In a vegetable distribution center in South Korea, a maintenance worker was mistaken by a robot for a vegetable box and carried to a conveyor belt, and was crushed to death with serious injuries. For a while, it aroused public concern and discussion about the possible harm caused by artificial intelligence and future robots to humans. Actually, this time it's really the robot that is to blame.
The application of robots in modern factories has become more and more common, the so-called "robot" here is called "process robot", Japan's Fanuc, Yaskawa, Germany's Kuka, including China's Xinsong, etc. are professional industrial robot manufacturers. However, the original origin of the name of industrial robot is only because the machine imitates the human arm and sets six degrees of freedom, which can "freely" grasp the workpiece in a certain space. The kind of "arm" with more degrees of freedom is called "manipulator", and the one that realizes automatic control is called "robot", and there is an additional title of "person".
In modern enterprises, the use of industrial robots to replace people to work has become common. For example, in large domestic automobile companies, there are hundreds or thousands of robots working in a workshop, more than workers. The reason for this design is that the robot can be competent for complex and difficult working environments, never "complains", and works quickly and is not easy to make mistakes. Industrial robots are completely programmed by people, just a few actions, repeated thousands of times.
The result of the application of robots is to greatly improve production efficiency and product quality, but in turn squeeze people's employment opportunities, a factory of five or six thousand people, relying on the application of robots, can reduce the number of people to hundreds. You have the opportunity to go to the welding workshop of the automobile manufacturing plant to see, on a production line, hundreds of robots are working at the same time, which is very spectacular. The operator is to do "assistance" throughout the process and control the work of these robots.
As for the accident in South Korea, it was actually an accident caused by a maintenance worker who violated safety regulations and accidentally triggered the working conditions of the robot while the equipment was still working automatically, and the equipment worked according to the set program.
Under normal circumstances, maintenance personnel must first turn off the power supply. When it is really necessary to simulate the maintenance of live electricity, at least two people must be working, one person dedicated to monitoring and emergency operations. And it is necessary to anticipate the possible maintenance risks and take preventive measures. If the maintenance is carried out in accordance with the safety regulations, this kind of accident will not occur.
The responsibility of the industrial robot is that under the conditions set by humans, as long as they are satisfied, they will receive the work instructions, and they will start working no matter what the "workpiece" is. As for how to correctly identify people or things on the Internet, it is related to the detection method, if it is only detected or not, it does not matter whether it is a person or a thing. In order to accurately judge whether it is a person or a thing, it is necessary to invest in the image recognition system, which can be done technically, but does it require such a high-cost investment for a cabbage handling?
Industrial robots can only be regarded as automation at most, and the "robots" under the future artificial intelligence technology discussed by everyone are not the same thing at all.
Therefore, in this accident, the robot expressed its reluctance to take the blame. Why bother wronging robots for human mistakes?