Effective Fusion of Disaster-Relief Agent in RoboCupRescue Simulation

Abstract

The RoboCupRescue Simulation project is one of the responses to recent large-scale natural disasters. In particular, the project provides a platform for studying disaster-relief agents and simulations. We designed and implemented an agent based on the results of a combinational experiment of various modules taken from teams that participated at RoboCup 2017. We developed a new fusional agent with better modules in the Agent Development Framework. This paper presents the results of the combination experiment in detail. We confirm that our fusional agent based on the experimental results obtained a better score than the champion agent at RoboCup 2017.