基于改进的混合蛙跳算法机器人路径规划的研究

摘要

对于采用混合蛙跳算法进行机器人路径规划容易产生局部最优、收敛速度慢等问题，把人工鱼群算法融合到混合蛙跳算法。人工鱼群算法的追尾行为有利于加快收敛速度，群聚行为提高了跳出局部最优解的能力。把机器人路径规划问题转化为最小化问题，使用改进的算法对其进行优化，最终得到全局最优路径。仿真结果表明，与改进前的算法相比，改进后算法具有较强的全局寻优能力和较快的收敛速度。%In order to overcome the defects of ShuffIed Frog Leaping AIgorithm (SFLA) such as sIow searching speed in the Iate evoIution and easiIy trapping into IocaI extremum.An improved SFLA based on the basic idea of ArtificiaI Fish-Swarm AIgorithm (AFSA) is put forward in this paper in which the foI ow behavior of fish swarm is used to acceIerate the optimization speed and the swarm behavior to improve the capacity of out of IocaI extremum.This paper transforms the probIem of robot path pIanning into a minimization one,using the improved aIgorithm to optimize and uItimateIy getting the gIobaI optimaI path.