The act of ‘driving’ through a car or other means of transportation is often done when you go to school or to work. To reduce the laborious amount of driving and time, autonomous vehicles have been developed. An autonomous vehicle does not require human labor and automatically moves in accordance to a given program. The basic technology behind autonomous driving is SLAM (simple localization and mapping) which enables location estimation and mapping. In addition, various algorithms are used in deep learning artificial intelligence.
Currently, the trend of autonomous vehicles is at a level where partial autonomous driving is possible only in certain sections. Various technologies have been commercialized and adapted, for example, the adaptive cruise control technology(ACC) function that automatically keeps a distance from the vehicle in front of you, and automatic emergency braking technology (AEB) that responds to the emergence of obstacles. However, there still needs to be a driver when driving, and there is still too much process to go through in autonomous driving. Then, what parts should be further developed in autonomous driving?
First of all, artificial intelligence systems must be developed. Deep learning technology is also an important technology for autonomous driving, and it is very important for autonomous vehicles to learn deep learning for an enhanced performance. Deep neural networks (DNN) and voice recognition, which are used to identify objects, are utilized through deep learning. Basically, the development of AI is necessary.
In addition, since there are countless scenarios that occur while driving, the development of artificial intelligence systems is required to learn the whole data of scenarios.
Second of all, when an accident occurs, the responsibility as a driver or as an automobile company should be clear. Therefore, a sufficient technology must be developed as well as a successful implementation of institutional laws. For example, if an automobile is set upon a case like the trolley dilemma, decisions should be clear. Trolley dilemna provides two unfortunate outcomes whereas you have choose only one depending on your ethical beliefs. When two five passersby can be saved if you change the orientation, one passerby still dies as a result. Therefore, the ethical problem of autonomous vehicles should be resolved. You drive more than 300 hours a year. If artificial intelligence technology and ethical problems are resolved, autonomous driving technology will be very helpful which would improve the quality of human life and save time in the future.
By. Sungju Park