航天器姿态的神经网络动态逆控制

针对航天器姿态系统,提出了一种基于自适应神经网络动态逆的控制算法。该算法针对滚转、俯仰和偏航三个姿态子系统,设计了两组神经网络:第一组是BP网络,用来逼近三个姿态通道的非线性项,可获得姿态逆模型;第二组是非线性自适应神经网络,用于在线实时地补偿逆模型存在的误差和外加干扰。详细分析了非线性自适应神经网络的拓扑结构、学习规则和调整算法。给出了应用该算法的具体实例,通过仿真实验证明该算法的有效性。

Adaptive neural-network-based inverse method and its application in spacecraft attitude control

To study the attitude system of spacecraft,an adaptive inverse control algorithm is presented.Two kinds of neural networks are designed for each attitude sub-system with a BP network approximating the inverse dynamics and another adaptive neural network compensating the inverse error and disturbances.The adaptive network weights are updated using a Lyapunov-based design.The topologies of these networks are investigated and the learning rules are analyzed.The effectiveness of the proposed method is demonstrated through the simulation on a satellite attitude control system.