Knowledge-Guided Competitive Evolutionary Algorithm for Multi-Solution Sensor-Weapon-Target Assignment Problem
针对现代防空系统中传感器-武器-目标分配问题，提出一种知识引导竞争进化算法，通过提取雷达信道冲突和资源分配模式来引导搜索，并保留多个高质量战术方案，实验表明优于现有方法。
The Sensor-Weapon-Target Assignment (SWTA) problem in modern air defense systems requires simultaneous optimization of sensor allocation, weapon assignment, and temporal coordination under complex operational constraints. Traditional optimization approaches typically converge to single solutions, failing to capture the multimodal nature of SWTA landscapes where multiple distinct assignment strategies can achieve comparable interception probabilities. This limitation restricts tactical flexibility essential for robust operational planning in dynamic combat environments. To address these challenges, this paper proposes a novel Knowledge-guided Competitive Evolutionary Algorithm (KCEA), integrating with a knowledge extraction mechanism and a competitive selection strategy. The knowledge-guided component systematically identifies and utilizes radar channel conflicts and resource allocation patterns to bias evolutionary search toward promising solution regions, while the competitive selection mechanism ensures preservation of multiple high-quality tactical alternatives throughout the optimization process. Comprehensive experimental validation on benchmark instances demonstrates that KCEA significantly outperforms state-of-the-art methods.