Fast Model-Free Learning for Controlling a Quadrotor UAV With Designed Error Trajectory

Abstract

Traditional model-based control methods typically require accurate system dynamics. However, when controlling a complex non-linear system such as a quadrotor unmanned aerial vehicle (QUAV), the dynamics are unknown and it is challenging to tune the control parameters manually. This paper proposes a novel model-free learning method that combines the advantages of a model-based method, i.e., sliding mode control (SMC), with the iterative learning control (ILC) method. Specifically, we selected a designed sliding surface to obtain the expected tracking error trajectory as the learning objective, and the system tracking errors of the angles of the QUAV constitute the state space. Then, the policy of converging to the sliding surface is learned by an ILC algorithm. We have provided theoretical proof of the convergence, and validated the proposed method with real-world experiments where the sine wave signals of roll and pitch angles were tracked. The results have demonstrated the effectiveness of the method with less tracking errors as well as faster learning speed compared with a baseline PID controller and a sliding mode controller.INDEX TERMS UAV, sliding mode control, iterative learning control, non-linear control, model-free.

Robot Motion Planning Based on Improved RRT Algorithm and RBF Neural Network Sliding

“…Yin et al [22] introduced an adaptive sliding mode control based on lateral deviation, effectively reducing the chattering phenomenon. An et al [23] combined iterative learning and sliding mode control, resulting in superior trajectory tracking with smaller tracking errors and faster learning rates compared to traditional methods. Addressing high-precision trajectory tracking for robotic arms, Xian et al [24] proposed a continuous sliding mode control (CSMC) scheme based on time-varying disturbance estimation and compensation, avoiding the chattering of traditional sliding mode control and enhancing disturbance resistance.…”

“…For the challenges posed by the complexity of a 2-DOF robotic arm's real-world operations and uncertainties in system parameters, making accurate mathematical modeling problematic, this paper introduces a sliding mode control based on RBF neural networks. The RBF neural network approximates uncertainties in the model [23], while the sliding mode controller adjusts parameters in real-time. Drawing upon the Lyapunov stability principle, the paper proves the asymptotic stability of the designed RBF-SMC controller, enabling the robotic arm system to converge to the desired trajectory within a finite time and achieve highprecision tracking performance.…”