USING SITUATIONAL AWARENESS FOR ADAPTIVE DECISION MAKING IN AGENT-BASED SIMULATION


Topcu O.

Winter Simulation Conference (WSC), Nevada, United States Of America, 3 - 06 December 2017, pp.1276-1287 identifier

  • Publication Type: Conference Paper / Full Text
  • City: Nevada
  • Country: United States Of America
  • Page Numbers: pp.1276-1287
  • Middle East Technical University Affiliated: Yes

Abstract

In an agent-based simulation, a plausible decision in a specific context cannot stay be valid in the face of the changing situation. Therefore, the result of the decision making process is mostly related to the agent's situational awareness and its adaptation to the new context and situation of the environment. A sound estimation of the situation requires a clean understanding of the operational domain, not only the data taken from sensors. In this paper, it is aimed to improve decision making by increasing the situational awareness of an agent by incorporating the decision making mechanism with the prior knowledge about the problem domain, such as the existing rules. Specifically, an existing adaptive decision making architecture, which is based on the deliberative coherence theory, is adopted to be driven by situational awareness. Moreover, a case study incorporating unmanned surface vehicles that are aware of the international sea traffic rules is demonstrated.