基于神经网络的自由漂浮空间机器人鲁棒控制

针对带有模型误差及外界扰动的自由漂浮空间机器人轨迹跟踪问题,提出了一种基于神经网络的自适应鲁棒控制策略。采用对神经网络状态空间进行划分后与滑模变结构结合的控制器，对不确定非线性进行自适应学习，逼近误差作为外部干扰由鲁棒控制器消除。该方法从整个闭环系统的稳定性出发，利用H_∞理论设计的鲁棒控制器及神经网络权值的在线调整规则保证了系统的稳定性，并能使系统L2增益小于给定的指标，具有较好的控制精度及动态特性。仿真分析进一步证明了该自适应鲁棒控制算法的有效性。

Robust control of free-floating space robot based on neural network

The trajectory tracking of a class of free-floating space robot manipulators with disturbance and model parameter uncertainties is considered. An adaptive robust control algorithm based on neural network is proposed. The state sections of a neural network are compartmentalized. It is used to adaptive learn and compensate the unknown system by syncretize with sliding model variable structure controller, and approach errors as disturbance are eliminated by a robust controller. The robust controller based on H_∞ theory and weight adaptive laws on-line based on Lyapunov are obtained. Above these assure the stability of the whole system, and L2 gain is also less than the index. 〖JP2〗The control scheme possesses great control accuracy and dynamic function. The simulation results show that the presented adaptive robust control algorithm based on neural network is effective.