The robot watched as Shikhar Bahl opened there frigerator door. It recorded his movements, the swing of the door and more,analyzing this data and readying itself to imitate(模仿)what Bahl had done. It failed at first, missing the handle completely at times, grabbing it in the wrong spot or pulling it incorrectly. But after a few hours of practice, the robot succeeded and opened the door.
"Imitation is a great way to learn, "said Bahl, a Ph. D. student at the Robotics Institute (RI) in Carnegie Mellon University. " Having robots actually learn from directly watching humans remains an unsolved problem in the field, but this work takes a significant step in enabling that ability."
Bahl worked with Deepak Pathak and Abhinav Gupta, both faculty members in the RI, to develop a new learning method for robots called WHIRL, short for In-the-Wild Human Imitating Robot Learning. WHIRL is an efficient computation program for visual imitation. People constantly perform various tasks in their homes. With WHIRL, a robot can observe those tasks, gather the video data it needs and then go about practicing and learning to accomplish the tasks on its own.
The team added a camera and their software to an off-the-shelf robot, and it learned how to do more than 20 tasks—from opening and closing appliances, cabinet doors and drawers to putting a lid on a pot, pushing in a chair and even taking a garbage bag out of the bin.
Current method for teaching a robot a task typically relies on reinforcement (强化) learning. In reinforcement learning, the robot is typically trained on millions of examples in imitation and then asked to adapt that training to the real world.
This learning model works well when teaching a robot a single task in a structured environment, but it is difficult to extend and deploy (调动). WHIRL can learn from any video of a human doing a task. It is capable of being easily expanded, not limited to one specific task and can operate in realistic home environments.