Controlling robots by nothing other than the human mind sounds like something straight out of a science fiction book. Previous efforts which were made to build a human and robot symbiosis ended in failure. A group at MIT’s Computer Science and Artificial Intelligent Laboratory (CSAIL) department and researchers at Boston University have collaborated to create what they are calling a response system. This system involves an electroencephalography (EEG) cap that can be worn over the head. This cap captures brain movements and transports it to the robot in a few seconds. “Imagine being able to instantaneously tell a robot to do a certain action, without needing to type a command, push a button or even say a word,” stated Daniela Rus, the CSAIL Director.
This study involved the use of a robot called Baxter. The robot was ordered to assemble paint cans among the spools of wire. The major highlight of this system is the method by which the robot can understand the human controller’s thoughts even without the individual having to modify its normal thought process. If any error is detected by the controller, that detection is passed on to the robot allowing it to rectify itself from making similar mistakes in the future. Researchers involved brain signals known as error related potentials (ErrPs) in this study to make it more natural. These signals are generated automatically in the human mind every time the person notices a mistake taking place.
No comments :
Post a Comment