Our robotic team for this project consists of two types of robots.
The first type is a larger, heavy-duty robotic platform,
called the ranger. It is used to transport
and deploy a number of small, mobile sensor platforms called
scouts, the second type of robot, into the environment.
Many reconnaissance and surveillance tasks require the use of
multiple small yet highly capable robots. The individual robots
must be easily deployable and able to move efficiently yet traverse
obstacles or uneven terrain. They must be able to sense their
environment, act on their sensing, report their findings, and accept
commands to operate in a coordinated manner.
We call these robots scouts.
Rangers add functionality to the team in two ways. First, using
their greater size and power, rangers add range and mobility to the team.
Rangers carry the scouts to the mission site and launch them into position
using a spring-powered ``gun.'' Second, using their greater computational
resources, rangers add intelligence to the team. Rangers do complex
perception (e.g., vision processing) on behalf of the scouts and coordinate
all of the robots in the team.
Together, the scouts and rangers form a hierarchical team capable
of carrying out complex missions in a wide variety of environments.
The original scout design was a ball.
The idea for the scout came out of earlier work I have done with
students for AAAI robotic competitions.
One of the tasks was to find and collect squiggle balls using autonomous
robots. Squiggle balls move straight, but change direction
in unpredictable ways when they encounter an obstacle. They have
no computer, just a set of batteries and an eccentric motor.
So, the idea was to replace an uncontrollable and fast moving ball
with a rolling robot ball and get the same ability to cover terrain fast.
The second idea was to increase the terrain coverage by launching the
robots with a launcher. This would allow them to reach farther away places.
No matter how a ball lands it is ready to roll and get its job done.
The project started from these two fundamental insights. The design of
the scout was subsequently modified to make it into a cylindrical shape,
so it could fit standard grenade launchers, but the basic idea remained
This project has been funded by DARPA/MTO and is currently funded by NSF.
You can find
more information about this project, including multiple videos.
Ryan Morlok and Maria Gini,
"Dispersing robots in an unknown environment," DARS 2004, Toulouse,
France, June 2004.
P. E. Rybski, S. A. Stoeter, M. Gini, D. F. Hougen, and N. Papanikolopoulos,
"Performance of a Distributed Robotic System Using Shared
IEEE Trans. on Robotics and Automation, special issue on
Multi-Robot systems, Vol 22, No. 5, pp 713-727, October 2002.
Paul E. Rybski, Sascha A. Stoeter, Michael D. Erickson,
Maria Gini, Dean F. Hougen, Nikolaos Papanikolopoulos,
"A Team of Robotic Agents for Surveillance",
Agents'2000, Barcelona, Spain, June 2000.
Dean F. Hougen, Saifallah Benjaafar, Jordan C. Bonney,
John R. Budenske, Mark Dvorak, Maria Gini, Howard French, Donald G. Krantz,
Perry Y. Li, Fred Malver, Brad Nelson, Nikolaos Papanikolopoulos,
Paul E. Rybski, Sascha A. Stoeter, Richard Voyles, Kemal Berk Yesin,
"A Miniature Robotic System for Reconnaissance and Surveillance",
IEEE International Conference on Robotics and Automation,
San Francisco, CA, April 2000.
Sascha A. Stoeter, Paul E. Rybski, Michael D. Erickson, Michael Wyman,
Maria Gini, Donald G. Krantz, Dean F. Hougen, Nikolaos Papanikolopoulos,
"A Robot Team for Exploration: Design and Architecture",
IAS-6, Venice, Italy, July 2000.
John Budenske, Architecture Technology Corporation
Don Krantz, MTS
Dean F. Hougen, postdoc
Ryan Morlok, undergraduate
Paul Rybski, PhD
Sascha A. Stoeter, PhD
Michael Wyman, undergraduate
Copyright: © 1999-2006 by the Regents of the University
Department of Computer Science and
Engineering. All rights reserved.
Comments to: Maria Gini