3rd EAI International Conference on Management of Manufacturing Systems

Research Article

ML/AI in Intelligent Forest Fire Observer Network

Download159 downloads
  • @INPROCEEDINGS{10.4108/eai.6-11-2018.2279681,
        author={Ljiljana Šerić and Darko Stipaničev and Damir Krstinić},
        title={ML/AI in Intelligent Forest Fire Observer Network},
        proceedings={3rd EAI International Conference on Management of Manufacturing Systems},
        publisher={EAI},
        proceedings_a={MMS},
        year={2018},
        month={12},
        keywords={artificial intelligence machine learning artificial perception forest fire detection},
        doi={10.4108/eai.6-11-2018.2279681}
    }
    
  • Ljiljana Šerić
    Darko Stipaničev
    Damir Krstinić
    Year: 2018
    ML/AI in Intelligent Forest Fire Observer Network
    MMS
    EAI
    DOI: 10.4108/eai.6-11-2018.2279681
Ljiljana Šerić,*, Darko Stipaničev1, Damir Krstinić1
  • 1: Universitiy of Split, FESB
*Contact email: ljiljana@fesb.hr

Abstract

This paper discusses advantages of using machine learning and artificial intelligence (ML/AI) techniques in forest fire observer. Observer network was proposed as a framework for artificial perception software system relying on sensor network deployed on monitoring locations and used for detection of phenomenon. The architecture of the system distinguishes three main parts where data is analyzed - network observer or proprioceptor who checks upon validity of system parts and data available from sensors, phenomenon observer for detection of scenario taking place in the environment that is monitored and system observer for analysis of system usefulness and identification of future improvement of the system. In each of these parts ML/AI techniques are used. This framework was implemented as vital part of intelligent forest fire monitoring and surveillance system deployed in coastal part of Croatia and some results are presented.