我所李润梅老师课题组完成的论文《Distributed Path Planning for Multi-UAV Systems Integrating Fish Swarm Behavior Strategies》于2025-10在《IEEE International Conference on Intelligent Rail Transportation (ICIRT)》正式发表。论文摘要如下：
Abstract: With the rise of the low-altitude economy, multi-UAV systems hold strong potential for tasks such as railway inspection. To address the challenges of coordinated path planning in complex environments, this paper proposes a distributed framework that integrates a task-driven planning model with artificial fish swarm behaviors and a Double Deep Q-Network (DDQN). The model incorporates UAV kinematics, terrain variations, and obstacle distributions to ensure safe and efficient navigation. Typical swarm behaviors are represented as acceleration vectors and fused through adaptive behavior weights, forming a flexible action space. A shared DDQN learns the optimal behavior weight distribution from local and neighboring states, enabling decentralized decision-making with centralized training. Simulations in representative 3D scenarios validate the method's effectiveness, demonstrating improved path efficiency and safety compared to baseline approaches.
引用信息：Jiaan Nie, Runmei Li, Zhang Hui. Distributed Path Planning for Multi-UAV Systems Integrating Fish Swarm Behavior Strategies. IEEE International Conference on Intelligent Rail Transportation (ICIRT), 2025-10. DOI: 10.1109/icirt66379.2025.11216753