Resource Optimization Allocation of UAV Swarm Based on Dispersion Quantification and DDPG Algorithm

Abstract

: In the fields of dynamic target protection and autonomous system cooperative control, achieving the optimal allocation of limited resources through intelligent decision-making has always been a core challenge in engineering practice and theoretical research. This study focuses on the complex problem of UAV swarms using smoke grenades to jam tracking targets in three-dimensional space, aiming to extend the target's movement time. A complex system model integrating multi-body dynamic coupling, spatiotemporal resource allocation, and environmental uncertainty adaptation is constructed. By defining the discrete degree quantitative index of target C, the dual-objective optimization of UAV utilization rate and smoke grenade effectiveness is realized based on the reinforcement learning DDPG algorithm.