The Robot Autonomy and Interactive Learning (RAIL) research lab, directed by Prof. Sonia Chernova, focuses on the development of robotic systems that operate effectively in complex human environments, adapt to user preferences and learn from user input. The development of robots that work cooperatively with people is of critical importance for furthering advancements in manufacturing, medicine, healthcare, defense, and consumer applications. Central to this goal is the development of technologies that are adaptable to changing task and user needs. Our research spans the fields of robot learning, adjustable autonomy, semantic reasoning, multi-robot teaming, human-robot interaction and cloud robotics.


RAIL Lab

Pictured above is one of our research spaces, which enables remote users to control and teach our robots in real time in a real-world environment. Using a web-based interface, users are able to visualize the robot's environment and its internal state information, provide both low level and high level input for mobile manipulation tasks, as well as interactively teach the robot. This capability is enabled through the integration of two key open source packages, the Robot Management System (RMS), a backend system for remote user study management, and Robot Web Tools, a collection of software packages and the rosbridge protocol for building web-based robot apps (collaboratively developed with Brown University and industry partners).

-

Recent Happenings and Highlights

[11/2014] Adrian Boteanu's paper on"Solving and explaining analogy questions using semantic networks" has been accepted to AAAI'15.

[10/2014] Anahita Mohseni-Kabir, Sonia Chernova, Charles Rich, Candy Sidner, and Daniel Miller, "Interactive hierarchical task learning from a single demonstration" to appear in HRI 2015.

[05/2014] Sonia Chernova and Andrea Thomas have a new book on Robot Learning from Human Teachers! Check it out here!

[04/2014] Sonia Chernova receives ONR Young Investigator Award for work on "Multi-Representational Learning from Demonstration through Sequential User Study Development".

[09/2013] Sonia Chernova and Andrea Thomaz (GTech) receive NSF NRI award for work on "Learning from Demonstration for Cloud Robotics".

[08/2013] Sonia Chernova receives ONR award for work on "Collaborative Robot Learning from Demonstration Using Hierarchical Task Networks and Attributive Motion Planning" in collaboration with Charles Rich, Candace Sidner and Dmitry Berenson at WPI.


WPI Logo

© 2013 RAIL Suffusion theme by Sayontan Sinha