A Deep Reinforcement Learning Approach for Optimizing the Placement of 5G-Enabled UAVs in Disaster Scenarios

In disaster situations, maintaining reliable communication is critical to support effective search and rescue operations. This paper presents the development and implementation of a Deep Q-Network (DQN) and Proximal Policy Optimization (PPO) methods for managing a fleet of unmanned aerial vehicles (UAVs) in a search and rescue (SAR) mission. The objective is to maximize the coverage of individuals in a designated area to restore communication using a 5 G network in a disaster scenario where traditional communication infrastructures have failed. The proposed methods are evaluated in a simulated 3D environment, demonstrating their effectiveness in optimizing coverage and enhancing the overall SAR mission performance.