Maja J Matarić

Ph.D., Computer Science and Artificial Intelligence, MIT, 1994

Bio

Lab

Publications

Book

Courses

Outreach

CV

Movie

Student Contact Message

Research goals: My research is aimed at endowing robots with the ability to help people. I am inspired by the dual goals of 1) gaining novel insights into human behavior and cognition through human-machine interaction and 2) developing robotic systems capable of providing personalized assistance individually (in convalescence, rehabilitation, training, and education) or in teams (in habitat monitoring and emergency response). My Interaction Lab focuses on socially assistive systems capable of aiding people through social interaction rather than through physical contact. Our research involves intelligent human-robot interaction (HRI), control, and learning in complex, dynamic, and uncertain environments by integrating on-line perception, representation, and interaction with people. To address the inherently multidisciplinary challenges of this research area, our work draws on theories, models, and collaborations from neuroscience, cognitive science, social science, health sciences, and education.

Projects: The Interaction Lab projects focus on assistive human-robot interaction. Under the unbrella of socially assistive robotics, we also pursue projects in humanoid control and learning through interaction, imitation and demonstration; and activity modeling for individuals, small groups, and crowds. My earlier work focused on synthesis and analysis of adaptive group behavior for distributed multi-robot control and learning. More information about our research projects is found here. Interested students should look here.

Methodology: With our reseach focus on real-time assistive human-robot interaction, my group's research efforts are brought together under the behavior-based methodology, using the notion of composable basis behaviors or primitives as building blocks for structuring complex systems and reducing the dimensionality of the associated control and learning problems. In all projects, we evaluate our models and methodologies on complex embodied agents and robots and faithful simulations, and conduct human-subjects experiments using rigorous experimental design and evaluation.

Contact Information