Utilizing AI-powered Drones for Efficient Flood Rescue Operations

• Published in: 2024 3rd International Conference on Automation, Computing and Renewable Systems (ICACRS) pp. 957 - 964
• Main Authors: Kumar, M. Jaswanth, Bitra, Dhanush, Titus, Rohan, Jazz, Dill, Kiliroor, Cinu C
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.12.2024
• Subjects: Accuracy; Artificial intelligence; Computational modeling; Computer vision; Deep learning; Disaster management; Disaster Relief; Disasters; Drone-Assisted; Drones; Flood Relief; Object detection; Transforms
• DOI: 10.1109/ICACRS62842.2024.10841559

In times of severe earthquakes and floods, rapid and effective aid is crucial. Traditional methods often face challenges due to accessibility issues and the vast scale of disasters.The proposed Autonomous Aerial Humanitarian Assistance and Disaster Relief (A2-HADR) System aims to transform disaster response through the use of drone technology. The proposed solution addresses the obstacles in rescue operations by employing drones equipped with Artificial Intelligence (AI) to provide immediate support. The advanced system utilizes state-of-the-art Computer Vision technology to autonomously detect people and obstacles from heights of 50-100 meters at various angles and deliver essential supplies like food, clothing, and rescue gear to those in need. We have executed and assessed a range of advanced object detection algorithms, such as YOLOv8, YOLOv9, and Detectron2. After thorough evaluation, Detectron2 emerged as the most effective model among those tested, showcasing exceptional accuracy and resilience.