UAV Fleet Size Optimization in Disaster Management: A Comparative Analysis using HFFPSO
Indu Chandran, Kizheppatt Vipin · 2023
Disasters, whether they are of natural origin or caused by human activities, can have devastating consequences on both human lives and infrastructure. In many cases, these disasters can render certain areas inaccessible to first responders due to safety concerns or challenging terrain. To address this challenge, researchers are exploring the potential of unmanned aerial vehicles, commonly known as drones, as a promising solution for disaster monitoring, assessment, and recovery efforts. Current relief operations depend on intelligent robotic agents to evaluate the damage, expedite timely rescues, and generate maps to assist the affected individuals. To effectively carry out these time-sensitive missions, it is crucial to deploy an appropriate number of reliable UAVs in the affected area. Furthermore, the adoption of efficient path planning techniques is essential to ensure swift coverage of the area. The goal of this research is to establish a well-balanced method for calculating the ideal quantity of unmanned aerial vehicles (UAVs) necessary to effectively survey a specific area while maximizing coverage within a predefined time frame. This involves developing a coverage trajectory through the processes of area decomposition and task assignment, utilizing a Hybrid Firefly-Particle Swarm Optimization (HFFPSO) approach. The study considers two distinct architectures with a homogeneous set of UAVs, and their performance is evaluated and compared in terms of mission latency. The study makes use of PX4-SITL flight stack in ROS framework, and the vehicle dynamics are visualized in the Gazebo physical environment. The research compares the effectiveness of the proposed approach to benchmark optimization strategies using comprehensive theoretical analysis and experimental evaluations.