Welcome
Digital library of construction informatics
and information technology in civil engineering and construction
 

Works 

Paper itaec-2004-9:
AVSAR: a collaboration system for disaster search and rescue operations using autonomous vehicles

Facilitated by the SciX project

Pe?a-Mora F, Craig M

AVSAR: a collaboration system for disaster search and rescue operations using autonomous vehicles

Abstract:The disaster relief community is increasingly focused on issues of critical physical infrastructure in search and rescue operations. As the disaster relief and civil engineering community attempts to expand its abilities in this arena, it is being confronted with constraints related to manpower, risks to human personnel, and system stability. The community can address these barriers by integrating autonomous vehicles and intelligent software agents into its traditionally human elements. The military has been actively pursuing this goal in order to minimize human casualties and expand its functionality, and a technology transfer to the disaster arena would be greatly beneficial. The transition from the military to the disaster relief community is a logical step because of the great number of similarities between the two areas. Both are concerned with operations carried out in hostile, chaotic environments, where many participants from different areas of expertise collaborate to reach an objective, and both are constrained by the quality of intercommunication and the effectiveness of their equipment. Experience gained by the military in the field of autonomous vehicles has shown that while the ratio of autonomous vehicles to humans remains low, there is little trouble in directly controlling these vehicles as personnel can be dedicated to this task alone. However, as the number of autonomous vehicles increases to include personal human assistants and entire teams of vehicles, the task of control and collaboration becomes increasingly difficult. To date, most autonomous vehicle control work has been done with a one-to-one structure where one human controls one vehicle. While this works well when the vehicles are relatively simple and the number of vehicles is small, it does not translate well into the ideal situation of large populations of complex autonomous vehicles. Under these circumstances, intelligent software agents, residing both on the autonomous vehicles and on the communication devices, are needed to handle the task of distributed decision-making. This autonomous decision making ability is particularly critical for the cases where the autonomous vehicles fall out of contact with their human commanders or remote experts such as geotechnical, structural, and earthquake engineers. This paper examines past work done for and by the military in the area of autonomous vehicle systems and examines its application to the field of disaster relief involving critical physical infrastructures. It then presents a system that meets the needs of a combined human - intelligent software agent - autonomous vehicle SR (Search and Rescue) team, operating on critical physical infrastructure in an unstable and hostile environment. The collaboration infrastructure includes an information policy layer and a client application layer that address the need for inter-user communication and flexible command structures, which can be dynamically arranged to meet the situational need.

Keywords:collaborative environments, disaster relief, search and rescue, autonomous vehicles, intelligent software agents, self-organization, control structures, information policy

Full text:content.pdf (1,079,294 bytes) (available to registered users only)

Series:itaec:2004 (browse)
Cluster:
Class:
Similar papers:
Sound:N/A.

 

hosted by University of Ljubljana



includes

W78




© itc.scix.net 2003
Home page of this database login Powered by SciX Open Publishing Services 1.002 February 16, 2003