论文部分内容阅读
针对机械手存在的扰动等未知模型,提出了基于RBF神经网络的自适应控制策略。采用RBF神经网络对机械手动力学模型在线自学习,并根据Lyapunov稳定性理论建立了网络权值自适应学习律,确保了网络逼近误差的收敛及系统的稳定。以平面转动双臂机械手轨迹跟踪为例进行仿真,结果表明该方法能够有效地补偿建模误差,实现了无需模型的机械手自适应控制,提高了系统的控制性能及对外部不确定扰动的鲁棒性,对实际工业机械手的自适应控制具有一定的可操作性。
Aiming at the unknown model such as disturbance of manipulator, an adaptive control strategy based on RBF neural network is proposed. RBF neural network is used to self-learn the robot dynamics model. Based on the Lyapunov stability theory, an adaptive learning algorithm of network weights is established to ensure the convergence of network approximation errors and the stability of the system. The simulation results show that this method can effectively compensate the modeling error, and realize the robot adaptive control without model and improve the control performance of the system and the robustness against external uncertain disturbance It has certain maneuverability to adaptive control of real industrial manipulator.