Environmental Engineering Reference
In-Depth Information
This project is highlighted because it is NASA's showcase for new tech-
nology, and the software technologies are truly revolutionary. The project
will demonstrate substantial autonomy in space-based missions with the goal
to establish a virtual presence in space. Cooperative autonomy technolo-
gies [
22
,
39
,
43
,
108
,
194
] could augment the DS1 autonomy architecture and
further this goal.
7.6.1 The Mobile Robot Laboratory at Georgia Tech
The Georgia Tech Mobile Robot Laboratory (MRL) has been working on the
fundamental science and current practices of intelligent mobile robot systems.
The MRL has the goal of facilitating the technology transfer of their research
results to real world problems.
Many of the MRL projects should be of interest to those attempting to set
up a cooperative autonomy laboratory. The MRL has studied online adaptive
learning techniques for robotic systems that allow robots to learn while they
are actively involved in their operating environment. This type of learning is
intended to be fast, similarity-based, and reactive. The MRL has also studied
oine learning where the robot system reasons deeply about its experiences
and learns as a result of this analysis. This type of learning is intended to be
slower, case-based and explanation-based, deliberative, and goal-oriented.
Georgia Tech has pursued autonomous vehicle research projects supported
by the Defense Advanced Research Projects Agency (DARPA). One effort
mixed autonomous robot behavior with human controllability. Other research
addressed multiagent systems that achieve tasks in the context of hostile en-
vironments.
Georgia Tech has developed several software packages that allow users
to create robot control architectures for a specific domain and then test the
control architecture in a simulated robot environment. One is written in Java
and is designed to be portable. The simulation environment is compatible with
off-the-shelf robotic hardware and allows the control architecture developed
in the simulator to be run directly on a physical robot.
7.6.2 Cooperative Distributed Problem Solving Research Group
at the University of Maine
The University of Maine's Cooperative Distributed Problem Solving Research
group is centered on determining and devising the features, characteristics,
and capabilities that are sucient to bring about collaboration among groups
of autonomous and semiautonomous agents toward the accomplishment of
tasks. Their work has involved underwater robots, which share many of the
same challenges as spacecraft:
•
Operations in hostile environments
•
Self-contained operations
