School of Information Science & Engineering, Central South University, Changsha, Hunan 410083, China; School of Electrical & Information Engineering, Changsha University of Science & Technology, Changsha, Hunan 410004, China.
School of Information Science & Engineering, Central South University, Changsha, Hunan 410083, China; Hunan Engineering Laboratory for Advanced Control and Intelligent Automation, Changsha, Hunan 410083, China.
ISA Trans. 2014 Jan;53(1):173-85. doi: 10.1016/j.isatra.2013.08.010. Epub 2013 Sep 7.
The referenced quadrotor helicopter in this paper has a unique configuration. It is more complex than commonly used quadrotors because of its inaccurate parameters, unideal symmetrical structure and unknown nonlinear dynamics. A novel method was presented to handle its modeling and control problems in this paper, which adopts a MIMO RBF neural nets-based state-dependent ARX (RBF-ARX) model to represent its nonlinear dynamics, and then a MIMO RBF-ARX model-based global LQR controller is proposed to stabilize the quadrotor's attitude. By comparing with a physical model-based LQR controller and an ARX model-set-based gain scheduling LQR controller, superiority of the MIMO RBF-ARX model-based control approach was confirmed. This successful application verified the validity of the MIMO RBF-ARX modeling method to the quadrotor helicopter with complex nonlinearity.
本文所参考的四旋翼直升机具有独特的配置。由于其参数不准确、非理想对称结构和未知的非线性动力学，它比常用的四旋翼直升机更为复杂。本文提出了一种新的方法来处理其建模和控制问题，该方法采用基于多输入多输出（MIMO）径向基函数神经网络（RBF）的状态相关自回归（ARX）（RBF-ARX）模型来表示其非线性动力学，然后提出了一种基于 MIMO RBF-ARX 模型的全局线性二次调节器（LQR）控制器来稳定四旋翼直升机的姿态。通过与基于物理模型的 LQR 控制器和基于 ARX 模型集的增益调度 LQR 控制器进行比较，证实了基于 MIMO RBF-ARX 模型的控制方法的优越性。这一成功的应用验证了 MIMO RBF-ARX 建模方法对具有复杂非线性的四旋翼直升机的有效性。
