知识引导的智能优化算法在航路规划中的应用  

Application of Knowledge-conducting Intelligent Optimization Algorithms to Path Planning

作　　者：刘钢[1] 老松杨[1] 侯绿林[1] 谭东风[1] 

机构地区：[1]国防科学技术大学信息系统工程重点实验室,湖南长沙410073

出　　处：《湖南大学学报（自然科学版）》2013年第1期103-108,共6页Journal of Hunan University:Natural Sciences

基　　金：国家自然科学基金资助项目(70571084;61074121);国家部委'十一五'科研计划资助项目(513040404-1)

摘　　要：为了提高智能优化算法对航路规划问题的求解质量和效率,提出了一种知识引导型智能优化算法的航路规划求解框架.与传统引导进化不同的是,考虑到以往所用知识的局限性,该框架并不是采用从前期优化过程中挖掘出来的知识,而是采用航路规划特定领域知识.为了描述引导方式,将智能优化算法形式化定义为3个引导对象的集合,从而将引导方式划分为7类单独或组合形式.根据航路规划特定领域知识的各自特点选择对应的引导方式,并将其结构化为能够改进算法性能的元策略.以粒子群优化算法为例对求解框架进行验证,仿真实验结果表明,特定领域知识的引导能够非常显著地提高算法的全局搜索性能和收敛速度.In order to improve the quality and efficiency of intelligent optimization algorithms for sol- ving path planning, the framework of knowledge-conducting intelligent optimization algorithms for solving path planning was proposed. Considering the limitaion of the knowledge adopted previously, the frame- work does not adopt the knowledge mined from previous iterations, which is different from the convention- al conducting evolution, but adopts the costomizing domain knowledge of path planning. In order to de- scribe the conducting manner, the intelligent optimization algorithms is defined formally as a set composed of 3 conducting objects, and the conducting manner is divided into 7 kinds of form which are separated or combined accordingly. A corresponding conducting manner is adopted according to the characteristic of dif- ferent costomizing domain knowledges of path planning, and these knowledges are transformd to struc- tured recta-strategy which can improve the performance of the algorithm. The solving framework is veri- fied by taking particle swarm optimization algorithm for instance, Simulation experiments results have in- dicated that the conducting of costomizing domain knowledge can improve the algorithm＇s global search ca-pabilities, and the algorithm possesses better convergence rate.

关 键 词：知识 引导进化 智能优化算法 引导对象 引导方式 航路规划 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]