摘要
滑模控制响应快,对系统参数和外部扰动呈不变性,可保证系统的渐进稳定性,但其缺点是控制存在很强的抖动。在一般滑模控制的基础上引入径向基函数神经网络(RBFNN),利用滑模控制的特点设定目标函数,将切换函数作为RBFNN的输入,切换项增益K的绝对值作为其输出。利用RBF神经网络的在线学习功能,消除了控制的抖动,同时使系统具有很强的鲁棒性。对单关节机器人的仿真研究表明,在存在模型误差和外部扰动的情况下,该方案能大大降低抖振现象。
The sliding mode control has quick response and takes on invariability to system parameter and external disturbance, which can assure the asymptotic stability of system, but has the shortcoming of strong fluctuations of control. The radial basis function neural network （RBFNN） was introduced to the common sliding mode control, in which the switching function was regarded as the input of RBFNN while the output was the absolute value of the switching gain K, and the object function was set by using the characteristic of sliding mode control. The fluctuations of control of the system were eliminated by using the learning function of neural network, which make the system have strong robust. The results of simulation about a single-link robotic manipulator show that the scheme can diminish chattering of control under the condition of existing model error and external disturbance.
出处
《机床与液压》
北大核心
2009年第8期122-124,共3页
Machine Tool & Hydraulics
基金
河北省自然科学基金项目(F2007000223)
河北省科学技术研究与发展计划项目(07212106D)
