Swarm Intelligence-Based Energy Optimization for UAV Mission Efficiency

Abstract

This study investigates the effectiveness of swarm intelligence–based optimization methods for improving energy efficiency and minimizing mission duration in multi–unmanned aerial vehicle (UAV) swarms. The problem is formulated as a route-sequencing optimization task in which the visiting order of target points is optimized to reduce total energy consumption while ensuring complete mission feasibility. A classical nearest-neighbor assignment serves as the baseline and is compared against two evolutionary approaches: the single-objective Particle Swarm Optimization (PSO) and the multi-objective Non-dominated Sorting Genetic Algorithm II (NSGA-II). A Python-based simulation environment was developed to evaluate algorithmic performance under varying payload and wind conditions, including Zero, Constant, and Ornstein–Uhlenbeck (OU) stochastic wind models. Experimental results indicate that route sequencing optimization substantially decreases overall energy demand. NSGA-II, in particular, successfully constructs a well-defined Pareto front for the energy–time objectives, offering mission planners a flexible spectrum of trade-off solutions. In the OU + 150 g scenario, the Knee-point solution (E = 26.8 Wh, T = 93 s) achieved approximately 13% lower energy consumption at the cost of only a 7% increase in mission duration relative to the baseline. Across all methods, mission completion remained at 100%, and coverage ratios improved notably. These findings confirm that swarm intelligence–based optimization techniques provide robust and efficient tools for balancing energy consumption and operational time in UAV swarm mission planning.