Seed Grant Profile
Robust People Following via RFID for Assistive Mobile Robots
The proposed research will advance the state of the art in robotics and assistive technology by developing sensing technology and algorithms that will enable an assistive mobile robot to robustly follow a person with ALS in everyday settings.
Mobile companion robots that remain in close proximity to a specific person would be a valuable platform for personalized health care services such as non-invasive monitoring and assistive manipulation. Currently, robots are unable to robustly follow a specific user through everyday situations and environments due to sensing limitations. In particular, robots have great difficulty detecting and recognizing specific people, which is likely to lead to errors when multiple people are in the presence of the robot. We will address this fundamental challenge by developing novel sensing technology that makes use of an RFID wristband worn by the user, and novel robotic behaviors that make use of this information to unambiguously identify and follow a specific user. This system will function by tracking the data signal emitted from the RFID wristband to provide both identifying information and information about the location of the user relative to the robot. As a proof of concept, we will develop and test this technology in the context of an assistive mobile manipulator for people with ALS.
Investigators: Charles Kemp (GT, Robotics), Matt Reynolds (GT, Computing), Jonathan Glass (Emory, Neurology)