中文摘要：

针对挠性航天器带有执行机构饱和的姿态控制问题，提出了一种将自适应变结构和智能控制相结合的智能鲁棒控制方法。首先，在基于非线性和低阶模态的动力学模型基础上，针对挠性模态不可测的特点，给出了仅利用输出信息的智能自适应变结构输出反馈控制器的设计方法，其中利用神经网络控制来补偿执行机构饱和非线性和采用自适应控制技术克服确定不确定性的界的困难，并基于Lyapunov方法分析了滑动模态的存在性及系统的稳定性。最后，将该方法应用于挠性航天器的姿态机动控制，在反作用飞轮存在饱和特性约束的情况下，完成姿态机动的同时，可有效地抑制挠性附件的振动。

英文摘要：

In this paper, neural network （NN） based robust controller for rotation maneuver is considered for an orbiting flexible spacecraft, taking into account the actuator saturation. The actuator saturation is assumed to be unknown and treated as the system input disturbance. With the aniversal approximating property and /earning capability of NN, the NN - based saturation compensator is design and inserted into a feed-forward to compensate the saturation nonlinearity. Then an adaptive variable struc- turo controller, which only uses the output information, is designed to reject the disturbance, deal with uncertainty and ensure that the system trajectories globally uniformly ultimately bounded, in which need for knowing the system parameters in advance is eliminated by using an adaptive update technique. Systematic design procedure is also developed for synthesis of the NN adaptive variable structure attitude control strategy. To study the effectiveness of the corresponding control scheme, the traditional methods are also developed for the control system. Both analytical and numerical results are presented to show the theoretical and practical merit of this approach.