This project targets research on Human-Robot Interaction (HRI) between children with Autism Spectrum Disorder (ASD) and engineered, autonomous, facially-expressive, human-like robots that can assess and adapt to personalized therapeutic goals. Adaptive partner robots specifically targeting ASD hold the potential to revolutionize treatment and improve the lives and productivity of millions of Americans. We envision that the next generation ASD robot will be used primarily in the home, in addition to specialized treatment facilities. The envisioned robotic partners will serve as therapists, instructors, passive observers, or active learning peers, thus supplementing the role of important individuals: parents, caregivers, or teachers. The further exploration of the application of groups of robots and children in therapy will enrich and deepen our research understanding and could produce meaningful future outcomes for individuals with ASD and their families. The proposed interactions represent a unique opportunity to capture and objectively measure the natural behavior of ASD children in multiple contexts, irrespective of the age, language or cognitive ability, because it is based on both motor and emotional responses.
The research undertaken in this project has several novel components, including formulating new quantitative, non-invasive markers, eventually integrated into a new ASD scale for diagnosis and treatment based on real-time evaluation of motions and emotions during interaction between robot and child. The ASD severity scale will be based on a dynamic time-warping motion metric and on a continuous emotion metric assessed by the robot during interaction. The scale will be used by the robot to adapt therapeutic activities in a personalized manner for each child in one-one one therapy sessions, and for several children and robots in group therapy sessions. With the help of industry partners, we will embed the resulting novel motor and cognitive behaviors into a nonthreatening, partner robot, specifically designed for children with ASD, endowed with a friendly appearance, behavioral switching and neuroadaptive controllers. The robot will be capable of collecting large amounts of data that may be useful to elucidate unique connections between the motor, sensory and emotional brain functions of the ASD population.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.