基于RBF逼近不确定项的机械手自适应控制研究

针对机械手控制系统中的不确定因素,提出了RBF神经网络逼近不确定项的自适应控制策略。在逆动力学计算力矩方法的基础上,设计了鲁棒自适应控制器。利用RBF神经网络对模型中的不确定项分块进行逼近,并用Lyapunov稳定性理论建立了网络权重自适应学习律,证明了系统的全局稳定性;最后进行了仿真,结果表明该方法能够有效的消除模型不确定性的影响,准确地实现了轨迹跟踪。

Research on Self-adaptive Control of Robotic Manipulator Based on Uncertainties Approximated by RBF

According to the uncertain factors in the control system of robotic manipulators,a self-adaptive control strategy based on uncertainties approximated by the RBF neural network was proposed. By means of computed torque control method based on inverse dynamics,the robust adaptive controller was developed. The block uncertainties of model was approximated by using RBF neural network,and the adaptive learning law of network weights was developed based on Lyapunov stability theory,the global stability of system was guaranteed：In the end,the results of simulation verified that the influence of model uncertainties can be effectively eliminated,the trajectory tracking was accurately realized.