Research on Cooperative Smoke Screen Deployment Strategy for Multi-UAV Interference Based on Effective Occlusion Detection and Hybrid Optimization
Year: 2026
Type: conference-paper
Abstract: In the dynamic and complex domain of drone operations, smoke-based countermeasures have emerged as a cost-effective method for disrupting electro-optical detection systems. However, in scenarios involving multiple unmanned aerial vehicles (UAVs) and smoke countermeasures against fast-moving targets, existing studies face several challenges, including accurately evaluating shielding effectiveness, optimizing deployment strategies within complex decision-making frameworks, and maintaining algorithmic stability. To address these issues, this study proposes a robust occlusion determination model for cylindrical targets based on the principle of full line-of-sight blocking. The target surface is discretized into 500 representative feature points, and the intersections between their line-of-sight sets and the incoming target are examined to provide a geometrically precise evaluation of occlusion effectiveness. Furthermore, an occlusion indicator function is introduced to compute the total effective occlusion duration, enabling the model to adapt to diverse operational scenarios, ranging from single-UAV/single-target engagements to coordinated multi-UAV/multi-target missions. To address high-dimensional and heavily constrained optimization problems with non-differentiable objective functions, a novel hybrid optimization algorithm (ABC-DE) is proposed, integrating Differential Evolution (DE) with the Artificial Bee Colony (ABC) algorithm. This hybrid approach achieves rapid convergence while effectively avoiding local optima. Ablation experiments conducted in a representative multi-UAV coordinated deployment scenario demonstrate that, compared with the standard DE algorithm, the proposed hybrid algorithm exhibits superior convergence speed, trajectory stability, and robustness to random initialization, thereby confirming its effectiveness and reproducibility.
Authors: Xiaoguo Zhang, Rui Mao
Institution: Guilin University of Electronic Technology
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7204815461