Abstract
Over the last 2 years, the DARPA Robotics Challenge (DRC) has driven the development of robot systems and software capable of assisting humans in responding to natural and man-made disasters – a problem that demands beyond-state-of-the-art approaches to manipulation, locomotion, perception, and human-robot interaction (HRI). The December 2013 DRC Trials demonstrated robots capable of manipulating doors and valves, cutting through walls, climbing ladders, driving vehicles, and traversing challenging terrain, all aided by a remote operator using only feedback from the robot sensors suite over a limited, field-realistic communications link. This talk will describe the design considerations, implementation and performance of the software that Team MIT is developing to command and control the Atlas humanoid robot built by Boston Dynamics. Our design emphasizes human interaction with an efficient motion planner, where operators expressed desired robot actions in terms of affordances fit using perception and manipulated in a custom user interface. Ongoing progress towards the June 2015 DRC Finals will also be presented.
Short Bio
Maurice Fallon is a Research Scientist in the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT. His research is focused on probabilistic methods for localization and mapping. In particular, he is the perception lead on MIT’s Darpa Robotics Challenge team – a multi-year competition developing technologies for semi-autonomous humanoid robotic exploration and manipulation in disaster situations. From 2004-2008 he completed his PhD studies at the University of Cambridge where he studied multi-target localization with acoustic sensors. In 2015 he will take up a Chancellor’s Fellowship and Lectureship at the Department of Informatics, University of Edinburgh.