"People will launch driverless cars in five years," Ford CEO Mark Fields said at CES.

Over the years, driverless cars have always been a little far away from us - many people believe that such cars will not be available until at least another ten years. If even the conservative auto industry believes that driverless cars will be launched in five years, we may be closer to this dream.

The CES in January this year is no longer the world of Samsung, Toshiba and a group of Chinese iPhone manufacturers. All large automobile manufacturers have also passed the Website production Booths have been set up - their purpose is also very unified, and they are all here to discuss the technology of driverless vehicles. However, in fact, it is not only the automobile industry that is optimistic about this market, but also many chip manufacturers.

Nvidia's keynote speech was very boring, describing the company's new generation of chips and the latest progress in deep learning technology: how to use the camera car to drive automatically without using expensive sensors. All of Nvidia's products are focused on the automotive industry.

　　 high pass Although it has not invested so much energy, it has also announced that many concept cars use the company's chips. Intel talked about the automatic driving of UAVs, but the company will undoubtedly apply these sensors to cars.

Google clearly wants to play a role in this area, as does Tesla. However, although these projects have attracted wide attention, the automobile industry is still likely to beat them in the market. These enterprises are testing all kinds of driverless cars. Audi even drove five journalists from Los Angeles to Las Vegas in driverless cars. Ford said that its driverless cars are now on the road, and other enterprises have also launched similar attempts.

But five years is not far away, and there are still many unanswered questions in the field of driverless vehicles. For example, once an accident occurs, who is responsible? How to avoid driver negligence, and how to hand over manual control to the driver again? How to avoid driver carsickness? Is driving fun? In the long run, automobile manufacturers may also face greater risks: driverless cars will change the relationship between human beings and cars, making us more dependent on carpooling services. If this happens, car sales will undoubtedly be affected.

But there are other questions to be answered. Since almost all driverless vehicles rely on highly accurate maps, drivers must keep in touch with manufacturers at all times so as to obtain regular updates. Does this mean that drivers must subscribe to these services? Will automobile manufacturers form an alliance to share map data with each other?

In addition, there are many regulatory barriers to overcome. All my communication objects at CES believe that what really hinders the development of driverless cars is not technology.

But the old American automakers may not be the first to launch driverless cars. "The key may not be who goes on the market first, but who has the last accident," a Ford spokesman said this week. In a country like the United States that likes to litigate, only a few major accidents will set back the entire industry for several years. Before 100% safety is ensured, consumers will not buy these cars - since all drivers think that their driving skills are good, driverless cars must be outstanding in this regard.

Therefore, cars in the next few years may gradually add more automatic driving functions. Adaptive cruise control will evolve gradually to achieve automatic driving in traffic jams. Lane departure warning will also become a true lane keeping function to help drivers on the expressway. Assisted parking will also evolve into a true automatic parking. All these technological advances will gradually converge. It is hoped that this gradual development will enable regulators to recognize the industry trend and eventually allow driverless cars to run on the road.

One day, the expressway or some urban roads may even prohibit people from driving cars. After all, human prediction ability and driving efficiency are far lower than computers.