Self-Tuning Sliding Mode Control for an Uncertain Coaxial Octorotor UAV
针对存在执行器故障的不确定同轴八旋翼无人机，提出一种自整定滑模控制策略，结合梯度下降法和径向基神经网络，实现鲁棒自适应跟踪控制，并通过仿真验证有效性。
Sliding mode control (SMC), which is well known to its insensitivity to parameter uncertainties and robustness against various disturbances, has been widely adopted for some types of underactuated systems, including multirotor unmanned aerial vehicles (UAVs). This article is absorbed in developing a self-tuning SMC strategy to perform the robust and adaptive tracking control for an uncertain coaxial octorotor UAV (COUAV) with the actuator faults. Such a strategy is composed of manifold coefficients, which are adjusted by employing gradient descent method. It is also composed of continuous controllers that are designed by utilizing the radial basis function neural network (RBFNN) to approximate the total disturbances, where the approximation errors of RBFNN are estimated by using the adaptive control method. Moreover, the developed SMC strategy allows tuning the adaptive coefficients of the constructed manifolds, eliminating the inherent chattering phenomenon, and addressing the robust and adaptive tracking control problems for an uncertain COUAV. Comparative simulation results are carried out to illustrate the effectiveness of the developed SMC strategy.