Autonomous Drone-Based Fire Detection and Suppression Using YOLOv8 and 2D Blueprint Mapping for Industrial Applications

Lately, spotting and controlling fires quickly has become more vital than ever to limit damage and save lives. Traditional methods often suffer from delayed response times and limited coverage, making them less effective in preventing the escalation of fires. In this paper, we focus on developing and implementing an autonomous drone system that enhances the efficiency of fire detection and firefighting efforts. The system integrates advanced machine learning techniques with live CCTV camera feeds to detect the presence of fire accurately. Upon detection, the system pinpoints the precise location of the fire and autonomously navigates a drone equipped with fire extinguishing tools, such as the AFO (Auto Fire Off) ball, to the affected area. Our proposed solution leverages a convolutional neural network (CNN) to process real-time video footage and accurately identify fire. The drone’s navigation system employs a combination of GPS and computer vision to traverse the environment and reach the fire site quickly and efficiently. Integrating these technologies ensures a rapid and reliable response to fire incidents. This study explores the design, implementation, and testing the autonomous drone platform, highlighting the challenges encountered and the solutions developed to address them. The findings suggest that integrating machine learning and autonomous navigation in fire response systems can lead to more effective and timely interventions, ultimately reducing the impact of fires on communities and infrastructure.