<![CDATA[The efficient utilization of limited onboard energy remains a fundamental challenge in cooperative multi-UAV data collection missions. Existing routing approaches typically optimize surrogate objectives, such as travel distance or aggregate energy consumption, which do not directly reflect mission effectiveness. The present paper proposes a Mission-level Energy-aware Genetic Algorithm (ME-GA) that directly maximizes mission-level energy efficiency, defined as the ratio of successfully delivered sensing data to total energy consumption. The proposed framework integrates a mission-level simulator into the fitness evaluation, explicitly modeling UAV propulsion, sensing, data buffering, wireless communication, and return-to-base feasibility under energy constraints. Extensive simulations involving up to 9 UAVs and 100 Points of Interest (PoIs) under both grid and random spatial layouts demonstrate that ME-GA consistently achieves high and stable energy efficiency while maintaining near-complete task satisfaction and high data delivery reliability. In comparison to GA-based baselines, the proposed approach enhances energy efficiency by approximately 5–15% across the evaluated scenarios along with a reduction in total travel distance by up to 40% in larger fleet sizes. Overall, the results demonstrate that mission-level energy efficiency serves as a unified and physically meaningful objective for multi-UAV optimization, enabling robust and scalable performance across diverse operational scenarios.]]>
