An Intelligent Strategy for Tactical Movements of UAVs in Disaster Scenarios

Abstract

Unmanned Aerial Vehicles (UAVs) are envisioned as flexible and fast-deploying communication network for disaster scenarios, where the typical communication infrastructure is likely to be malfunctioning. A few works propose UAVs for building communication links autonomously between rescue team's members in disaster scenarios. The techniques used are usually based on navigation, positioning, and signal strength processing. However, these techniques may not be enough if the objective is to provide communication services to the maximum number of victims and rescuers and not only to a few rescuers. In this situation, dissimilarity metrics, like the Jaccard distance, can provide information about whether the communication service provided to victims is efficient or not (e.g., providing a better distribution of the victims assigned to each UAV acting as service provider). We propose an intelligent strategy that allows UAVs to perform tactical movements in a disaster scenario, combining the Jaccard distance and artificial intelligence algorithms like hill climbing and simulated annealing. Our strategy maximizes the number of victims that are serviced by the UAVs while avoiding network disconnections. Also, a mobility model specifically developed for modelling the victims' movements within the incident site of a disaster scenario is proposed.