Momotaz Begum
University of New Hampshire, NH·Computer Science
Many older adults want to keep living in their own homes, and more people are acquiring disabilities earlier in life. Institutional care is expensive, so robots that help at home are becoming an attractive option.
- Fields
Computer Science
Engineering
- Topics
learning from demonstration
LfD
imitation learning
- Grants
- 1 known grant, through Aug 2028
What they work on
- Develop learning-from-demonstration (LfD) algorithms that learn a task's safety features from human video demonstrations and filter out demonstrations that would teach the robot unsafe behavior.
- Extend optimal control theory to learn, from few examples, feedback-based robot policies that reliably reach their goal despite changes in the environment during a task.
- Create and validate a questionnaire measuring how safe end users feel after using a robot trained with safe LfD.
- Validate safe-LfD robots for aging-in-place tasks in a simulated home and in community settings, gathering data on real-world use.
More about the research
Every person's needs and every home are different. A useful home robot would therefore need to learn new tasks from the people around it. Those people include elderly individuals and their caregivers, who are not programmers. Learning from demonstration (LfD) lets a robot learn a task by watching a person do it. Current LfD methods, however, do not guarantee that the robot learns safe behavior, and they rarely work outside simulations and controlled labs.
Why it matters
Older adults who wish to age in their own homes, along with their caregivers, could benefit from assistive robots they can safely teach new care tasks without expert help. Safer learning algorithms and a validated measure of how safe users feel could move home robots toward reliable real-world use. Data from community use could also show which problems appear only outside the lab.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Cagdas Onal
Worcester Polytechnic Institute, MA·Engineering
Develop a modular, origami-inspired continuum robot arm (CoBRA) with customizable design algorithms, so the arm can be tailored to specific assistive tasks in homes and healthcare settings.
Jaime Fernández Fisac
Princeton University, NJ·Engineering·Early-career
Develop a framework based on dynamic game theory in which a robot guarantees safety against the worst case within an Operational Design Domain (ODD), the conditions designers or regulators say it must handle, while modeling how agents gain information over time.
Shuran Song
Stanford University, CA·Computer Science·Early-career
Develop hierarchical task representations, built in hyperbolic space (a tree-like curved geometry), learned from large-scale unlabeled instructional videos to model the steps of everyday activities such as cooking and repair.
The school
University of New Hampshire: 5 labs on PI Pages, $230M a year on research, 79 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.