Unlike conventional robots, soft robots are far more flexible and can easily adapt to their environments. Because they have these properties, their behaviors are difficult to study and control. We first design and build robots made out of silicone rubber. We then actuate them in order to observe how their morphologies affect their behaviors, tracking their movements with Qualisys Motion Capture cameras. This process allows us to gather their tracking data. We then run the data against a Quality Diversity Algorithm that can autonomously generate a behavioral repertoire of diverse and efficient behaviors so that our robots can be efficient and autonomous.
Presenting
Primary Speaker
Hyojin Han
Faculty Sponsors
John Rieffel
Abstract Details
Presentation Type
Faculty Department/Program
Faculty Division
Do You Approve this Abstract?
Approved