摘要
提出了一个火力分配的非线性整数规划模型,该模型充分考虑了多火力点联合毁伤概率对实际射击效果的影响。应用基本蚁群指派算法可以求解此NP[1]问题,但难以满足战术决策的实时性要求。为此,提出一种自适应的信息素更新方法和消除无效分配的变异策略,提高了收敛速度。采用删除已访问节点的方法取代比较禁忌表的方法,进一步降低了计算代价。计算机实验结果表明,该算法在优化性能和时间性能上都取得了良好的效果,满足战术决策的实时性要求。
A nonlinear integer-programming model for weapon-target assignment was proposed. The model took influence of joint damage probability of multi-weapon on actual fire effect into consideration. This NP-hard problem can be solved by basic ACA algorithm, but real-time requirement of tactical decision was difficult to satisfy sufficiently. So an adaptive method of updating pheromone and a mutation strategy of clearing up invalid assignment were introduced to increase convergent speed. The calculation cost was further reduced by method of deleting visited node instead of comparison taboo list, Computer experimental result showed that both optimization performance and time performance are improved by use of proposed algorithm, which meets the real-time requirement of tactical decision.
出处
《火炮发射与控制学报》
北大核心
2006年第4期1-5,11,共6页
Journal of Gun Launch & Control
关键词
整数规划
毁伤概率门限
火力分配
变异蚁群算法
integer-program
threshold of damage probability
weapon target assignment
mutation ant colony algorithm
