Initially, say the researchers, their robot does not know what its shape is. However, after a brief period of "babbling," and within about a day of intensive computing, their robot creates a self-simulation and then is able to use it to perform tasks and detect self-damage.
This, say the researchers, is not unlike how humans are able to imagine themselves, and acquire and adapt their self-image over their lifetime. Until now, most robots still learn using human-provided simulators and models, or by laborious, time-consuming trial and error.
"But if we want robots to become independent," says Hod Lipson, professor of mechanical engineering, and director of the Creative Machines lab, "to adapt quickly to scenarios unforeseen by their creators, then it's essential that they learn to simulate themselves."
For their study, the researchers used a four-degree-of-freedom articulated robotic arm. Initially, it moved randomly and collected approximately one thousand trajectories, each comprising one hundred points. The robot then used deep learning - a machine learning method based on learning data representations as opposed to task-specific algorithms - to create a self-model.
The first self-models, say the researchers, were quite inaccurate, and the robot did not know what it was, or how its joints were connected. But after less than 35 hours of training, the self-model became consistent with the physical robot to within about four centimeters.
The self-model performed a pick-and-place task in a closed-loop system that enabled the robot to recalibrate its original position between each step along the trajectory based entirely on the internal self-model. With the closed-loop control, the robot was able to grasp objects at specific locations on the ground and deposit them into a receptacle with 100% success.
Even in an open-loop system, which involves performing a task based entirely on the internal self-model, without any external feedback, the robot was able to complete the pick-and-place task with a 44% success rate, say the researchers.