The team from Disney Research have made it possible for robots to learn how to move from random motion data.
They have achieved this by using a machine learning technique to break down short pieces of motion data and create a simpler… pic.twitter.com/rEFj19VS95
They have achieved this by using a machine learning technique to break down short pieces of motion data and create a simpler version of the movements.
Then, they use this simplified version to train a control system that can take movement instructions and produce realistic motions that work in the real world.
The ability to train robots to perform a wide range of motions, including unseen and complex ones, could impact industries like manufacturing, healthcare, and service robotics. Robots could handle more dynamic tasks, adapt to new environments, and interact more naturally with humans.
Science fiction movie makers have seen robots dancing for a long time. In his 1927 film Metropolis, Fritz Lang gives a robot the personality and likeness of a young dancer; the resulting robot dances for the upper caste men of the city.
RoboShiko! Sumo Exercises Still Good For Robots
'... the expressionless face before me was therefore that of the golem-wrestler, Rolem, a creature that could be set for five times the strength of a human being.' - Roger Zelazny, 1966.
Technovelgy (that's tech-novel-gee!)
is devoted to the creative science inventions and ideas of sf authors. Look for
the Invention Category that interests
you, the Glossary, the Invention
Timeline, or see what's New.
Smart TVs Are Listening!
'You had to live -- did live, from habit that became instinct -- in the assumption that every sound you made was overheard...'
RoboShiko! Sumo Exercises Still Good For Robots
'... the expressionless face before me was therefore that of the golem-wrestler, Rolem, a creature that could be set for five times the strength of a human being.'