Tracking Control of Quadrotor by using Adaptive Sliding-Mode Control based on Chebyshev Neural Networks

In this paper, a method is proposed for the control of a quadrotor based on sliding mode control by using Chebyshev neural networks. The proposed approach is a combination of the sliding mode controller and the Chebyshev neural network approximator that the neural network weights are tuned in real-time by using robust adaptive techniques. In this research, the dynamic model of the quadrotor is divided into two subsystems for the purpose of the position and orientation tracking control: a fully-actuated subsystem and an underactuated subsystem. For the former, the sliding surfaces are designed by using one state variable, and for the latter, the sliding manifolds are defined by a linear combination of two state variables. In this paper, the system stability is analyzed by Lyapunov theory-based techniques and the accuracy of the controller performance will be illustrated by the simulation results.