Butterfly.
O'Reilly Emerging Technology Conference.
Books Safari Bookshelf Conferences O'Reilly Network
   

Arrow Home
Arrow Registration
Arrow Speakers
Arrow Keynotes
Arrow Tutorials
Arrow Sessions
Arrow At-a-Glance
Arrow BOFs
Arrow Community
Meetings
Arrow Events
Arrow Exhibitors
Arrow Sponsors
Arrow Hotel/Travel
Arrow Venue Map
Arrow See & Do
Arrow Press
Arrow Join Mailing List 


April 22-25, 2003, Santa Clara -Explore. Invent. Connect.
Butterfly.

Session

Swarming of Unmanned Air Vehicles (UAVs)
Paolo Gaudiano, Ph.D., Icosystem

Track: Nanotechnology and Hardware
Date: Friday, April 25
Time: 3:15pm - 4:00pm
Location: Lafayette/San Tomas/Lawrence

Unmanned Air Vehicles (UAV) swarms rely on local, distributed control and communication strategies of the type observed in insects and other natural groups. This leads to greater robustness than traditional, centralized control approaches, while yielding substantial reductions in code size and communication requirements.

These practical advantages make it possible to deploy UAV swarms with shortened development times and at a minute cost compared to other existing approaches.

Guadiano shows quantitative metrics to evaluate the effectiveness of the proposed approach and to allow for systematic comparison with other traditional approaches.

Guadiano also describes how to generate desired emergent behaviors through the manipulation of local control and communication rules. The same principles are also applicable to other types of swarms, such as wheeled or legged robots to be deployed in search-and-rescue missions, inspection and repair in hazardous areas, surveillance, and intelligence gathering.

Guadiano adds, "Our company, Icosystem Corporation, has received support from the Air Force Research Laboratory to study the problem of controlling large numbers of Unmanned Air Vehicles. We have been working on this project since June of 2002. We chose to work on this project because it combines Icosystem's focus on swarm intelligence, and because of my own background in robotics and control.

"The presentation might generate some controversy around the topic of using technology to increase US military superiority. Until recently, most people had never heard of UAVs. Much of this changed last year, with the news that a remote-controlled Predator UAV had been used to kill a suspected al Qaeda operative. The Predator and most other UAVs deployed by the military to date are controlled from a remote location by one or more operators. Hence the only difference between a UAV and a regular war plane is that the UAV operators are not risking their lives. What will happen when the military gains the ability to control swarms of UAVs? On the one hand, it has been shown that improvements in technology can dramatically reduce the likelihood of 'collateral' casualties. On the other hand, many people feel strongly against scientists developing technology that will be used for explicit military operations.

"However, not all military or government applications are bellic in nature. Unmanned Air Vehicles, and other types of Unmanned Vehicles in general, have been used for a variety of other missions. For instance, robots helped to search the ruins of the World Trade Center, and they can be used to clear minefields or to clean pipelines. Being able to control swarms, or teams of Unmanned Vehicles could lead to novel peacetime applications. For instance, the US Fisheries may want to use a swarm of UAVs to track schools of fish or whales in the ocean; teams of robots could be used to explore and clean up a hazardous site. Having the ability to control large teams with one or a few operators could yield tremendous time and cost savings.

"We believe that any system comprised of a large number of interacting elements can benefit from a 'swarm intelligence' approach. One of the core tenets of our work is that often a decentralized, bottom-up approach to controlling a system is much more effective than a traditional, centralized approach. We have applied these concepts to a variety of technology problems, such as distributed data storage in a computer network, and the creation and management of ad hoc wireless networks."


Download presentation file


O'Reilly Home | Privacy Policy

© 2002, O'Reilly Media, Inc.