基于神经网络的机器人鲁棒跟踪控制

摘要

A robust control strategy based on neural network is proposed for trajectory tracking control of robotic systems specific to its parametric uncertainties and external disturbance.Robust compensation is designed to eliminate the effect of the uncertainties caused by the model parameters and disturbance, while a neural network is employed to learn the unknown upper bound of the uncertainty.The simulation results demonstrate that the proposed method can overcome the parametric uncertainties and disturbance effectively,with good robustness and control performance.%针对具有参数不确定以及外部扰动的机器人系统,提出了一种基于神经网络的鲁棒跟踪控制策略.鲁棒补偿控制器用于消除系统参数以及外部干扰引起的不确定性的影响,再利用神经网络学习系统不确定性未知上界.仿真结果表明,方法能有效克服机器人系统模型的不确定性和外部干扰,具有良好的鲁棒性和控制性能.