基于模糊神经网络的机器人滑模自适应控制

滑模控制（SMC）响应快，对系统参数和外部扰动呈不变性，可保证系统的渐近稳定性，但其缺点是控制存在很强的抖动；而模糊神经网络（FNN）具有模糊系统和神经网络共同的特点。将滑模控制和模糊神经网络控制有机结合，利用简单得到的学习信号对模糊神经网络进行在线学习，通过平滑切换函数实现直接自适应控制策略。对两连杆机械手的仿真研究表明，在存在模型误差和外部扰动的情况下，该方案既能达到高精度快速跟踪的目的，又能有效减小滑模控制的抖动问题。

Sliding Mode Adaptive Control for Robots Based on Fuzzy Neural Networks

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 being strong chattering in control is the shortcoming.The fuzzy neural network(FNN) has common feature of fuzzy systems and neural networks.The sliding mode control and FNN are combined.Simple learning signal is added to FNN to realize direct adaptive control strategy through smooth switching function.The simulation results of a two-link robotic manipulator show that the scheme can achieve tracking effect with high precision and speediness,as well as minishing chattering of control under the condition of existing model error and external disturbance.