计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (5): 216-219.

• 工程与应用 • 上一篇    下一篇

改进RBFNN求机械臂轨迹跟踪的研究

魏 娟1,杨恢先2,谢海霞3   

  1. 1.湘潭大学 信息工程学院,湖南 湘潭 411105
    2.湘潭大学 材料与光电物理学院,湖南 湘潭 411105
    3.琼州学院 物理系,海南 五指山 572200

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-02-11 发布日期:2011-02-11

Research of trajectory tracking for manipulator using improved RBF neural network

WEI Juan1,YANG Huixian2,XIE Haixia3   

  1. 1.College of Information Engineering,Xiangtan University,Xiangtan,Hunan 411105,China
    2.Faculty of Material and Photoelectronic Physics,Xiangtan University,Xiangtan,Hunan 411105,China
    3.Department of Physics,Qiongzhou University,Wuzhishan,Hainan 572200,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-02-11 Published:2011-02-11

摘要:

为了使机械臂准确跟踪目标轨迹,达到控制精度高、实时性好的目的,提出一种改进的径向基函数(RBF)模糊神经网络算法。该算法采用模糊遗传算法在线调整神经模糊控制器的参数,对其参数进行改进和优化,同时采用最近邻聚类算法对控制器的模糊规则库进行更新。仿真结果表明,该算法与传统的神经网络算法相比具有较好的性能,学习速度快,跟踪精度高,并具有良好的控制性能和自学习能力。

关键词: 踪迹跟踪, 径向基函数神经网络, 模糊遗传算法, 最近邻聚类算法

Abstract: In order to make a manipulator track the target trajectory accurately and get good real-time performance,an improved RBF fuzzy neural network algorithm is put forward.This algorithm uses a novel fuzzy genetic algorithm to regulate the parameters of fuzzy controller and the parameters are optimized.Simultaneously the nearest neighbor clustering algorithms are used to refresh the fuzzy rules.In the simulation,compared with traditional fuzzy algorithms,this improved algorithm gets better performance,learning speed,high precision tracking and has good control performance and self-learning ability.

Key words: trajectory tracking, Radial Basis Function Neural Network(RBFNN), Fuzzy Genetic Algorithms(FGA), Nearest Neighbor Clustering Algorithms(NNCA)