Anoj

autonomous vehicles: from laboratory concept to street fireworks, how far are we from the real future travel

When stuck on an elevated road during the morning rush hour, many people have had this fantasy: if they could release the steering wheel, lean back in their seats to catch up on the sleep they didn't get, and when they looked up again, the car would have been parked steadily downstairs from the company building. The scenes that used to only appear in science fiction movies have now quietly taken to the streets of reality - when you take a car in a city commercial district, you will find that there is no driver holding the steering wheel in the driver's seat. When the car encounters a red light, it will stop smoothly on its own. When a car changes lanes in the adjacent lane with a light, it will politely slow down and give way. When encountering an elderly person waiting to cross the road at a zebra crossing, it will stop earlier than a human driver to signal. In recent years, when we talked about autonomous vehicles, our attitude gradually turned from novelty to doubt: can it really be reliable enough to hand over our safety? Why has the concept been shouted for so many years, but we still rarely trust it completely in our daily commuting?
Many people's perception of autonomous vehicles still stays in the imagination of "the algorithm is powerful enough to replace human driving". They think that as long as all human driving actions are taught to machines, they can build a car that does not need human control. But those who truly work in the industry know that the hardest thing is never to make the car achieve a perfect score on an ideal closed test field, but to deal with unexpected scenarios on real roads where even experienced human drivers have to frown - a thin layer of water accumulates on the asphalt road during the rainy season, and the lane lines are blurred by soaking; The door of the van parked on the roadside suddenly opened, and a child holding a ball rushed straight onto the road; During the rush hour in the morning and evening, the delivery guy's electric car flexibly shuttles through the traffic flow, not following the traffic rules. These "long tail scenes" without fixed patterns are the last hurdle that autonomous driving technology needs to cross, and the core reason why we have not yet seen fully autonomous vehicles running on the streets. It's not that it can't solve a fixed math problem, but rather that in the ever-changing real world, like an experienced driver with over ten years of driving experience, it needs to make the most reliable judgment on every unexpected situation.
Talking about safety, the public's concern about autonomous vehicles is completely understandable. After all, if one were to entrust their own safety to a machine system, anyone would be more cautious. But few people realize that it is never absolutely safe for humans to hold the steering wheel on their own: staring ahead for two hours during a long distance can cause a loss of attention, being distracted by work-related troubles can slow down the reaction to sudden obstacles on the roadside, and driving with a lucky mentality after drinking can lead to significant deviations in distance judgment. The relevant traffic statistics have long made these issues clear, and the root cause of the vast majority of traffic accidents is never that human drivers cannot drive, but that people are naturally tired, distracted, and led by emotions. From the first day of design, the autonomous vehicles has made the natural weaknesses of these people into system redundancy - it will not get sleepy after driving hundreds of kilometers, will not miss the traffic lights at the intersection because of answering the phone, and will not step on the accelerator suddenly because of road rage. Just when a small accident occurs in every test in the industry, it will be infinitely magnified under the spotlight, just like when cars were first invented, everyone thought that slow-moving carriages were safe, and motor vehicles that run frighteningly fast will eventually lead to fatalities. The day when technology moved from the laboratory to the day when ordinary people fully trusted it, the gap between them was never the difference in technology itself, but the gradual change in cognitive habits of an entire generation.
Of course, we don't need to shape the autonomous vehicles into a flawless perfect future. Nowadays, the autonomous taxis we experience on the streets still occasionally make jokes about getting stuck in the middle of the road and not daring to move: when a car next to us gets stuck, it will strictly follow the rules and stop in place, not knowing how to be flexible until a long queue forms behind it; When encountering roads with deep water on rainy days, it will stop directly on the roadside due to inaccurate sensor recognition, and wait for staff to remotely come and take over. These seemingly "dull" small flaws actually indicate that autonomous driving is still in its growth stage - it is like a novice driver who has just learned how to drive, strictly adhering to every traffic rule, but has not fully grasped the complexities of human relationships that require flexibility and adaptability. What is even more alarming is the impetuous atmosphere in the industry. Many car manufacturers promote the auxiliary driving function that can only be used on highways as "fully autonomous driving", making car owners dare to let go of their hands and sleep on the highway. When the system fails to respond in emergency situations, they realize that they have been misled by marketing propaganda. The practice of packaging children who are still learning to walk as world champions not only undermines public trust in technology, but also slows down the true implementation of the entire industry.
In fact, we really don't have to think of autonomous vehicles as too science fiction and too far away. The ultimate change it will bring is never to turn cars into such cool technological toys, but to quietly dismantle the inconveniences we take for granted in our daily lives. In the future, the elderly with poor legs and feet want to go to the hospital for further consultation by themselves. Instead of waiting for their children to ask for a special leave to take a detour to pick them up, they can go downstairs and get on a smooth autonomous vehicles to reach their destination safely; Visually impaired individuals no longer need to be trapped in their familiar neighborhoods due to limited mobility. They can easily go to bookstores or parks to chat with friends by simply going out; The time we spend on the road is no longer a burden of having to concentrate on the road ahead. The two or three hours we spend stuck in commuting can be used to catch up on sleep, read half a book, and make a non distracting phone call with family. When the service of autonomous driving becomes widespread enough, the large parking lots that occupy prime locations in the city center can gradually be converted into community parks, children's playgrounds on the street, and small squares for the elderly to play chess. The space in the city is no longer stored for parking and traffic, but returns to its essence of being used by people.
The pace of technological development never stops because of the doubts of one person. Just like when the Internet first came into life, many people felt that surfing the Internet was an unprofessional business. When smart phones first became popular, many people felt that it would disrupt people's lives. The day when the autonomous vehicles finally came to our side would not be an earth shaking press conference. Instead, on an ordinary rainy day, we stood at the roadside and waited for the car. The car stopped steadily in front of us. After getting on the car, we didn't have to talk about today's road conditions with the driver. Only when we leaned against the seats and looked at the street view outside the window and slowly retreated back would we suddenly react: Oh, we had been quietly changed by this technology. It doesn't need to become an unattainable myth of the future, as long as it can make our daily lives easier, reduce the fatigue of running around, and give us more time to ourselves, it's already good enough.