Fire brigade response enhancement using drone swarms: Comparative analysis and beyond

Abstract

Addressing the increasing incidence of building fires is critical for enhancing public safety and effective emergency response. However, there is a notable gap in the literature regarding effective strategies for managing these fires. This study presents a prototype of affordable drone swarms designed to enhance firefighting efforts through live surveillance, data collection, and coordinated operations. The drones are organized into groups, each led by a controller to ensure effective communication and adapt dynamically by adding units as needed. They utilize an artificial potential field (APF) model for movement control, allowing them to navigate toward fire hotspots while avoiding obstacles. Through Reinforcement Learning (RL) drones learn to enhance their navigation skills and choose optimal fire response strategies in unpredictable situations. Equipped with thermal cameras, GPS, and live communication, the drones can efficiently monitor and extinguish fires. Strategic recharging stations enable continuous operation without human intervention. By integrating RL with traditional models, this approach bridges conventional fire fighting methods and modern drone technology, offering a scalable, adaptive, and intelligent solution to the growing challenge of building fires.