A Vehicle-Target Assignment Framework: Collaborative Missions of Multiple High-Speed Flight Vehicles With Multiple Targets

In scenarios where multiple high-speed flight vehicles (HFVs) perform collaborative tasks on multiple targets, the reasonable allocation of vehicle resources to the designated targets is crucial for enhancing the overall mission effectiveness. In this article, a HFV-target assignment (HTA) framework that incorporates an evaluation model and an enhanced assignment algorithm is proposed for collaborative missions. The evaluation model is designed to evaluate the overall mission effectiveness considering factors such as actual trajectory, mission success rate, and target value, which are collectively represented as distance cost and act payoff. In order to quickly obtain actual trajectories of HFVs, a data-driven prediction method based on multi-scale neural network fusion is suggested. Due to the complexity of real-world mission environments and the demand for real-time responsiveness, the target assignment for large-scale HFV swarm requires rapid and efficient solutions. In this paper, the HTA optimization problem is converted into a bipartite graph optimal matching problem and an iterative generalized Kuhn-Munkres (IGKM) algorithm is proposed. Simulation results compared with particle swarm optimization and CNN-based learning method reveal that the algorithm exhibits competitive performance in terms of solution stability and computational speed in both small-scale and large-scale scenarios.