1st International ICST Conference on Ambient Media and Systems

Research Article

Multi-agent Designs for Ambient Systems

Download510 downloads
  • @INPROCEEDINGS{10.4108/ICST.AMBISYS2008.2925,
        author={Kendall E. Nygard and Dianxiang Xu and Jonathan Pikalek and Martin Lundell},
        title={Multi-agent Designs for Ambient Systems},
        proceedings={1st International ICST Conference on Ambient Media and Systems},
        publisher={ICST},
        proceedings_a={AMBI-SYS},
        year={2010},
        month={5},
        keywords={Architecture description languages},
        doi={10.4108/ICST.AMBISYS2008.2925}
    }
    
  • Kendall E. Nygard
    Dianxiang Xu
    Jonathan Pikalek
    Martin Lundell
    Year: 2010
    Multi-agent Designs for Ambient Systems
    AMBI-SYS
    ICST
    DOI: 10.4108/ICST.AMBISYS2008.2925
Kendall E. Nygard1,*, Dianxiang Xu2,*, Jonathan Pikalek3,*, Martin Lundell4,*
  • 1: North Dakota State University Dept. of Computer Science Fargo, ND 58105-5164 1-701-231-8203
  • 2: North Dakota State University Dept. of Computer Science Fargo, ND 58105-5164 1-701-231-8185
  • 3: North Dakota State University Dept. of Computer Science Fargo, ND 58105-5164 1-701-231-8562
  • 4: University of Minnesota Crookston Dept. of Math, Sci., and Technology Crookston, MN 56716 1-218-281-8180
*Contact email: Kendall.Nygard@ndsu.edu, Dianxiang.Xu@ndsu.edu, Jonathan.Pikalek@ndsu.edu, mlundell@umn.edu

Abstract

Designing and developing software for an ambient intelligence (AmI) system involves difficult challenges related to the varied roles of many heterogeneous devices and communication channels, and intelligent user interfaces. Because ambient systems have unpredictable requirements and are context-aware, software designs must support dynamic and sustainable change. We argue that such designs should utilize formal methods and aspect-oriented techniques, to help in supporting model validation and verification. Features of an aspect-oriented, multi-agent, architectural description language are presented as a mechanism for reasoning about cross-cutting concerns.