基于高斯基的广义CMAC在机器人轨迹跟踪控制中的应用

General CMAC Model for Gaussian Functions' Controller and its Application on Robotic Tracking Control

介绍了基于高斯基的广义小脑模型关节控制器(CMAC)的快速算法,实现CMAC学习速率的自适应模糊调整,使其能根据控制系统的动态特性优化CMAC的学习控制,以适应控制系统工况的变化和不确定因素的干扰,并把它与自适应神经元PID控制器相结合实现前馈反馈复合控制,避免了控制器的输出产生振荡或进入饱和状态,增强了系统的鲁棒性.仿真结果表明:这种控制方案能够准确的实现系统的高精度实时位置跟踪和力的控制.

Fast-algorithm of general cerebellar model articulation controller（CMAC） based on Gaussian functions is introduced, and the learning-rate of CMAC is adjusted using fuzzy algorithm so that it can be adapted to the change of the dynamic feature of the control system. It is also applied on robotic together with adaptive PID, so the control strategy can avoid controller＇s output producing surge or entering saturation condition and it enhances the robust of the system. Simulation results show： the presented method can improve the system＇s robust and real time performance.