At mit, our approach to control a robot by thought

At the Massachusetts Institute of technology have developed a method of robot control by thought. The staff of the laboratory CSAIL (Computer Science and Artificial Intelligence Laboratory) used the EEG signals, processing them using algorithms that exploit the technology of machine learning. The goal is to identify the so-called «potential errors» — ErrPs. This potential is generated by the brain when a person sees that the robot acts incorrectly, and are used for the correction of his actions.

In other words, the operator, watching the actions of the machine, mentally agrees or disagrees with them. The corresponding signals are recorded by a robot that adapts to human desire.

Because the value of ErrPs in proportion to the degree of error, scientists believe that their method is suitable for complex tasks with several choices. Among the possible areas of application of technology researchers include the control of industrial robots and unmanned vehicles.

Source: MIT


(Visited 9 times, 1 visits today)
No tags for this post.