistic choice for intelligent driving.
So is it possible for 4D millimeter wave radar to replace lidar? It is pointed out that the cost of 4D millimeter wave radar is similar to that of traditional millimeter wave radar, and its performance is comparable to that of low beam lidar, which is more in line with the current demand of mass production.
The disadvantage of lidar is obvious, that is, heavy fog, rain and snow weather effect is poor, and can not obtain external images. Cai Qinghan said that if the lidar operates in a foggy environment with visibility of only 0.1 km, it has almost no way to detect it. If the lidar operates in an environment of heavy rain, the detection range will be reduced by about 50%. The measured results of 4D millimeter wave radar show that even in rainy weather, the detection range of 4D millimeter wave radar can still reach 300 meters, which is an important reason why 4D millimeter wave radar is more suitable to be used as autopilot sensor than lidar.
At present, the market consensus: self-driving technology can not rely on a single sensor to dominate the world, multi-sensor fusion is a popular solution. Because there are many segments of the market and different levels of autopilot, cameras and radars will co-exist because their advantages and disadvantages are very complementary. The special one is lidar. The solution of 4D millimeter wave radar can reduce or replace the use of lidar.
Edit / irisz
Comment(0)
Reason For Report