Robots can perform repetitive tasks with a high degree of precision. Change some aspect of its environment and the bot will not adapt, and errors will soon follow. Researchers from Hasegawa Groups Labs, at the Tokyo Institute of Technology, have created a bot that can make decisions based on prior data in a dynamic environment.
The experiment was to pour a glass of water, add ice, and place it for a person. While being away of its surroundings and situation, the bot had to decide on what order to do all the operations. Instead of teaching the bot what to do the researchers said, "if you teach this robot just the things that it can't do, it incorporates those things as new knowledge, and it can solve the problem overall, by including that knowledge." Not only does the bot have input through visual, auditory, and tactile data, it can also connect to the internet and other robots for more data on its require task. The whole system allows the robot to learn and adapt in an ever changing world, much like a human being. The system developed is called SIONN (Self-Organizing Incremental Neural Network).
The team gives an example of how it could make something it never has before, "For example, suppose this robot doesn't know how to make tea, and it's sent to an elderly person who lives alone. And suppose that person asks it to make a cup of green tea. The robot doesn't know how, so it asks robots around the world how to make tea. Suppose, for example, that a robot in the UK tells it how to make British-style tea. We think this robot will become able to transfer that knowledge to its immediate situation, and make green tea using a Japanese teapot."
Eavesdropper