UE5 simulation methodology for autonomous uav swarms: multi-agent learning and communication resilience assessment in emergency scenarios
DOI:
https://doi.org/10.31548/Keywords:
UAVs, swarm of autonomous agents, Unreal Engine 5, emergency rescue scenarios, communication disruptions, multi-agent training, telemetryAbstract
The article discusses an approach to building a pilot simulation environment in Unreal Engine 5 to study the behavior of a swarm of unmanned aerial vehicles in emergency response tasks. The relevance of the work stems from the fact that when searching for people, inspecting damaged areas, and avoiding obstacles, communication between devices can be unstable, and centralized control schemes can become inefficient. The study aims to develop a simplified UE5-oriented methodology that enables setting mission scenarios, simulating packet loss, latency, and short periods of communication failure, and collecting telemetry data to compare control approaches. The technique is based on a modular plugin architecture that includes Scenario Runner, Agent Framework, Communication Stress Injector, Role Manager, and Telemetry Logger. Four scenarios are proposed for the initial analysis: covering the open area, searching for targets in the urban environment, avoiding obstacles in the ruin corridor, and a mixed search-and-return mission. The result is a data structure, a set of metrics, and a simulation example of evaluating mission success, execution time, number of collisions, communication quality, and recovery from a stressful window. The proposed approach does not replace field testing, but creates a basis for early validation of ideas and preparation for further hardware validation.
Recieved: 20.03.2026. Recieved: 03.06.2026. Accepted: 22.06.2026
References
1. Rashid, T., Samvelyan, M., Schroeder, C., Farquhar, G., Foerster, J., & Whiteson, S. (2018). QMIX: Monotonic value function factorisation for deep multi-agent reinforcement learning. In Proceedings of the 35th International Conference on Machine Learning (pp. 4295-4304). PMLR.
2. Yu, C., Velu, A., Vinitsky, E., Wang, Y., Gao, Y., Wu, Y., & Wang, J. (2022). The surprising effectiveness of PPO in cooperative multi-agent games. arXiv. https://arxiv.org/abs/2103.01955
3. Gronauer, S., & Diepold, K. (2022). Multi-agent deep reinforcement learning: A survey. Artificial Intelligence Review, 55, 895-943. https://doi.org/10.1007/s10462-021-09996-w
4. Zhao, X., Yang, R., Zhong, L., & Hou, Z. (2024). Multi-UAV path planning and following based on multi-agent reinforcement learning. Drones, 8(1), 18. https://doi.org/10.3390/drones8010018
5. Liu, B., Wang, S., Li, Q., Zhao, X., Pan, Y., & Wang, C. (2023). Task assignment of UAV swarms based on deep reinforcement learning. Drones, 7(5), 297. https://doi.org/10.3390/drones7050297
6. Kong, X., Zhou, Y., Li, Z., & Wang, S. (2023). Multi-UAV simultaneous target assignment and path planning based on deep reinforcement learning. Frontiers in Neurorobotics, 17, 1302898. https://doi.org/10.3389/fnbot.2023.1302898
7. Phadke, A., Medrano, F. A., & Chu, T. (2023). Increasing operational resiliency of UAV swarms. Aerospace Research Communications, 1. https://doi.org/10.3389/arc.2023.12420
8. Phadke, A., Medrano, F. A., & Chu, T. (2022). Towards resilient UAV swarms. Drones, 6(11), 340. https://doi.org/10.3390/drones6110340
9. Zhang, X., & Liu, J. (2024). Research on UAV swarm network modeling and resilience assessment methods. Sensors, 24(1), 11. https://doi.org/10.3390/s24010011
10. Bai, G., Li, Y., Fang, Y., Zhang, Y. A., & Tao, J. (2020). Network approach for resilience evaluation of a UAV swarm by considering communication limits. Reliability Engineering & System Safety, 193, 106602. https://doi.org/10.1016/j.ress.2019.106602
11. Wang, X., Zhao, Z., Yi, L., Ning, Z., Guo, L., Yu, F. R., & Guo, S. (2024). A survey on security of UAV swarm networks: Attacks and countermeasures. ACM Computing Surveys, 57. https://doi.org/10.1145/3703625
12. Liu, C., Zhang, Y., Niu, G., Jia, L., Xiao, L., & Luan, J. (2022). Towards reinforcement learning in UAV relay for anti-jamming maritime communications. Digital Communications and Networks. https://doi.org/10.1016/j.dcan.2022.08.009
13. Epic Games. (2026). Unreal Engine 5 Documentation. https://dev.epicgames.com/documentation/en-us/unreal-engine
14. Epic Games. (2026). Networking and Multiplayer in Unreal Engine. https://dev.epicgames.com/documentation/en-us/unreal-engine/networking-and-multiplayer-in-unreal-engine
15. Epic Games. (2026). Artificial Intelligence in Unreal Engine. https://dev.epicgames.com/documentation/en-us/unreal-engine/artificial-intelligence-in-unreal-engine
Published
Issue
Section
License
Copyright (c) 2026 Energy and Automation
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
All materials are disseminated under the terms of the Creative Commons Attribution 4.0 International Public License, which permits others to distribute the manuscript with proper acknowledgement of the authorship and the original publication in this journal.
