The award is the MAST CTA, the Micro Autonomous Systems and Technology Collaborative Technology Alliance, and will incorporate more than $22 million in funding during 10 years from the Army Research Laboratory.
Penn will lead the multi-university research initiative tasked with creating cohesive teams of autonomous robots that can function effectively in urban and complex terrain. Participating universities will build new robotic systems, study novel approaches for robotic control and perception and develop new algorithms for intelligent operation of robots in unstructured environments. The goal is autonomous machines that operate with little or no direct human supervision and can support security or rescue personnel operating in dangerous environments. In venues as disparate as buildings and caves, the machines must be able to organize into sub-teams and clear and secure areas, track hostile targets in a three-dimensional environment and find victims or explosive devices by crawling, climbing, flying or hovering.
Vijay Kumar, chair of the Department of Mechanical Engineering and Applied Mechanics, will lead the consortium, consisting of the University of California, Berkeley; Georgia Institute of Technology; the University of New Mexico; Massachusetts Institute of Technology; the Australian Center for Field Robotics; Vanderbilt University; and the University of Maryland. Using expertise in algorithms, artificial intelligence, communications, control theory, perception, robotics, sensing, signal processing and systems engineering, the Penn program, Autonomous Multifunctional Mobile Microsystems, will design a networked group of mini aerial and ground vehicles and sensors operating in dynamic environments. The project will look to nature for inspiration to create these machines and develop intelligent group behaviors that allow the network to perform tasks that cannot be performed individually.
Scaling down today’s robots to tomorrow’s micro-robots raises many technical challenges including perception and control with small sensors, actuators and processors; communication between members of the team with low-power antennas; and the development of intelligent group behaviors with small-scale, resource-constrained robots.
“Our goal is to combine scientific principles with new engineering technologies to make autonomous aerial and ground robots work together, work independently, adapt, survey and ultimately become a reality in the field,” said George Pappas, professor of electrical and systems engineering and deputy dean Penn Engineering.
The team consists of Kumar; Pappas; Kostas Daniilidis, director of Penn’s GRASP Laboratory of Robotics Research and Education; and Ali Jadbabaie, Mark Yim and Dan Lee of the GRASP Laboratory.