An international team of researchers is using AI to train a robotic dog to respond to its owner’s requests.
The project is supported by a one-year seed grant from the Institute for Future Technologies (IFT), a partnership between New Jersey Institute of Technology (NJIT) and Ben-Gurion University of the Negev (BGU).
Assistant Professor Kasthuri Jayarajah of NJIT’s Ying Wu College of Computing is researching how to design a social assistance model for the Unitree Go2 robot dog that dynamically adapts its behavior and interactions based on the characteristics of the people it interacts with.
The overall goal is to bring dogs “alive” by employing wearable-based sensing devices that can detect physiological and emotional cues specific to the owner’s personality and characteristics, or even temporary states such as pain or comfort levels.
The invention could help combat loneliness among older people and could also be useful in treatment and rehabilitation.
Co-principal investigator Shelly Levy-Tzedek, an associate professor in BGU’s School of Physiotherapy, is an experienced researcher and leader in rehabilitation robotics, with a focus on studying the effects of aging and disease on body control.
The researchers note that as wearable devices become increasingly accessible, everyday models such as earphones can be reused to extract information about the wearer’s brain activity, subtle facial expressions, and other states.
The goal of this project is to combine wearable and traditional robotic sensors to objectively and passively track user attributes.
While socially assistive robots show promise, cost and computing limitations make their continued use over the long term difficult, Jayarajah said.
The researchers added: “Robots like the Unitree Go2 are not yet up to large AI tasks.
“Compared to large GPU clusters, they have limited processing power, not as much memory, and limited battery life.”
Image: New Jersey Institute of Technology
