According to recent WHO (World Health Organization) statistics, more than 1 million people in the world die yearly in car accidents. South Korea is no exception. Last year in South Korea, 4,100 people died in car accidents. Almost 90% of all car accidents are caused by human error: falling asleep at the wheel, drunk driving, using a cell phone while driving, and careless driving habits.
Many automobile manufacturers have concentrated on the development of self-driving cars as a solution to human error. A self-driving car is a vehicle that is capable of sensing its environment and navigating without human input. Elon Musk, CEO of Tesla Motors, said he expects that by the end of 2017, a Tesla vehicle would be able to drive safely in full autonomous mode from LA to New York “without the need for a single touch on the wheel.”
The Principle Behind the Wheel of Self-Driving Cars
So how can a car drive itself? According to Sridhar Lakshmanan, a professor at the University of MichiganDearborn, there are three requirements for self-driving cars to operate properly.
The first is a positioning system that works like a GPS (Global Positioning System). This system defines the “mission” of the self-driving car by setting the starting and ending point of the trip. This system then chooses the fastest route to take in order to get to its destination. To ensure more accuracy, experts have been working to bridge the gap between the real position of the car and the position that the car recognizes.
The second system that is required is one that recognizes the dynamic conditions of the road: ADAS (Advanced Drive Assistance System). ADAS works as a complement to the GPS system, offsetting its limitations. “A self-driving car requires a second level of intelligence with the ability to fill in additional details in the map,” says Lakshmanan. He calls this system “a differential GPS.” “If you think of the map as having a static view of the world, the sensor system provides a dynamic fill-in to that map and these two, together, provide what is called a ‘world model’ for that self-driving car,” he says. These two help self-driving car grasp road condition and decide effective driving strategy. ADAS contains several technologies including AEB (Autonomous Emergency Braking), LDWS (Lane Departure Warning System), LKAS (Lane Keeping Assist System), and BSD (Blind Spot Detection). Also, in the near future, technology V2X (Vehicle to Everything) helps cars communicate with each other so they can comprehend driving strategy of near cars, which will make driving more flawless.
The third system is the AI (ArtificialIntelligence) technology system that utilizes the information from the other two systems and transforms them into an action such as steering or accelerating. However, several accidents occurred because of the wrong decisions made by self-driving cars’ AI systems.
The Advantages of Self-Driving Cars
The greatest advantage of selfdriving cars is a safety. Christopher Hart, chairman of U.S. National Transportation Safety Board, said in an interview with MIT Technology Review, “I am very optimistic that self-driving cars will cut into the auto accident death toll.” Nathaniel Fairfield, a Google chief engineer, claims that “more than 90% of car accidents are caused by humans, and if Google AI commercialized, the risk would decrease.” Safety can be ensured because the three systems serve to eradicate careless human driving mistakes that usually cause car accidents in the first place.
Self-driving cars are also predicted to improve quality of life. While the car drives itself, drivers can rest and do what they want. According to researchers in the US, self-driving cars will save commuters up to 50 minutes a day. That is almost an extra hour that can be used for enjoying daily activities that one could not normally have done while driving.
Self-driving cars also have an economic advantage. Since self-driving cars help to prevent accidents caused by human error, car insurance premiums are likely to be decreased. Mckinsey, a consulting firm in the US, said “Commercialization of self-driving cars will have a 190-billion-dollar economic effect.” As crash avoidance technology gradually becomes standard in selfdriving cars, insurers will become better at identifying the cause of an accident, and therefore be able to determine the exact amount that they would need to pay out to a claimant. Being able to calculate the payout precisely will make it possible for insurance companies to charge less.
The Disadvantages of Self-Driving Cars
However, there are negative consequences to self-driving cars. One is the loss of jobs. If cars can drive themselves, those who make a living out of driving will lose their jobs. Transportation companies would no longer need to hire bus drivers, taxi drivers, or subway drivers. It is expected that 500,000 people will lose their jobs in America alone. Therefore, countermeasure should be prepared for such drivers.
Moreover, self-driving cars are exposed to the danger of hacking. Some experts say if someone hacks the selfdriving car, he or she can control the car by other mobile vehicle freely. In effect, China’s Keen Security Lab succeeded in remotely hacking into a Tesla Model S by exploiting the vehicle’s in-car web browser. Also, hackers can collect various personal information like moving route, voice recording, etc.
The Current State of Self-Driving Car Technology
The U.S. National Transportation Safety Board categorizes the technology of self-driving cars into four levels. Level 1 is a function-specific automation.
At this level, automation involves one or more specific control functions. Level 2 is a combined function automation. This level involves an automation of at least two primary control functions. Level 3 is a limited selfdriving automation. Cars at this level enable the driver to cede full control of all safety functions under certain conditions. Drivers are expected to be available for occasional control. Level 4 is a full self-driving automation. Vehicles are designed to perform all driving functions and monitor roadway conditions for an entire trip.
Currently, vehicles with level 4 automation are referred to as an autonomous vehicles. Google, which has the most advanced technical skills, is near level 4. In South Korea, Hyundai is near level 3. Naver has teamed up with Toyota and is now planning on showcasing a car with level 4 automation.
Although a self-driving car technology is being developed at a fast pace, there are still certain components of driving that prevent these cars from becoming commercialized. Traffic, the route of other cars on the road, and risk of malfunction are some of the variables that still make self-driving cars risky. Therefore, more research and data that deal with these variables are needed.
What We Need to Keep in Mind
Ya-Qin Zhang, president of Baidu, the biggest Chinese Internet company, said “Self-driving cars will become the future computing platform. Take note of the advent of Internet Vehicles in the era of the Internet of Things.” However, because human lives are at stake, we should thoroughly verify the safety of self-driving cars. Therefore, although self-driving car technology is developing at a fast pace, it should be phased in with numerous evaluations and trials to ensure safety.