Abstract The deployment of Unmanned Aerial Vehicles (UAVs) in conjunction with Mobile Edge Computing (MEC) has come to be a viable approach to solve some challenges that face the internet of things systems, including energy consumption, latency, and data processing efficiency. However, trajectory planning optimization for UAVs in the MEC systems remains a challenging issue due to energy restrictions. This study introduces a trajectory planning algorithm, Fungal Growth–Differential Evolution (FGODE), seeking to minimize overall energy consumption without compromising task offloading efficiency and UAV mobility. The approach employs a hybrid optimization algorithm that combines the Fungal Growth Optimizer (FGO) and Differential Evolution (DE) algorithms to effectively maintain between searching new regions and refining promising solutions. The method also utilizes an optimized population size-based encoding mechanism to properly represent candidate solutions. Furthermore, a low-complexity greedy mechanism is employed to sequence the stop points along each UAV’s trajectory, while elite opposition-based learning and Gaussian mutation are utilized to accelerate convergence and mitigate premature stagnation. Several experiments have been conducted to compare with several algorithms. Experimental findings show that FGODE delivers more competitive results than state-of-the-art algorithms across several performance metrics, displaying higher optimization capability.
Othman et al. (Wed,) studied this question.