摘要
针对结构化环境中移动机器人路径规划问题,提出一种基于粒子群的路径规划算法。该算法利用适应度函数描述环境约束及路径的距离信息,适应度函数通过神经网络计算;由路径节点构成粒子,通过混合粒子群算法进行寻优。最后,通过计算机仿真验证了该算法是合理的,并且可应用于机器人的实时导航。
A novel path planning approach based on particle swarm was presented aiming at mobile robots in structured environments. The information of environment constrains and path length was integrated in the fitness function which was computed by neural net work, the path nodes was viewed as a particle, so with the quality of optimization of hybrid particle swarm algorithm, a best path was found. Finally by computer simulation, it is proved that the algorithm is rational and can be used in mobile robot real-time navigation.
出处
《计算机应用研究》
CSCD
北大核心
2007年第3期181-183,186,共4页
Application Research of Computers
基金
国家"863"计划资助项目(2005AA420050)
关键词
路径规划
粒子群
适应度函数
神经网络
path planning
particle swarm
fitness function
neural network
