Design of Neural Network-based Backstepping Controller for the Folding-Boom Aerial Platform Vehicle

Abstract

Abstract-In this paper, the robust trajectory tracking problem is addressed for the work platform of folding-boom aerial platform vehicle in the presence of uncertainties and disturbances. The control objective is to make the work platform move along a desired reference trajectory and make the vibration inhibit at the same time. Since neural network system can approximate any nonlinear function with arbitrary accuracy over a compact set in the light of the universal approximation theorem, a neural network-based backstepping controller, which composed of backstepping control and neural network, is proposed for the trajectory tracking control of the work platform in the case of modeling uncertainties and disturbances. According to Lyapunov stability theorem, the stability and convergence of the overall system can be guaranteed by the derived control law. In addition, simulation results demonstrate that the proposed controller is effective for suppressing the vibration and reducing trajectory tracking error of the work platform. Keywords-aerial platform vehicle; model uncertainties; trajectory tracking control; backstepping controller; neural network I、INTRODUCTIONFolding-boom aerial platform vehicle is a kind of device which can lift people to the height for installation or maintenance [1], the scheme of which is shown in Figure.1.As it requires vey high safety, both the stability of movement and the accuracy of positioning of work platform should be guaranteed.Since the light-long beam is widely used in the arm system, the influence of elastic deformations of beam should be considered. Therefore, in the literature [1], the beams are seen as flexible and the dynamics equations of the arm system of folding-boom aerial platform vehicle are set up based on flexible multi-body dynamics theory and Lagrange's equation. The establishment of the model lays foundation of the research of steady movement and accurate positioning of work platform.Due to the robust performance, integrator backstepping control has been applied to many nonlinear systems successfully, such as single link flexible manipulator [2] and multiple link rigid manipulator [3], etc. In addition, in [4], the backstepping control scheme has been used for the control of work platform of folding-boom aerial platform vehicle effectively. However, this method can only be used for the accurate dynamic model of the arm system of folding boom aerial platform vehicle.In fact, there exist various uncertainties due to external disturbances and approximation of the modeling [5,6]. As a result, a robust adaptive control scheme is developed for a class of uncertain nonlinear systems by the combination of backstepping control method and fuzzy control method in [7]. Moreover, as neural network can be used to approximate any nonlinear function over a compact set with arbitrary accuracy [8], it has attracted a wide spread attention. In [9,10], the combination of backstepping design and neural network has been used in the control of uncertain nonlinear systems.In this paper,...

A Backstepping Controller with the RBF Neural Network for Folding‐Boom Aerial Work Platform

“…Furthermore, the simulation results illustrate that the effectiveness of the presented controller for restraining the vibration and attenuating the tracking error when there exist model uncertainties. Although we had acquired the similar results in the previous article [24], we have made some improvements on the control performance in this study.…”