受到外部扰动的空间机械臂基于模糊递归神经网络的控制策略

Control method for space manipulator with external disturbance based on recurrent fuzzy neural networks

针对载体位置不受控、姿态受控的空间机械臂运动控制问题,对具有不确定系统参数的空间机械臂在受到外部扰动工况下动力学建模以及控制算法进行了研究。运用拉格朗日第二类动力学方程及被控系统动量守恒关系,对该工况下空间机械臂运动特性进行了分析,建立了系统在受到外部扰动时的动力学模型,并对动力学特性进行了归纳,提出了一种基于模糊递归神经网络的控制策略。根据李雅普诺夫第二类方法,证明了整个闭环控制系统的运动渐进稳定性。使用计算机数值仿真实验与计算力矩法进行对比,阐述了整个控制系统的有效性。研究结果表明,在存在外部扰动及系统参数不确定时,所设计的控制方案相比于传统计算力矩控制方法具有更好的轨迹跟踪效果和更高的可靠性。

Aiming at the control problem of space manipulator system with external disturbance and uncertain parameters when the attitude of based was controlled and its location was uncontrolled,combining with the system of conservation of momentum,the system's dynamic model under external disturbance was established by using the Lagrange dynamics equation and assumed modes methods were investigated. A method based on recurrent fuzzy neural networks was proposed. This method could solve space manipulators with external disturbance and uncertain parameters control problems,and through the second Lyapunov method to prove the asymptotic stability of the closed-loop control system.The results indicate that this control strategy has better tracking performance and higher reliability than the computed torque control method through the computer numerical simulation experiment.