COLLOQUIUM

Fluid Coordination of Human-Robot Teams

Julie Shah, MIT

Computer Science and Robotics Engineering

Faculty Candidate

Friday, February 26, 2010
11:00 a.m. - 12:00 p.m.
Fuller Labs 320

Abstract:

Collaboration between humans and robots is becoming indispensible to our work in many high-intensity domains, ranging from surgery to space exploration. To harness the relative strengths of humans and robots, we must develop robots that seamlessly integrate with human group dynamics. Robots should preserve the essential qualities of a good human partner by robustly anticipating and adapting to other team members and avoid constraining their human partners’ flexibility to act. The robot partner must be capable of reasoning quickly online, and adapting to the humans’ actions in a temporally fluid way.

In this talk, I present a capability named Chaski that enables a robot to work with a human teammate under a flexible plan containing choices in the task assignment and timing of activities. Chaski generalizes the state-of-the-art in dynamic plan execution to provide a powerful framework for explicitly modeling and efficiently reasoning on temporal information for human-robot interaction. This capability is efficiently realized by an incremental algorithm that reasons on perturbations over possible futures. Chaski enables a human and robot to work together under different models of teamwork: as Equal Partners and as Leader & Assistant. I develop models that distinguish these two styles of teamwork based on the predictability of the partner. Finally, I present recent work applying Chaski to perform multi-manipulator coordination using two Barrett Whole Arm Manipulators, and describe ongoing work to demonstrate temporally fluid human-robot teaming using the Mobile-Dexterous-Social (MDS) robot.

______

Julie Shah is a PhD candidate in the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology (MIT). Her research interests include mixed-initiative human-robot collaboration, multi-agent coordination, dynamic plan execution under uncertainty, and temporal reasoning.

Host: Prof. Michael Gennert

Refreshments will be served.

Maintained by webmaster@cs.wpi.edu
Last modified: = 02/22/2010