Recently, Huawei’s princess Yao Anna is very popular, and even President Ren has come out to apologize for the trademark registration of Yao Anna. Princess Huawei made her debut, and the list she got was not a mobile phone, not a base station, but a car.
Huawei also took the consumer BG’s Mr. Yu to the automotive field, and communications and automobiles are beginning to approach two fields that are far apart.
In fact, as early as last year when the 5G speculation was booming, some people connected 5G with autonomous driving. By 2021, several domestic automakers have already put 5G and auto products on the ground.
So, will 5G technology really disrupt autonomous driving? Let’s take a look.
Where are the real advantages of 5G technology?
Since the popularization of 5G technology, the argument about 5G chicken ribs has always existed. The new features of 5G include EMBB (enhanced mobile broadband), URLLC (ultra-high reliability and low latency) and MMTC (mass machine communication) standards.
However, when people actually use 5G mobile phones, they find that the experience is not much different from 4G. When only downloading big games, the 5G network speed is indeed much faster.
The legendary high bandwidth does exist, but the low latency is not obvious, because the application on the mobile phone is not only from the base station to the mobile phone. The base station to the terminal has low latency. The base station passes through the layered network to the target website on time, and the delay is the same. It was almost the last time.
However, if the end-to-end transmission is within the range of the base station, 5G technology has an overwhelming advantage over 4G technology.
For the automotive industry, there is a 5G-V2X standard in the 5G era. The 5G-V2X-based Internet of Vehicles introduces four types of applications: vehicle formation driving, advanced driving, sensor expansion and remote driving.
Among them, sensor expansion requires the interaction of on-board sensors or on-board dynamic video information between vehicles, between vehicles and roadside units (RSU), between vehicles and pedestrians, and between vehicles and V2X application servers, so as to obtain a more comprehensive current Environmental information of the road.
In V2V (vehicle-to-vehicle) and V2I (vehicle-to-roadside unit) direct communication, 5G low-latency communication can be directly used. This is of great significance to autonomous driving.
Several routes for autonomous driving
For autonomous driving, the automotive industry has actually studied very, very early. Before the L1 to L4 classification, some manufacturers have tried autonomous driving on highways.
It is Tesla that makes autonomous driving really into the public eye. Tesla used Mobileye’s technology in the early years, focusing on camera recognition and calculation, assisting millimeter wave radar and ultrasound, and realizing low-level autonomous driving. At present, Tesla’s unmanned driving route focuses on improving computing power. It has developed high-performance AI computing chips by itself, using camera machine vision recognition as the main source of information, and using the improvement of computing power to improve the reliability of autonomous driving.
Real L4 level unmanned driving requires multi-line lidar, and it is more accurate to rely on lidar to identify objects. However, in the early years, multi-line lidars were very expensive, and a radar was more expensive than a car, so the L4 level unmanned driving in the early years was limited to concept products. Later, solid-state lidar became cheaper and cheaper, and lidar began to be used. The ET7 released by Weilai uses solid-state lidar.
With the development of lidar in the future, autonomous driving with lidar as the main sensor will also be yours.
With the development of 5G, the sensor expansion of 5G-V2X has brought a new path. Whether using a camera or millimeter-wave radar or lidar, the detection range is very limited. And 5G-V2X’s sensor expansion and inter-terminal communication can greatly expand the range of perception.
SAIC’s upcoming MARVEL R uses the Mobileye EyeQ4H chip (Mobileye later parted ways with Tesla, and the EyeQ4H chip is a chip using the MIPS instruction set) and Huawei’s 5G Barong 5000 solution. Great Wall’s new car with Yao Anna as its spokesperson also uses the 5G-V2X solution.
SAIC MARVEL R and Great Wall’s new cars have introduced 5G-V2X and use V2V (vehicle-to-vehicle) and V2I (vehicle-to-roadside unit) for direct communication. Can solve the problem of sensor detection distance.
After the popularization of 5G equipment, the information collected by roadside cameras, other vehicle information, and traffic light information can all be directly connected to the car through the low latency of 5G, which is equivalent to the expansion of the car’s sensor from the line of sight to the entire 5G base station and low Delay the area covered by the network.
With this scheme, the level of autonomous driving can be raised by one level. Even for people driving, there is an improvement in the driving experience. The road and vehicle conditions can be sensed through the network, and reasonable route planning can be made to avoid congestion. Direct early warning of visual blind spots to avoid accidents.
Will 5G technology revolutionize autonomous driving?
Although the 5G-V2X standard is good, we have to see that the usage scenarios described by 5G-V2X are highly dependent on infrastructure construction. V2V (car to car) and V2I (car to roadside unit) communicate directly. Enough cars are needed to support 5G-V2X, various transportation infrastructures are needed to support 5G-V2X, and 5G base station coverage is needed. If a larger range of information coverage is required, then high-speed and low-delay network construction between base stations is also required.
It will take time for 5G-V2X to be popularized in car terminals, and it will take time for various infrastructure constructions.
Fortunately, China’s 5G construction is far ahead of the world, and China is also very willing to build various 5G infrastructure. China has 5G industry leaders such as Huawei and ZTE, and Chinese automakers such as SAIC and Great Wall are also very willing to invest in their own 5G-V2X is supported in the product.
Therefore, China is likely to be the first to realize the 5G revolution to autonomous driving.
Moreover, China’s electronic consumer industry has advantages and it is easy to integrate various autonomous driving routes.
Tesla has the fastest chip, and Chinese automakers can buy nVIDIA chips to achieve the fastest AI computing power.
Multi-line lidar is expensive. China can develop its own lidar technology to make lidar cheaper. Huawei is already engaged in lidar.
In the future, with the popularization of 5G-V2X, self-driving cars made in China can instantly adjust driving strategies based on road conditions and surrounding vehicle conditions. This is not only a question of improving the safety of autonomous driving, but also a huge improvement in the experience of manned driving.
Even if you are a driver, the car can warn you which route is congested, which red light time is shorter, how far in front of you there is a car accident, how far behind you is a car speeding, and how many cars are in the blind area of the intersection you are going to cross Will there be a hidden danger of accident at the speed of the vehicle? This experience will be disruptive.
For consumers, under the same conditions, they can try their best to buy models that support 5G-V2X. Supporting this standard will have the possibility of software and service upgrades in the future, just like the advantages of smart phones over feature phones.
