Large-Scale UAV Formation Reconstruction Method Based on Dynamic Grouping

Formation reconstruction for large-scale unmanned aerial vehicle (UAV) swarms faces critical challenges in computational complexity and safe navigation within high-density environments. To address the O(N3) computational bottleneck of traditional assignment algorithms, this study proposes a Dynamic Grouping Task Assignment (DGTA) method based on a hierarchical sorting strategy. Furthermore, an Integrated Hierarchical Control (IHC) framework is developed by coupling fuzzy logic velocity regulation and linear trajectory prediction with an improved Artificial Potential Field (APF) method. Numerical simulation results demonstrate that the DGTA method significantly enhances efficiency; specifically, when the swarm is partitioned into three complete groups, the task assignment time for a 600-UAV swarm is reduced by over 90% compared to non-grouping approaches. Additionally, the IHC framework reduces hazardous incidents by more than 27% in congested scenarios. This DGTA-IHC structure successfully balances global optimality with real-time safety, providing a scalable solution for the autonomous coordination of ultra-large-scale swarms.