World’s first robot that has a brain and thinks
Japanese scientists have developed the world’s first “thinking” robot that can tilt its path and avoid walls and obstacles.
The robot has brain-like nerve cells that were grown in a laboratory. When these cells are electrically stimulated, the device successfully navigates through small mazes. The researchers say a method called “physical reservoir computing” enables the device to sense brain waves and avoid obstacles. “These neurons, or nerve cells, are grown from living cells,” he said.
“They behave like the physical memory of a computer to create coherent messages.
This is the first time intelligence has been “trained” into robots. As the robot moves along a spiral path, information about the environment is transmitted to it.
If an AI device turns or goes in the wrong direction, an electric shock is delivered to the nerve cells that have been killed. In a series of experiments, the device constantly received messages until it successfully completed the task.
The round device – about 3 inches (7.5 cm) in diameter and 2 inches (5 cm) high – fits in the palm of your hand. Has. ”
This research opens the door to designing devices that can solve problems with human-like thinking.
“These results show that goal-directed behaviour can be generated without additional learning by sending a message to the physical system,” said Professor Takahashi, a computer engineer at the University of Tokyo. “It’s not black and white, but relies entirely on test pulses and electrical errors.”
Robots are expected to take over much of the work now done by humans in the next 50 years.
“In an experimental run, live culture [cells] were put into the carrier robot to help it find a way to pass through the helix,” Professor Takahashi said.
He is inspired by the idea that intelligence arises from a mechanism that creates coherence from a chaotic situation.
Using this principle, the robot creates a “memory” of information that helps it understand and solve the problem.
According to Professor Takahashi, “The brain of an elementary school student is not able to solve the mathematical questions of the university entrance exam, probably because the dynamics of the brain or the calculation of the physical reservoir is not sufficient.”
“The ability to do things is determined by the richness of the set of peripheral patterns that the network generates.”
The Japanese group hopes this success will lead to the development of supercomputers that can adapt to the human brain.
The study also hopes to shed light on new dimensions of brain mechanism and the cause of diseases such as Alzheimer’s and Parkinson’s.