Addressing disasters in smart cities through UAVs path planning and 5G communications: A systematic review

Abstract

UAVs are increasingly incorporated in a wide range of domains such as disaster management and rescue missions. UAV path planning deals with finding the most optimal or shortest path for UAVs such that minimum energy and resources are utilized. This paper examines the path planning algorithms for UAVs through a literature survey conducted on 139 systematically retrieved articles published in the last decade that are narrowed down to 36 highly relevant articles. As retrieved from the shortlisted articles, the path planning algorithms include RRT, Artificial Potential, Voronoi, D-Star, A-Star, Dijkstra, MILP, Neural Network, Ant Colony Optimization, and Particle Swarm Optimization that are classified into four main types: Model-based, Conventional, Learning-based, and Cell-based. Most of the disaster-related articles are focused on the post-disaster phase only and use conventional and learning-based algorithms with applications to localize victims and optimize paths. Regarding the UAV communication network (UAVCN), the key challenges are communication issues, resource allocation, UAV deployment, defining UAV trajectory, and content security. UAV path planning’s key barriers are path optimization, path completeness, optimality, efficiency, and achieving robustness. Accordingly, a holistic IoT-powered UAV-based smart city management system has been recommended in the current study where all the smart city key components are integrated to address disasters like floods, earthquakes, and bush fire. The proposed holistic system can help prepare for disasters and mitigate them as soon as these arise and help enhance the smart city governance.