摘要
提出一种模糊神经网络,并将其应用于两关节机械手轨迹跟踪控制。该网络采用三角形隶属度函数,将模糊控制与神经网络相结合,利用神经网络实现模糊推理。这种模糊神经网络能够在线调节输出隶属度函数中心以及关节间耦合权值,使得控制器具有更好的学习与自适应能力。仿真结果表明,这种网络能很好的用于机器人的轨迹跟踪控制中,是一种行之有效的控制方法。
A fuzzy neural network is proposed for two-link robot manipulator tracking control.This network uses triangle-shaped membership function,fuzzy control and neural network are combined,and fuzzy inference can be realized by neural network.The fuzzy neural network can adjust output membership function centers and coupling weights of each joint on line,so that the controller has better learning and adaptive ability.The results of simulation show that this network can control robot tracking perfectly,and this control method is effective.
出处
《齐齐哈尔大学学报（自然科学版）》
2010年第4期1-5,共5页
Journal of Qiqihar University(Natural Science Edition)
关键词
模糊神经网络
三角形隶属度函数
两关节机械手
轨迹跟踪
fuzzy neural network
triangle-shaped membership function
two-link robot manipulator
tracking control
