基于广义双曲正切模型的机器人模糊自适应控制

利用广义模糊双曲正切模型的全局逼近特点，设计一种直接模糊自适应控制器用于机器人轨迹跟踪控制．模糊控制器的输入变量经过平移变化后得到的广义输入变量能够覆盖整个输入空间，因此，模糊控制器能以任意精度逼近系统的最优控制；由于双曲正切模型的特殊结构，在保证跟踪精度的同时，避免了因模糊子集数目增加而带来的计算负担的增加，满足了机器人实时控制的需要．仿真结果表明，该控制算法具有较强的鲁棒性和较好的跟踪性．

Fuzzy Adaptive Control of Manipulators Based on the Generalized Hyperbolic Model

A universal approximator and a direct fuzzy adaptive controller are proposed for the trajectory tracking control of robotic manipulator by taking advantage of the property of the generalized hyperbolic model. The input generalized variables of the fuzzy controller, derived from the translation of the input variables can cover the whole input space. So, this controller can approximate the optimal control by arbitrary accuracy. Furthermore, due to the special structure of the fuzzy hyperbolic model, the control strategy avoids a heavy computation burden when the fuzzy subsets are augmented hut taking a satisfied tracking performance. It meets the requirement of robotic manipulator real-time control. Simulation studies show that this control algorithm possesses strong robustness and satisfactorg tracking accuracy.