Polymer electrolyte membrane fuel cell (PEMFC) air-feed subsystem is usually affected negatively by system uncertainty and unavailable variable. This paper investigates a discrete neural network control with an observer for the oxygen excess ratio (OER) control. The control goal is to avoid oxygen starvation and maintain the optimal net power. Specifically, by utilizing coordinate transformation and Euler approximation, the discrete strict-feedback form is obtained, and the backstepping technique can be applied. To estimate the unavailable variable with the measurable system output, a discrete neural network observer is proposed. Besides, the discrete neural network controller is designed to tackle the system uncertainty and achieve an ideal OER tracking. Finally, the system tracking error is proved to be semi-globally uniformly ultimately bounded by Lyapunov stability theory. Numerical simulations and hardware-in-loop (HIL) experiments are presented to demonstrate the effectiveness and superiorities of the proposed controller.

Polymer electrolyte membrane fuel cell (PEMFC) air supply systems are usually affected negatively by model uncertainties, external disturbance and unmeasured variables. In this paper, we propose a composite adaptive type-2 fuzzy controller based on a high-gain observer and a disturbance observer for oxygen excess ratio (OER) of PEMFC air supply systems. Firstly, the derivatives of system output, which are unavailable due to limited sensors, are estimated via the high-gain observer. Then, interval type-2 fuzzy logic systems (IT2 FLSs) are adopted to approximate the unknown system dynamics and the disturbance observer is designed to estimate compound disturbance including unknown external disturbance and fuzzy approximation error. Finally, in order to improve the tracking performance, two composite adaptive updating laws are constructed by utilizing the estimated tracking error and the modeling error. Theoretical analysis shows that the system tracking error is uniformly ultimately bounded by Lyapunov stability theory. Numerical simulations and hardware-in-loop (HIL) experiments are presented to demonstrate the effectiveness and superiority of the proposed controller.

This article addresses the steering control problem for steer-by-wire (SbW) systems subject to the unknown uncertainty, external disturbance and unavailable variable. Before the controller design, an adaptive neural network-based observer and a disturbance observer are constructed to estimate the angular velocity signal and the compound disturbance, respectively. Then, to guarantee the transient and steady-state performance of steering tracking error within the quantitative boundary, a prescribed performance function is constructed by user-designed tracking accuracy and settling time. Finally, the controller is designed based on the backstepping scheme and the neural network with a composite learning scheme is proposed for the approximation of lumped uncertainty. The Lyapunov stability theory shows that the signals involved in the system are semiglobal uniformly ultimately bounded and the tracking error converges to a preset range at finite time. Different numerical simulations and experiments are implemented to verify the effectiveness of the developed control scheme.

This technical brief emphasizes on the control of polymer electrolyte membrane fuel cell (PEMFC) air supply system. The control objective is to improve the net power output through adjusting the oxygen excess ratio within a reasonable range. In view of the problem that the PEMFC air supply system is difficult to achieve accurate modeling and stable control, a robust adaptive controller is proposed by utilizing exact linearization and radical basis function (RBF) neural network (RBFNN) system. This controller does not need the complete structure and parameters of PEMFC system. The unmodeled dynamics of PEMFC system can be approximated by RBFNN in which the adaptive learning law can be derived based on Lyapunov theory, and the external disturbance as well as the approximation error of RBFNN can be attenuated through robust control. The stability analysis shows that the system tracking error is uniformly ultimately bounded. Finally, the effectiveness and feasibility of controller are validated by hardware-in-loop (HIL) experiment.

A crucial role is played by steering-angle prediction in the control of autonomous vehicles (AVs). It mainly includes the prediction and control of the steering angle. However, the prediction accuracy and calculation efficiency of traditional YOLOv5 are limited. For the control of the steering angle, angular velocity is difficult to measure, and the angle control effect is affected by external disturbances and unknown friction. This paper proposes a lightweight steering angle prediction network model called YOLOv5Ms, based on YOLOv5, aiming to achieve accurate prediction while enhancing computational efficiency. Additionally, an adaptive output feedback control scheme with output constraints based on neural networks is proposed to regulate the predicted steering angle using the YOLOv5Ms algorithm effectively. Firstly, given that most lane-line data sets consist of simulated images and lack diversity, a novel lane data set derived from real roads is manually created to train the proposed network model. To improve real-time accuracy in steering-angle prediction and enhance effectiveness in steering control, we update the bounding box regression loss function with the generalized intersection over union (GIoU) to Shape-IoU_Loss as a better-converging regression loss function for bounding-box improvement. The YOLOv5Ms model achieves a 30.34% reduction in weight storage space while simultaneously improving accuracy by 7.38% compared to the YOLOv5s model. Furthermore, an adaptive output feedback control scheme with output constraints based on neural networks is introduced to regulate the predicted steering angle via YOLOv5Ms effectively. Moreover, utilizing the backstepping control method and introducing the Lyapunov barrier function enables us to design an adaptive neural network output feedback controller with output constraints. Finally, a strict stability analysis based on Lyapunov stability theory ensures the boundedness of all signals within the closed-loop system. Numerical simulations and experiments have shown that the proposed method provides a 39.16% better root mean squared error (RMSE) score than traditional backstepping control, and it achieves good estimation performance for angles, angular velocity, and unknown disturbances.