Exadaktylos, Stylianos (University of Cyprus), Kolios, Panayiotis (University of Cyprus), Eliades, Demetrios (University of Cyprus)

An integrated modelling framework for drone-based flood monitoring

Scheduled for presentation during the Regular Session "Path Planning II" (ThA4), Thursday, June 23, 2022, 10:30−10:50, Divona-2

2022 International Conference on Unmanned Aircraft Systems (ICUAS), June 21-24, 2022, Dubrovnik, Croatia

This information is tentative and subject to change. Compiled on April 25, 2024

Abstract

The consequences of natural catastrophes, such as floods, can be devastating. To mitigate their impact, it is critical to carry out a rapid analysis of the affected region in the moments after they occur, in order to enhance situational awareness. For this purpose, Unmanned Aerial Vehicles (UAVs) may be considered as the technological basis onto which intelligent systems capable of providing support to rescue teams are built. In this work, we present a novel design of a single-agent architecture that can monitor environments affected by a flood event, and provide support to first responders in the decision-making process. In this framework, the autonomous UAV agent is capable of selecting regions based on a hydrological models and generate trajectories for monitoring and coverage. While monitoring the selected areas, the agent collects real-time data regarding water depth and water velocity. Exploiting the data collection process, the agent updates the selected regions, hence constantly updating its path. To demonstrate the proposed framework, we present a simulation study of a flash-flood which could affect the city of Nicosia.