Research on obstacle avoidance and target search of unmanned aerial vehicles in flood search and rescue work

Abstract

The increasing complexity of flood rescue missions demands innovative solutions, with drones emerging as pivotal tools. This research explores dronebased obstacle avoidance and target identification, emphasizing dynamic and uncertain environments characteristic of flood scenarios. Integrating multisensor systems such as LiDAR, thermal imaging, and visual cameras, the study aims to enhance obstacle detection and survivor localization. Novel algorithms, including realtime path planning and sensor fusion, are proposed to optimize drone navigation and precision in turbulent conditions. Additionally, the research addresses challenges like computational efficiency, energy optimization, and adaptive decisionmaking. By leveraging advanced AI and machine learning techniques, the project seeks to contribute to autonomous drone applications in disaster management, offering scalable, costeffective solutions to improve operational safety and efficiency in critical rescue operations. The outcomes are expected to provide actionable strategies for deploying drones in realworld emergencies, ensuring faster response and reduced human risk.