Who is the hero behind the scenes? The technology behind autonomous driving

Not long ago, Google's AlphaGo defeated Korean chess champion Lee Sedol on the chessboard, and people exclaimed that the era of computers replacing human brains is coming. Driven by this wave, people naturally think of the accelerated popularization of fully autonomous vehicles. Of course, there are still many variables in the future. Instead of unrealistic fantasies, let's take a look at the heroes behind the scenes who are making unremitting efforts behind its popularization.

-Anti-interference "sensory organs"

Whether it is the current assisted driving system headed by lane keeping and adaptive cruise control, or the driver-centric technology that has been mass-produced and can drive autonomously in some cases, or even the future fully autonomous driving, the basis for making accurate judgments is similar to the way we understand the world through organs, which requires collecting a lot of information. Of course, in cars, these "organs" have become radars, infrared sensors and cameras.

It is precisely because the various types of information they obtain are the fundamental basis for the processing system to make judgments, and the quantity and quality of the information they collect will directly affect the accuracy of the final judgment. Therefore, in recent years, sensors, as the most important behind-the-scenes heroes in autonomous driving technology, are constantly evolving.

- Faster network

A few years ago, when our communication was limited to making phone calls and sending text messages, ordinary 2G networks were sufficient to meet our needs. When people are eager to obtain more information, WeChat and video have become new means of communication, and the network has been continuously upgraded to 3G and 4G. The demand for surrounding information by vehicles is also in an explosive growth stage, which puts higher requirements on the network.

With the development of cloud technology, V2V and V2C technology, the network has become the link for the transmission of various types of information between vehicles and between vehicles and the environment. In other words, if the information transmission between vehicles and other members of the environment is blocked due to the transmission network, it will inevitably affect the accuracy of future fully automatic driving judgments.

-More detailed maps

With the continuous development of autonomous driving technology, the various information collected by R&D personnel has continuously optimized the judgment logic of the active driving system, from the original feedback based on the current road conditions to the prediction based on the road conditions in advance. In short, the change is from passive acceptance to active choice. Of course, in addition to the stable and high-speed network, the hero behind all this is the high-precision map.

The information collected by traditional radar, infrared, and cameras can only be part of the conditions required for judgment. The processing system urgently needs a broader field of view, or in other words, the future processing system needs to have information about the entire driving environment, including all environmental information including surrounding buildings. In addition to real-time traffic information, high-precision maps are a carrier that can provide richer traffic information and are also one of the foundations for predictive decision-making for autonomous driving. However, due to other factors, there are still certain obstacles to opening up access rights in a short period of time.     - More powerful hardware
 

When various sensors have collected a wealth of information, it is time for the processor to show its prowess. Obviously, this is also a test of the processor's performance. When the information is growing exponentially, how to quickly and accurately analyze and classify this complex information and issue commands to the next actuator requires further development of the existing hardware system.

-Smarter logic program

From collecting information to processing information and finally executing it, this interlocking process is just like what we talked about at the beginning, which is very similar to the process of human cognition, judgment and influence on the world. Therefore, the ability of R&D personnel has also become one of the factors affecting the development of autonomous driving.

At the same time, due to the need to adapt to the driving habits of people in different countries and regions and collect relevant information, R&D personnel have begun to drive self-driving test vehicles on open roads for small-scale tests. Taking Volvo as an example, next year it will select 100 experiencers in Gothenburg to improve the entire system through their driving conditions. This test will also be carried out in my country in the future.

- More secure system

When it comes to vehicle safety, most people think of the passive safety performance of the vehicle first. When fully automatic driving becomes popular, the incidence of traffic accidents may be greatly reduced. So, can the development of safety come to an end? Of course not. On the contrary, vehicle safety development will develop in a new direction.

On the other hand, vehicle safety is also reflected at the software level, or in other words, the development of vehicle safety will include more content, such as the stability of the system that uses various types of data as the basis for judgment in dealing with hacker attacks, and the stability of the entire system in various complex environments. These issues require the cooperation of R&D personnel in more fields.

Summary of the full text:

With the continuous development of various technologies, in recent years, they have gradually moved from theoretical development models in the laboratory to open road testing. Its popularization can not only free the driver's hands and reduce the probability of traffic accidents, but also play a positive role in reducing fuel consumption. In order to achieve the long-term goal of fully automatic driving, many behind-the-scenes heroes, led by sensors, high-precision maps, and logical judgments, have already made unremitting efforts. The day when we realize the dream of fully automatic driving is not far away.