Drone Swarm Coordination Using Reinforcement Learning for Efficient Wildfires Fighting

被引：0
作者：
Blais M.-A. [1 ,2 ]
Akhloufi M.A. [1 ,2 ]
机构：
[1] Dept. of Computer Science, Université de Moncton, Moncton, NB
[2] Perception, Robotics, and Intelligent Machines (PRIME), Université de Moncton, Moncton, NB
来源：
SN Computer Science | / 5卷 / 3期
基金：
加拿大自然科学与工程研究理事会;
关键词：
Drones; Intelligent system; Reinforcement learning; Swarm; Wildfires;
D O I：
10.1007/s42979-024-02650-6
中图分类号：
学科分类号：
摘要：
Natural events, such as wildfires, pose a serious threat to the human population and cause significant environmental and economic damage. As climate change increases the frequency and intensity of extreme natural events, more efficient solutions are required to mitigate their impacts. One proposed solution is the usage of a swarm of drones and unmanned ground vehicles to extinguish wildfires. Controlling a swarm of drones is a challenging task that requires a complex control system capable of providing a good coordination between multiple vehicles. Artificial intelligence, particularly reinforcement learning, has been proposed as a solution for the autonomous control of a swarm. In this paper, we propose a conceptual system based on reinforcement learning and swarm robotics to efficiently combat wildfires. Our system consists of an overwatch drone and multiple payload drones which are responsible for carrying fire suppressants. The goal of our research is to guide a swarm of payload drones to a wildfire while preserving a tight swarm grouping. Each drone is individually controlled and aware of the position of the goal and the positions of other agents. To compare different algorithms, we tested a DQN, Rainbow DQN and FQF using a simulation and four AirSim scenarios. With our approach, we achieve impressive results in controlling a swarm of drones towards a common goal. Our research demonstrates the possibility of utilizing reinforcement learning to control a swarm of drones for wildfire fighting. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.
引用
收藏
相关论文
共 113 条
  • [51] Yang Y., Juntao L., Lingling P., Multi-robot path planning based on a deep reinforcement learning DQN algorithm, CAAI Trans Intell Technol, 5, 3, pp. 177-183, (2020)
  • [52] Wang D., Deng H., Pan Z., Mrcdrl: multi-robot coordination with deep reinforcement learning, Neurocomputing, 406, pp. 68-76, (2020)
  • [53] Zhang L., Sun Y., Barth A., Ma O., Decentralized control of multi-robot system in cooperative object transportation using deep reinforcement learning, IEEE Access, 8, pp. 184109-184119, (2020)
  • [54] Huttenrauch M., Sosic A., Neumann G., Guided deep reinforcement learning for swarm systems, Arxiv Preprint Arxiv, 1709, (2017)
  • [55] Venturini F., Mason F., Pase F., Chiariotti F., Testolin A., Zanella A., Zorzi M., Distributed reinforcement learning for flexible and efficient UAV swarm control, IEEE Trans Cogn Commun Netw, 7, 3, pp. 955-969, (2021)
  • [56] Hammond T., Schaap D.J., Sabatelli M., Wiering M.A., Forest fire control with learning from demonstration and reinforcement learning, International Joint Conference on Neural Networks (IJCNN)., pp. 1-8, (2020)
  • [57] Wang Z., Schaul T., Hessel M., Hasselt H., Lanctot M., Freitas N., Dueling network architectures for deep reinforcement learning, In: International Conference on Machine Learning., pp. 1995-2003, (2016)
  • [58] Haksar R.N., Schwager M., Distributed deep reinforcement learning for fighting forest fires with a network of aerial robots, In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)., pp. 1067-1074, (2018)
  • [59] Pickem D., Glotfelter P., Wang L., Mote M., Ames A., Feron E., Egerstedt M., The robotarium: A remotely accessible swarm robotics research testbed, In: 2017 IEEE International Conference on Robotics and Automation (ICRA)., pp. 1699-1706
  • [60] Pham H.X., La H.M., Feil-Seifer D., Nefian A., Cooperative and distributed reinforcement learning of drones for field coverage, Arxiv Preprint Arxiv, 1803, (2018)
12345678910