The Information Science Colloquium welcomes Dr. Bradley Hayes, a postdoctoral associate in the Interactive Robotics Group within the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology. Brad's research interests center around developing the algorithms necessary for building supportive, interactive, and intuitive robotic systems that are capable of performing complex collaborative tasks in environments shared with humans. His work combines theoretical advances and practical applications of machine learning, task and motion planning, human teaming psychology, and human-robot interaction.

Brad received his Ph.D. in Computer Science from Yale University in 2015, advised by Brian Scassellati. His work has been featured on TedXCambridge, Phys.org, CBC News, Wired, the BBC, Popular Science, MIT Technology Review, and the Boston Museum of Science.

TitleLearning to be a good teammate: Algorithms for Efficient Human-Robot Collaboration

Abstract: Robots capable of fluent collaboration with humans will bring transformative changes to the way we live and work. In domains ranging from healthcare to education to manufacturing, particularly under conditions where modern automation is ineffective or inapplicable, human-robot teaming can be leveraged to increase efficiency, capability, and safety. Despite this, the deployment of collaborative robots into human-dominated environments remains largely infeasible due to the myriad challenges involved in creating helpful, safe, autonomous teammates.

In this talk I will present my recent work in overcoming these challenges, toward realizing flexible, communicative robot collaborators that both learn and dynamically assist in the completion of complex tasks through the application of novel learning and control algorithms. In particular, I will cover approaches to hierarchical task modeling, task and motion planning, and cooperative inverse reinforcement learning within the theme of human-robot teaming, focusing on the interpretable learning, synthesis, and execution of supportive behaviors. I will conclude with insights gained from real-world deployments of such systems and a perspective on future directions for human-robot teaming research.

 

Click Here to Watch Video