摘要
无人机(UAV)指派问题是一种具有多约束条件的复杂任务分配问题。随着问题规模和约束数量的增加,其复杂性加剧,尤其是对于目前常用的,基于线性规划类的方法而言,存在着维数爆炸和优化求解困难加剧的问题。提出了一种通用的UAV任务指派模型,将UAV指派问题转化为多约束条件下的优化问题。该模型通过构造可行解的方法,不但有效地减小了搜索空间,提高了搜索效率,而且适用于各种计算智能类的优化方法。通过4种典型的计算智能优化方法,即粒子群优化方法、遗传算法、差分进化算法和克隆选择算法的数值分析,结果表明该模型具有更好的适应性和可扩展性,与计算智能优化方法相结合,能有效地求解复杂UAV任务指派问题。
The task allocation of unmanned aerial vehicle (UAV) is a complex assignment problem with various constraints. With the increase of the size of scenarios and the number of constraints, UAV assignment problems become more complicated. Especially, the potential dimensional explosion and optimization difficulty are unavoidable to those algorithms based on linear programming. A new UAV assignment model was proposed, which transforms the UAV assignment problem into a multi- constraint optimization problem. The proposed model reduces the dimension of solution space effectively, improves the optimization efficiency, and is adapted to the other computational intelligence algorithms. Several computational intelligence algorithms, such as particle swarm optimization, genetic algorithm, differential evolution algorithm, clonal selection algorithm, were applied to accomplish the optimization work. Numerical experimented results illustrate that the model has better adaptability and extcnsibility, can solve complex UAV assignment problems combined with the computational intelligence algorithms.
出处
《兵工学报》
EI
CAS
CSCD
北大核心
2009年第12期1706-1713,共8页
Acta Armamentarii
基金
国家自然科学基金项目(60903005)
关键词
运筹学
无人机指派问题
粒子群优化方法
遗传算法
差分进化算法
克隆选择算法
operational research
unmanned aerial vehicle assignment problem
particle swarm optimization
genetic algorithm
differential evolution algorithm
clonal selection algorithm