Osamu Hasegawa, a professor at Tokyo Institute of Technology, has developed a system that allows robots to solve problems by themselves correlations from data taken from the environment and from internet.
"The Self-Organising Neural Network Incremental" or SOINN is an algorithm that allows robots to use their previous knowledge to carry out tasks.
Thus, SOINN is examining the environment to gather data they need in order to organize the information as a coherent set of instructions.
In a laboratory demonstration, when the robot was required to "serve water," he began to divide the task into a series of previously learned skills: how to hold a cup, how to take a bottle, pour water from a bottle in glass and how to put a glass down. Without having a special program related to water servant, alone finds the order in which must perform the tasks. SOINN determines that a task is beyond his abilities, ask for help and learn, by storing data for the next time to carry out the action.
Hasegawa's team goal is to try to merge these capabilities, creating a robot that performs searches on the internet to learn how to do certain actions.
Source: News




