基于RBF神经元网络的漂浮基空间机械臂关节运动自适应控制方法

讨论了栽体位置、姿态均不受控制情况下,漂浮基空间机械臂关节运动的控制问题.由拉格朗日第二类方法及系统动量、动量矩守恒关系,建立了漂浮基空间机械臂完全能控形式的系统动力学方程.以此为基础,借助于RBF神经网络技术、Ge-Lee(GL)矩阵及其乘积算子定义,对漂浮基空间机械臂进行了神经网络系统建模;针对空间机械臂系统所有惯性参数均未知的情况,设计了漂浮基空间机械臂关节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通常的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、学习,此外,由于充分利用了空间机械臂的系统动力学特性,因此在控制过程中不需要反馈、测量漂浮基的位置、移动速度、移动加速度以及姿态转角的角速度、角加速度.一个平面两杆漂浮基空间机械臂的系统数值仿真证实了该方案的有效性.

Adaptive Control for Free-floating Space Manipulator to Track Trajectory in Joint Space Based on RBF Neural Network

The control problem of free-floating space manipulator to track trajectory in joint space was discussed. Firstly, with the Lagrangian formulation and the relationship of the system linear, angular momentum conversations, the full-controlled dynamics equation of the system was 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 was approximated. And then with all unknown inertial parameters of the manipulator, the adaptive neural network control scheme for free-floating space manipulator to track trajectory in joint space was developed. It needs neither linearly parameterize the dynamics equation of the system, nor knows any actual inertial parameters. And the neural network needs not do training and learning online. In addition, because of making full use of the dynamics property of the system, the proposed control scheme also needs not feedback and measure the position, the linear and angular velocity and acceleration of the free-floating base. Lastly, a planar free-floating space manipulator was simulated to verify the effectiveness of the proposed control scheme.