Advancing Capabilities With UAVs: A Comprehensive Review of Sensor Fusion, Path Planning, Energy Management, and Deep Learning Algorithms

Abstract

The current engineering and construction of buildings has led to an increase in building height, which has created difficulties for search and rescue teams working with fire departments. These challenges are exacerbated by the limited reach of the fire ladder and water hose restrictions. Furthermore, traffic jams might make it difficult for the fire brigade to get to the site quickly, which could lead to the loss of life and property. To combat that issue, unmanned aerial vehicles (UAVs) can be employed for environment mapping in buildings during fire emergencies which will help detect fire and identify trapped humans. That is possible by equipping UAVs with path-planning algorithms, control strategies, and image-processing algorithms coupled with a robust sensor fusion algorithm. Well-designed UAVs can help overcome the limitations of traditional firefighting methods. This paper presents an elaborate review of the development of UAVs in terms of their data handling, control management, and energy storage. For the data handling and control management systems, the paper discusses effective sensor fusion algorithms, optimal path-planning strategies, control techniques, and fire detection algorithms. The use of hybrid energy storage for UAVs along with its energy management is elaborated. Through this review, researchers will acquire knowledge about the existing development in the application of UAVs in firefighting. Finally, the paper highlights advancements and limitations of various sensor fusion, path planning and energy management strategies to increase the flight time.