Pieter Abbeel, Professor of Electrical Engineering and Computer Science at the University of California, Berkeley and students have a new start-up, Embodied Intelligence Inc., which uses deep reinforcement learning and artificial intelligence to easily teach industrial robots new, complex skills. Using a VR headset the robot is virtually guided through a task, much like manipulating the arms of a puppet. With this technique a robot can be trained to a new task in a day, rather than requiring weeks to months to write new computer code to reprogram it.
“Right now, if you want to set up a robot, you program that robot to do what you want it to do, which takes a lot of time and a lot of expertise,” said Abbeel. “With our advances in machine learning, we can write a piece of software once—machine learning code that enables the robot to learn—and then when the robot needs to be equipped with a new skill, we simply provide new data.It completely changes the turnaround time because the amount of data you need is relatively small. You might only need a day of demonstrations from humans to have enough data for a robot to acquire the skill.”
The startup is manned by Abbeel, President and Chief Scientist), and graduate students Peter Chen (CEO), Rocky Duan (CTO), and Tianhao Zhang. It has already raised $7 million in seed funding since its founding in September.
“This is an amazing capability that we just developed here at UC Berkeley, and we decided we should put this into the world and empower companies still using techniques that are many years behind what is currently possible,” Abbeel said. “This will democratize access to robotic automation.”