Intelligent Adaptive Backstepping Control System for Magnetic Levitation Apparatus

Abstract

We propose an intelligent adaptive backstepping control system using a recurrent neural network (RNN) to control the mover position of a magnetic levitation apparatus to compensate for uncertainties, including friction force. First, we derive a dynamic model of the magnetic levitation apparatus. Then, we suggest an adaptive backstepping approach to compensate disturbances, including the friction force, occurring in the motion control system. To further increase the robustness of the magnetic levitation apparatus, we propose an RNN estimator for the required lumped uncertainty in the adaptive backstepping control system. We further propose an online parameter training methodology, derived by the gradient descent method, to increase the learning capability of the RNN. The effectiveness of the proposed control scheme has been verified by experiment. With the proposed adaptive backstepping control system using RNN, the mover position of the magnetic levitation apparatus possesses the advantages of good transient control performance and robustness to uncertainties for the tracking of periodic trajectories.Index Terms-Adaptive backstepping control, lumped uncertainty, magnetic levitation, recurrent neural network.

Cascade Control of Magnetic Levitation with Sliding Modes

“…Many magnetic levitation control design have been reported in the literature, including feedback linearization based controllers [4,6,[9][10][11], linear state feedback control design [6,12], the gain scheduling approach [13], observer-based control [5], neural network techniques [14], sliding mode controllers [8,15,16], backstepping control [17], model predictive control [18], cascade control [19] and PID controllers [20]. Since the governing differential equations are highly nonlinear, the nonlinear controllers are more attractive.…”

Performance analysis of Interval Type-2 FSM controller applied to a magnetic levitation system

“…Since these systems are able to suspend any object in air, by using sensors for observing position of levitated objects and then giving controlled amount of current to electromagnet, an object can be held into air.MLS can be connected in numerous application regions, for example, diverse fields such as frictionless deportments, extraordinary swiftness maglev trains, hovering of wind tunnel models, maglev anti-vibration systems, etc.In MLS, the presence of high nonlinearity and parameter uncertainty make it difficult to control the ball position through conventional controller. This is ambitious and fascinating job for control designer and analyst to control given MLS system [1][2].…”

Adaptive backstepping control for a class of semistrict feedback nonlinear systems using neural networks

“…As is well-known, neural networks (NNs) are always used as approximators for unknown smooth functions [1,2,4] in indirect adaptive control. Therefore, many strict feedback systems that are not in linear-in-parameters forms [8] can also be solved by using NNs.…”