Robots can now learn new tasks simply by watching a person, and copying them.
The breakthrough by Israeli startup DLR (Deep Learning Robotics) means people with no robotics knowledge or programming skills can now teach robots to do new things.
DLR unveiled its robot control software for the first time at last week’s Consumer Electronics Show in Las Vegas.
Its advanced machine learning algorithms enable robots to learn by observing and mimicking human actions, eliminating the need for complex instructions.
It uses artificial intelligence that combines computer vision, deep learning, and motion control algorithms to perceive the environment as a 3-D map in real time.
It can differentiate between a human hand and the objects being handled, so people’s demonstrations are automatically translated into robot control language.
The algorithm can adapt to a wide range of robots and applications, from industrial manufacturing to home automation.
While robots are currently used to perform the same task thousands of times, DLR’s robotic controller lets them perform different tasks without needing to program them.
The company says that automation with robots can now become plug and play.
“We are excited to introduce this cutting-edge technology to the world,” said Aviv Vana, Marketing Director of DLR.
“Our goal is to make it easy for anyone to unlock the full potential of robots, regardless of technical expertise. This revolutionary software will greatly expand the use and application of robots in various industries.”