Enerjiye Duyarlı Optimizasyon Kullanarak İHA Sürülerinde Güvenlik Odaklı Görev Ataması

Öz

Birden fazla İHA’nın görev aldığı senaryolarda görev ataması, özellikle enerji kısıtlı görevler sırasında kritik bir sorun oluşturmaktadır, çünkü İHA’ların bataryaları sınırlı kapasitelere sahiptir. Mevcut görev atama teknikleri çoğunlukla kaynakların yeterli olduğu varsayımıyla mesafe veya zaman minimizasyonuna odaklanmaktadır. Mesafe temelli bu yaklaşımlar, güvenli olmayan görev atamalarına yol açabilmekte ve bataryaların tükenmesi sonucunda görev başarısızlıklarına neden olabilmektedir. Bu çalışma, heterojen batarya kapasiteleri dikkate alınarak bir İHA sürüsünün görev atama problemini ele almaktadır. Görev atama problemi, her bir İHA’nın konumu ve başlangıç durum-yükü (SoC) ile temsil edildiği, İHA’lar ile hedefler arasında bire bir eşleştirme olarak modellenmiştir. Enerjiye duyarlı bir strateji, doğrusal olmayan ceza fonksiyonları kullanılarak batarya fizibilitesinin optimal atama probleminin maliyet fonksiyonuna dahil edilmesiyle geliştirilmiştir. Önerilen yeni problem formülasyonu, klasik optimal çözücülerin uygulanabilirliğini ve çözüm esnekliğini korurken enerji açısından kritik atamaların önemini azaltmaktadır. Geliştirilen stratejinin performansı, Monte Carlo benzetimleri ve açgözlü en yakın komşu ile mesafe temelli optimal atama yöntemleriyle yapılan karşılaştırmalı deneyler aracılığıyla değerlendirilmiştir. 500 Monte Carlo simülasyonu üzerinde gerçekleştirilen deneylerde önerilen yöntem %65.88 görev hayatta kalma oranı elde etmiştir. Bu oran, yalnızca mesafeye dayalı optimal atama yöntemi (%54.76) ve greedy temel yaklaşım (%49.09) ile karşılaştırıldığında belirgin bir iyileşmeye işaret etmektedir. Ayrıca önerilen yaklaşım, ortalama kaçınılabilir çarpışma sayısını 2.29’a düşürürken, bu değer mesafeye dayalı yöntemde 4.51 ve greedy yaklaşımda 5.65 olarak gözlenmiştir. Ancak bu performans artışı, toplam uçuş mesafesinin ortalama 551.57 m’ye yükselmesi karşılığında elde edilmiştir.

Anahtar Kelimeler

• Görev atama
• enerjiye duyarlı optimizasyon
• İHA sürüleri
• çoklu İHA sistemleri

Safety-Oriented Task Assignment in UAV Swarms Using Energy-Aware Optimization

Öz

The allocation of tasks in the presence of multiple UAVs is a critical concern, especially during energy-constrained tasks, as the batteries of the UAVs have limited capacities. The existing task allocation techniques mainly minimize the distance or time while assuming the availability of resources. These techniques, which rely on distance, might result in unsafe task allocation, which could eventually cause failure due to depleted batteries. This work deals with the task allocation of the UAV swarm considering the heterogeneous battery capacities. The problem of task allocation is considered to be one-to-one between the UAVs and the targets, with each UAV represented by its position and initial SoC. An energy-aware strategy for the optimal assignment problem is developed by incorporating battery feasibility into the cost of the optimal assignment problem with the use of nonlinear penalty functions. The new problem formulation down-plays the importance of the energy-critical assignments while maintaining flexibility and classical optimal solver applicability. The performance of the developed strategy is evaluated with the use of Monte Carlo simulations and comparative experiments with greedy nearest neighbor and optimal assignments. Across 500 Monte Carlo simulations, the proposed method achieved a survival rate of 65.88%, compared with 54.76% for the distance-only optimal method and 49.09% for the greedy baseline. It also reduced average avoidable crashes to 2.29, compared with 4.51 and 5.65 for the distance-only and greedy approaches, respectively, at the expense of a higher average total flight distance of 551.57 m.

Anahtar Kelimeler

• Task allocation
• Energy-aware optimization
• UAV swarms
• multi-UAV systems

Kaynakça

1. Pu, Q., Yang, L., & Li, T. (2025). Dynamic coverage path planning method for UAV formations in multi-region aerial tasks. Aerospace Science and Technology, 167, 1–19.
2. Kunarak, S. (2026). Energy management revolution in unmanned aerial vehicles using deep learning approach. Applied Sciences, 16, 1–22.
3. Li, J., Yang, X., Yang, Y., & Liu, X. (2024). Cooperative mapping task assignment of heterogeneous multi-UAV using an improved genetic algorithm. Knowledge-Based Systems, 296, 1–17.
4. Yuan, M., Chen, M., Zhou, T., & Han, Z. (2025). Energy learning hyper-heuristic algorithm for cooperative task assignment of heterogeneous UAVs under complex constraints. Defence Technology, 54, 1–14.
5. Park, H., Lee, S., & Cho, S. (2025). Distributed inference in IoT-based aerial network of UAVs. Internet of Things, 30, 1–10.
6. Guo, C., Huang, L., & Tian, K. (2025). Combinatorial optimization for UAV swarm path planning and task assignment in multi-obstacle battlefield environment. Applied Soft Computing, 171, 1–19.
7. Wang, Z., Gao, J., Li, W., Chen, Y., & Cheng, X. (2025). Swarm intelligence optimization-based task assignment for multiple autonomous underwater vehicles. Journal of the Franklin Institute, 362, 1–22.
8. Ding, W., Song, J., & Yu, J. (2026). A multi-UAV unfixed scale task assignment method with uncertainty based on deep Q network. Robotics and Autonomous Systems, 197, 1–14.

9. Fei, B., Liu, D., Bao, W., Zhu, X., & Li, X. (2025). Task-driven multi-UAV path planning via three-stage optimization strategy for urban region surveillance. Advanced Engineering Informatics, 68, 1–18.
10. Mondal, A., Talukdar, N., Hazra, A., & Mazumdar, N. (2025). Deep reinforcement learning based multi-UAV assisted data collection for deadline-sensitive IoT networks. Physical Communication, 73, 1–14.
11. Raju, M. R., Mothku, S. K., Somesula, M. K., & Chebrolu, S. (2025). Age and energy aware data collection scheme for urban flood monitoring in UAV-assisted wireless sensor networks. Ad Hoc Networks, 168, 1–10.
12. Liu, T., Sun, C., Zhang, Y., & Sun, W. (2025). Energy-efficient end-to-end optimization for UAV-assisted IoT data collection and LEO satellite offloading in SAGIN. Electronics, 15, 1–23.
13. Li, H., Wang, S., Du, Y., Li, R., Fan, X., & Luo, C. (2026). Energy-efficient hierarchical federated learning in UAV networks with partial AI model upload under non-convex loss. Sensors, 26, 1–29.
14. Han, Z., Fang, C., Wang, W., & Xu, J. (2025). Energy-efficient UAV routing problem based on approximate cellular decomposition for geohazards monitoring. Computers & Operations Research, 183, 1–15.
15. Zhang, C., Gui, X., Gupta, G. S., & Hasan, S. F. (2025). Placement optimization of multiple UAVs for energy-efficient maximal user coverage. ICT Express, 11(6), 1167–1172.
16. Slimani, K., Khoulji, S., Taherdoost, H., & Kerkeb, M. L. (2025). Real-time and energy-aware UAV routing: A scalable DAR approach for future 6G systems. Computers, Materials & Continua, 83(3), 4667–4686.
17. Lian, F., Li, B., Yang, Q., Zhu, H., & Du, D. (2025). WA-LPA*: An energy-aware path-planning algorithm for UAVs in dynamic wind environments. Drones, 9, 1–26.
18. Zhao, J., Chen, S., Chi, P., & Wang, Y. (2026). Dynamic merging and splitting for large-scale swarm navigation of UAVs and UGVs in unknown area. Robotics and Autonomous Systems, 198, 1–16.
19. Song, F., Zeng, Q., Zhu, X., Zhang, R., Ye, X., & Zhou, H. (2026). Cooperative navigation framework for UAV formations using LSTM and dynamic model fusion. Drones, 10, 1–30.
20. Lai, Y., Wu, H., Xia, Y., Fu, X., Duan, D., Wang, A., & Shi, M. (2026). Research on collaborative path planning of UAV swarms for urban logistics distribution in dense building environments. Expert Systems with Applications, 305, 1–22.
21. Medani, K., Gherbi, C., Mabed, H., & Aliouat, Z. (2025). Energy-efficient Q-learning-based path planning for UAV-aided data collection in agricultural WSNs. Internet of Things, 33, 1–17.
22. Mohammed, Y. I., Hassan, R., Hasan, M. K., Abbas, H. S., Khan, M. A., Baili, J., & Gupta, D. (2025). Optimizing UAV-assisted IoT sensor networks: A multi-objective approach to data collection and routing. Alexandria Engineering Journal, 115, 47–56.
23. Wang, C., Dong, M., Yuan, Y., & Li, G. (2026). Energy-efficient trajectory planning for UAV-assisted communication recovery using multi-agent graph reinforcement learning. Ad Hoc Networks, 184, 1–14.
24. Zhang, D., Li, W., Liu, C., He, X., & Li, K. (2025). A High-Efficiency Task Allocation Algorithm for Multiple Unmanned Aerial Vehicles in Offshore Wind Power Under Energy Constraints. Journal of Marine Science and Engineering, 13(9), 1–15.
25. Negassa, W. G., Gelmecha, D. J., Singh, R. S., & Rathee, D. S. (2025). Hybrid machine learning-based 3-dimensional UAV node localization for UAV-assisted wireless networks. Cognitive Robotics, 5, 61–76.
26. Li, H., Qiu, Z., Han, X., Zhang, M., Liao, D., & Jin, H. (2024). UAV flight path planning optimization. Telecommunication Systems, 87, 329–342.
27. Aggarwal, S., Budhiraja, I., Garg, S., Kaddoum, G., Choi, B. J., & Hossain, M. S. (2025). A blockchain-based secure path planning in UAV communication networks. Alexandria Engineering Journal, 113, 451–460.
28. Sezgin, A. (2025). Scenario-driven evaluation of autonomous agents: Integrating large language model for UAV mission reliability. Drones, 9, 1–21.
29. Sezgin, A., & Boyacı, A. (2025). Optimizing drone deployment for network coverage using a hybrid ant colony optimization and reinforcement learning approach. Journal of Aeronautics and Space Technologies, 18(2), 268–285.