Safety and Surveillance on Unmanned Aerial Vehicles Control Systems and Optimization Methods: A Systematic Review

Unmanned Aerial Vehicles (UAVs) have become an issue of high research activity because of their extensive variety of applications in agriculture, logistics, security, and emergency responses. Though they are advancing at a very fast rate, their operational capability is still limited due to a number of issues, the primary ones being short flight range, inadequate autonomy, and the difficulty of collaborating with UAVs. The short battery life, which has been a source of energy constraint, has led to the investigation of hybrid propulsion systems, improved energy management, and automated battery replacement via docking systems. No less important are the questions of autonomous navigation, supposed to be based on efficient path planning, collision avoidance, and optimization of the payload. The coordination of swarms of UAVs also adds complexity to the system, which, in turn, necessitates providing the system with dependable communication and safety measures. Recent research indicates that the application of artificial intelligence and optimization solutions, including Reinforcement Learning (RL) and Deep Q-Learning (DQL), can be used as a path-planning and coordination tool, but not in very dynamic or unpredictable environments. As UAVs transition from a hobby tool to an indispensable part of disaster management and precision agriculture, there is a growing need to overcome these obstacles. This systematic review investigates the current innovations within the area of UAV control and optimization, as well as the current constraints and future research opportunities to attain trustworthy, self-governing and power-efficient UAV systems.