Oct 18 2017
An alliance led by the University of Pennsylvania has received a five-year, $27 million grant from the U.S. Army Research Laboratory to develop new methods for creating resilient, intelligent and autonomous teams of robots.
Magnus Egerstedt, executive director of Georgia Tech’s Institute for Robotics and Intelligent Machines, is shown in the new Robotarium. Egerstedt is among the Georgia Tech researchers participating in a new Army Research Laboratory grant. (Credit: Christopher Moore, Georgia Tech)
These teams, comprising of multiple types of sensors and robots with different abilities, are designed to help humans in a broad range of missions in dynamically contested, changing and harsh environments. These include humanitarian missions, search and rescue of hostages and information gathering after natural disasters or terrorist attacks.
The award is part of the Distributed and Collaborative Intelligent Systems and Technology (DCIST) Collaborative Research Alliance of ARL. Penn Engineering, in collaboration with the Army Research Laboratory, the Georgia Institute of Technology and Massachusetts Institute of Technology’s Aeronautics and Astronautics Department, will lead this alliance. The consortium also includes a faculty from University of Southern California, University of California Berkeley and University of California San Diego.
DCIST includes imbuing teams of heterogeneous sensors and robots with the intelligence to learn and adjust to different settings and execute new tasks along with humans. Building resilience to disruption is integral to this vision.
Teams of robots and human first responders may eventually be employed to survey a disaster site for victims; however, ongoing hazards and unpredictable environments could destroy or damage some of the robots, or interrupt communications between them. If all robots were just preprogrammed and given particular instructions, it could cause gaps in their search. However, if the team were able to reconfigure itself following damage, the remaining robots could jointly decide how to reorganize and operate with human partners in order to complete the mission.
We want to have teams of robots that know how to work together, but can figure out how to keep working even if some of their teammates crash or fail, if GPS signal is unavailable, or if cloud services are disrupted. This means designing networks with loose, flexible connections that can change on the fly. That way, a single event can’t bring down the entire network. More importantly, we want them to learn to perform tasks they may have never performed and work alongside humans that they may never have worked with.
Vijay Kumar, Penn Engineering’s Nemirovsky Family Dean and Director for the DCIST program
The three important focus areas of the research are distributed intelligence and learning; building resiliency in group behaviors; and developing a cohesive team of sensors, computational resources, autonomous robots and human experts.
Through this exciting project, Georgia Tech will help develop novel tools and techniques that enable human operators to work effectively and safely in teams together with autonomous robots. These types of questions connect well with our expertise in the areas of human-robot interactions, distributed decision making and learning, and swarm robotics.
Magnus Egerstedt, Executive Director of Georgia Tech’s Institute for Robotics and Intelligent Machines and Julian T. Hightower Chair in Systems and Controls
Beyond Egerstedt, the Researchers of Georgia Tech associated with this multidisciplinary project are Panagiotis Tsiotras, Dean’s Professor in the School of Aerospace Engineering; Justin Romberg, Associate Chair for Research and Schlumberger Professor in the School of Electrical and Computer Engineering; and Sonia Chernova, assistant professor in the School of Interactive Computing.
With different types of assets collectively evaluating a complex, continuously changing situation and determining how best to designate their individual skills to a widely defined problem, such human-robot teams of the future would be perfect first-responders to hazardous situations.
“The technology we’re working will better allow humans to respond by projecting their intelligence without directly coming in harm’s way,” Kumar said.