Home News This robot dog just taught itself to walk

This robot dog just taught itself to walk


The team’s algorithm, called Dreamer, uses past experience to build a model of the world around it. Dreamer also allows robots to perform trial-and-error calculations in a computer program, rather than in the real world, by predicting potential future outcomes of their potential actions. This makes it faster to learn than purely by doing. Once the robot learns to walk, it continually learns to adapt to unexpected situations, such as resisting being pushed over by a stick.

“Teaching robots by trial and error is a difficult problem, which is made even more difficult by the lengthy training that such teaching requires,” said Lerrel Pinto, an assistant professor of computer science at NYU who specializes in robotics and machine learning. Dreamer shows that deep reinforcement learning and world models can teach robots new skills in a fraction of the time, he said.

The findings, which have not yet been peer-reviewed, clearly show that “reinforcement learning will be a cornerstone tool for the future of robotic control,” said Jonathan Hurst, a professor of robotics at OSU.

There are many benefits to removing the simulator from robot training. Hafner said the algorithm could be used to teach robots how to learn skills in the real world and adapt to things like hardware failures — for example, a robot could learn to walk with a malfunctioning motor in one of its legs.

Stefano Albrecht, assistant professor of artificial intelligence at the University of Edinburgh, said the approach could also be used for more complex things, such as autonomous driving, which requires complex and expensive simulators. A new generation of reinforcement-learning algorithms could “learn super-fast how the environment works in the real world,” Albrecht said.

Source link

Previous articleA1 and Celebrity Big Brother star Ben Adams marries girlfriend Sara in romantic summer wedding
Next articleWill the Ranbir Kapoor-starrer ‘Shamshera’ bring good tidings for a struggling Bollywood?


Please enter your comment!
Please enter your name here