Robot-assisted discovery of evacuation routes in emergency scenarios

Abstract

When an emergency occurs within a building, it is crucial to guide victims towards emergency exits or human responders towards the locations of victims and hazards. The objective of this work is thus to devise distributed algorithms that allow agents to dynamically discover and maintain short evacuation routes connecting emergency exits to critical cells in the area. We propose two Evacuation Route Discovery mechanisms, Agent2Tag-ERD and Tag2Tag-ERD, and show how they can be seamlessly integrated with existing exploration algorithms, like Ants, MDFS and Brick&Mortar. We then examine the interplay between the tasks of area exploration and evacuation route discovery; our goal is to assess whether the exploration algorithm influences the length of evacuation paths and the time that they are first discovered. Finally, we perform an extensive simulation to assess the impact of the area topology on the quality of discovered evacuation paths.