不同重力环境下空间机械臂神经自适应鲁棒控制

Neural Adaptive Robust Control of Space Manipulator under Different Gravity Environment

针对空间机械臂从地面装调到空间应用过程中重力项的变化问题,提出了一种神经网络自适应鲁棒补偿控制策略用于空间机械臂的末端控制,从而实现在地面重力环境下装调好的空间机械臂在空间微重力环境下实现在轨操控任务。通过神经网络在线建模来逼近系统模型中变化的重力项,逼近误差及系统的不确定性通过自适应鲁棒控制器来补偿。该控制策略不依赖于系统的模型,避免了回归矩阵的复杂计算及未知参数的估计,降低了计算量。基于李亚普诺夫理论证明了闭环系统的渐近稳定性。仿真结果表明该控制器对不同重力环境下空间机械臂的末端控制均能达到较高的控制精度,具有重要的理论研究和工程应用价值。

Considering change of gravity items from ground alignment under gravity environment to space applications under microgravity environment,a neural network adaptive robust control strategy is proposed for end control of space manipulator,so as to achieve the on-orbit tasks in space under microgravity environment for the space manipulator adjusted on the ground under gravity environment.The control scheme uses neural networks to approach the gravity items of system model on line.Approach errors and system uncertainties can be compensated by using an adaptive robust controller.The control strategy can not depend on the model,avoids the complex calculations of the regression matrix and the estimation of unknown parameter and reduces the computational complexity.The control scheme can guarantee the stability of closed loop system and the asymptotic convergence of tracking errors based on the Lyapunov theory.The simulation results show that the controller is effective in control accuracy for end control of space manipulator under different gravity environments,and has important value for theoretical research and engineering application.