Hierarchical reconnaissance-strike intelligent simulation for multiple moving targets by UAV swarms

Unmanned Aerial Vehicle (UAV) swarms for cooperative reconnaissance-strike missions have emerged as a pivotal component of modern warfare. However, existing simulation models often struggle to comprehensively characterize complex operational elements. To address this challenge, this study proposes a State-Behavior-Decision Hierarchy (SBDH) model, grounded in an in-depth analysis of swarm cooperative characteristics and the operational mechanism of the Observe-Orient-Decide-Act (OODA) loop. By integrating the strengths of Finite State Machines (FSMs) and Behavior Trees (BTs), the model establishes a tri-tier cooperative architecture comprising command, strike, and reconnaissance units. Through the coupling of state nodes, behavior-command nodes, and decision-transition nodes, it enables the parameterized construction of operational models. Furthermore, this paper introduces a grid tracking-back reconnaissance strategy and a spatial optimal matching strike strategy. The former achieves optimal area partitioning based on the Nondominated Sorting Genetic Algorithm II (NSGA-II) for the efficient search for multiple moving targets, while the latter ensures optimal firepower allocation and precision strikes through a spatial greedy optimization method. Simulation experiments evaluate the impact of key variables, including target motion parameters and UAV swarms operational parameters, on combat effectiveness metrics. The results validate the model’s applicability in complex scenarios, providing a significant theoretical reference for tactical-edge UAV swarms deployment.