自适应模糊控制及其在机械臂系统中的应用研究

【作者】 宋睿卓；

【导师】 师五喜；

【作者基本信息】 天津工业大学 ， 控制理论与控制工程， 2008， 硕士

【摘要】 自适应模糊控制是模糊控制理论与自适应控制理论相互交叉、相互渗透而形成的一个研究领域。由于自适应模糊控制系统可以在运行过程中不断修正自己的控制规则来改善控制性能,并且收敛速度快、鲁棒性好,非常适用于具有非线性、多变量等性能的复杂系统建模与控制,因而受到控制界的广泛重视。首先根据模糊逻辑系统的万能逼近性,对一类非线性系统设计了模糊辨识器,通过在线辨识使模糊辨识器参数进行自适应调整。文中证明了所设计的辨识方案可以通过调整设计参数,可使辨识误差减小。其次本文针对一类单变量和多变量非线性系统提出了自适应模糊控制方法。在单变量非线性系统中,用模糊逻辑系统逼近控制器中的未知函数和未知函数的倒数;在多变量系统中,用模糊逻辑系统组成的矩阵逼近控制器中的逆矩阵以避免控制器的奇异性问题。两种情况都采用误差补偿项对模糊逻辑系统的逼近误差进行补偿。文中证明了两种情况下的控制方案不但能使闭环系统的所有信号有界,而且可使跟踪误差收敛到原点的小邻域内。最后在推导的机械臂动力学模型的基础上,把本文提出的多变量系统自适应模糊控制方法应用于此模型的控制中,仿真结果说明此控制方案可使机械臂达到满意的跟踪效果。更多还原

【Abstract】 Adaptive fuzzy control is a research field which is interactive theory between fuzzy control and adaptive control. Adaptive fuzzy control has recently attracted increasing attention from control circles, because the system of adaptive fuzzy control can not only improve system performance by modifying itself rules but also has quick convergence speed and robust. It is very suitable to model and control for nonlinear and multiplicity systems.Firstly, an adaptive fuzzy identification method for a class of nonlinear dynamic system was presented based on universe approximation of fuzzy logic system. The parameters of fuzzy identification can be adapted by identification on line. It is proved that the proposed method can reduce the identification error by adjusting the value of designed parameters.Secondly, adaptive fuzzy control methods are presented for a class of uncertain SISO nonlinear systems and MIMO nonlinear systems. For SISO nonlinear systems, two fuzzy logic systems are used to approximate unknown functions and inverse of unknown functions. For MIMO nonlinear systems, the matrix, which is composed of fuzzy logic systems, is used to approximate the matrix inverse in the controller in order to avoid the controller singularity problem. In the two situations, the error compensator is used to eliminate approximations errors. It is proved that the proposed schemes not only guarantees the boundedness of all signals in the closed-loop system, but also makes the tracking error converge to a small neighborhood of the origin.Lastly, adaptive fuzzy control method for MIMO nonlinear systems is used to control dynamic model of robot. Simulation results demonstrate the effectiveness of this method.更多还原