Facebook Is Teaching Robots to Adapt to Any Environment in Real-Time
We’ve seen robots that can dance, climb stairs, and even fetch beer, but they are usually programmed to do these tricks. Well, a team of researchers from Facebook (yes, you read that right) has developed a new model for robotic locomotion that allows legged robots to adapt to any environment in a matter of seconds, without any prior knowledge.
Most legged robots usually achieve locomotion by being taught to navigate in different environments through a combination of hand-coding and learning techniques. What Facebook is proposing is a new method called Rapid Motor Adaptation (RMA). Researchers say that this is the first learning-based technology that allows a legged robot to adapt to its environment by exploring and interacting with its surroundings. Through this interaction, RMA employs end-to-end learning, including directly generating joint positions without the use of predefined leg motions or other prior knowledge. To simulate scenarios ahead of the real challenge, the team used something called “reinforcement learning” and “supervised lear
Most legged robots usually achieve locomotion by being taught to navigate in different environments through a combination of hand-coding and learning techniques. What Facebook is proposing is a new method called Rapid Motor Adaptation (RMA). Researchers say that this is the first learning-based technology that allows a legged robot to adapt to its environment by exploring and interacting with its surroundings. Through this interaction, RMA employs end-to-end learning, including directly generating joint positions without the use of predefined leg motions or other prior knowledge. To simulate scenarios ahead of the real challenge, the team used something called “reinforcement learning” and “supervised learn
