基于神经网络前馈的无人直升机非线性鲁棒控制设计

针对小型无人直升机的控制问题,设计了一种基于神经网络前馈的非线性鲁棒控制算法.算法主要由两部分组成:基于三层神经网络的前馈,用以补偿无人直升机姿态动力学模型中的不确定项;基于符号函数积分的鲁棒控制,用以补偿未知外界扰动;基于Lyapunov分析方法证明了控制器可实现姿态角的半全局渐近跟踪.在三自由度实验平台上对所设计的控制算法进行了实验验证,结果表明:提出的设计取得了较好的姿态控制效果,并对外界未知风扰具有较好的鲁棒性.

Design of nonlinear robust control of unmanned helicopter based on neural network feedforward

Aiming at control problem of micro unmanned helicopter,an algorithm of feedforward neural network-based non-linear robust control is proposed.The algorithm is composed of two parts,the first part is a three-layer neural network feedforward that is implemented for compensation of the uncertainties contained in helicopter attitude dynamic models;the other is robust control based on integration of signum function,which is developed for compensation of unknown external disturbances. Lyapunov-based analysis method guarantees the controller can realize semi-global convergence tracking of attitude angle.The proposed control algorithm is verified on a 3 degree-of-freedom helicopter testbed,which demonstrates the good control performance of the proposed control methodology and shows robustness in presence of unknown gust disturbance.