漂浮基姿态受控空间机械臂关节运动的自适应神经网络控制

讨论了载体位置无控、姿态受控情况下,空间机械臂姿态及关节协调运动的自适应神经网络控制问题.由拉格朗日第二类方法及系统动量守恒关系,建立了漂浮基空间机械臂的系统动力学方程.以此为基础,借助于RBF神经网络技术和GL矩阵及乘积算子定义,对空间机械臂系统进行了神经网络系统建模;之后针对空间机械臂所有惯性参数未知的情况,设计了空间机械臂载体姿态与机械臂各关节协调运动的自适应神经网络控制方案.提出的控制方案不要求系统动力学方程具有关于惯性参数的线性性质,且无需预知系统惯性参数的任何信息,也无需对神经网络进行离线训练和学习,因此更适于实时应用.通过对一个平面两杆自由漂浮空间机械臂系统的数值仿真,证实了方法的有效性. 

Adaptive Neural Network Control of Free-floating Space Manipulator With an Attitude Controlled Base

The control problem of coordinated motion of free-floating space manipulator with an attitude controlled base is discussed. Firstly, with the Lagrangian formulation and the relationship of the system linear momentum conversation, the dynamic equation of system is derived. Based on the above results and the RBF neural network technique, the Ge-Lee (GL) matrix and its product operator, the dynamics of free-floating space manipulator is approximated. And then with all unknown inertial parameters of the manipulator, the adaptive neural network control scheme of coordinated motion between the base's attitude and the manipulator's joints of free-floating space manipulator is developed. It need neither linearly parameterize the dynamic equation of system, nor know any actual inertial parameters. And the neural network need not do training and learning online too. Therefore, the control scheme is prone to real-time application. Lastly, a planar free-floating space manipulator is simulated to verify the proposed control scheme.