The Hasegawa Group located at the Tokyo Institute of technology has been able to integrate artificial intelligence into a semi-humanoid robot enabling it to think, learn, and act by itself. The robot uses a self-replicating neural network to learn in much the same way as its human counterparts. It is able to extrapolate based on its past experience and knowledge combined with observation of its surrounding environment, then make educated guesses. In that sense, it learns from its mistakes, just like a child, and consistently corrects and adjusts until it can successfully perform required tasks.
Self Organizing Incremental Neural Network (SOINN) applied to robotics-
According Osamu Hasegawa, an associate professor with the Imaging Science and Engineering Lab,
"So far, robots, including industrial robots, have been able to do specific tasks quickly and accurately. But if their environment changes slightly, robots like that can't respond. This robot remembers only basic knowledge, and it can apply that knowledge to its immediate situation. If it doesn't know enough, it stops, and reacts by saying, "I can't do this because I don't know how." So, 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."
Since the robots core intelligence has been implemented using the SOINN neural network, it packs considerable processing power in a small, fast package while still being able to draw on sensory data streams like vision, hearing, and touch, as well as information on the internet. Unlike imperfect humans, it is also able to share copies of its knowledge and experience with other similarly equipped robots. While people have to struggle with words, grammar, context, and language differences, the new robot system can communicate perfectly with its robotic partners, which could eventually enable its accumulated knowledge base and capability to expand at a tremendous rate.
Check out the DigInfo website for the full story.