@misc{Cheng_Nannan_Research,
 author={Cheng, Nannan and Li, Qing and Wang, Heng and Tong, Saisai and Fu, Xingjian},
 howpublished={online},
 publisher={Zielona Góra: Uniwersytet Zielonogórski},
 language={eng},
 abstract={This paper proposes an improved version of the marine predator algorithm (abbreviated as IMPA) to optimize unmanned aerial vehicle (UAV) dynamic adversarial game strategies, effectively addressing critical limitations of the original marine predator algorithm (MPA), including slow convergence speed and a pronounced tendency to fall into local optima.},
 abstract={The IMPA incorporates four key innovations: opposition-based learning (OBL) for enhanced initial population quality, an adaptive mechanism for dynamic population resizing, nonlinear step-size control, and an inertia weight strategy. These improvements collectively accelerate convergence, balance global exploration with local exploitation, and significantly improve the ability to escape local optima.},
 abstract={To validate the algorithm`s performance, a dynamic game model based on situation assessment and utility functions is established for red-blue UAV confrontation scenarios, and the IMPA is successfully applied to solve for Nash equilibrium. Comprehensive simulation results demonstrate that the proposed algorithm converges over 40% faster than the original MPA and other benchmark algorithms, and robustly achieves the Nash solution in all 100 consecutive tests, underscoring its superior effectiveness and reliability in dynamic adversarial environments.},
 title={Research on UAV dynamic adversarial game strategies using an improved marine predator algorithm},
 type={artykuł},
 keywords={improved marine predator algorithm, UAV dynamic adversarial game, convergence speed, local optima, Nash equilibrium},
}