International Journal of Adaptive Control and Signal Processing最新文献

This work presents a robust model reference adaptive control with a full adaptive super-twisting sliding mode action (RMRAC-FASTSM) and its discrete-time stability analysis using the Lyapunov criterion. The control structure differs from the literature by using a couple of adaptive gains to automatically adjust the weights on the nonlinear and integral components of the STSM action. This strategy alleviates the chattering phenomenon because it reduces the sliding mode action when the system reaches the steady state, improving the closed-loop system performance and simplifying the controller design process by reducing the number of tuning hyperparameters required during this stage. The proposed control strategy emerges from the union of robust model reference adaptive control and a full adaptive super-twisting algorithm, resulting in a novel adaptive and robust single-loop control strategy. The stability analysis provides the controller design constraints, while also demonstrating that tracking error tends to a residual set in the steady state, even when the system is subject to matched and unmatched dynamics. Furthermore, the boundedness of all closed-loop signals is also proved. The resulting controller is able to control a partially modeled system, where the only required system information is the sign of the high-frequency gain and the relative degree of the nominal system transfer function. Simulation results corroborate the robustness of the controller dealing with relevant system perturbations and its superiority in relation to other adaptive super-twisting controllers from the literature.

The problems of accuracy-preassigned finite-time synchronization and global exponential synchronization for delayed chaotic neural networks (CNNs) using an adaptive controller are studied in this article. A direct method based on system solutions is used to establish the sufficient criteria for achieving synchronization. And the explicit expression of settling time (ST) is acquired. Compared with the previous methods, our method reduces the computation and simplifies the proof process, and the obtained sufficient conditions are less conservative. In addition, the convergence time of the error system is more accurate, and the gains of the adaptive controller are theoretically convergent. Three numerical examples with related simulations that demonstrate the significance of the obtained results.

In a hybrid robot mechanism, disturbances and significant payload variations with bounded torque inputs constitute a significant problem. In view of these uncertain dynamics, a new model-free robust adaptive control is developed using: nonlinear proportional-derivative ideal velocity feedback (NPD-IVF); novel inertia-gain adaptation control; and time-delay control (TDC). The proposed control scheme has four components: a time-delay estimation (TDE), a new NPD-IVF desired error dynamics, an inertia-gain adaptation law, and a saturation function (applied to guarantee torque input boundedness). The TDE component is used to get the lumped system dynamics thereby establishing a practical model-free framework. The NPD-IVF is designed to enhance the controller, reduce inherent estimation error and improve the position tracking performance. A gain-adaptation law exploits a neuro-fuzzy network that leads to a real-time dynamic update of the inertia gain matrix; this tolerates significant variations in payload and disturbances. Cumulative effects of these four components result in the proposed control being input torque bounded, model-free, improved accuracy and robust. Experimental comparisons were made using a prototype hybrid mechanism for vehicle electro-coating conveying with various payload conditions to verify the effectiveness of the proposed control.

The problem of secure observed-based guaranteed cost control for networked control systems with mixed anomalies is studied in this article, and the anomalies are considered to exist in both the actuators and the communication channels. The mixed anomalies have brought new challenges to the secure guaranteed cost control because of the coupling property and diversity modes of anomalies. Thus, a secure generalized extended state observer is presented for the estimation of the system state and actuator anomalies. Then the systems with communication-channel-anomalies is modeled as switched systems and the secure observer-based guaranteed cost control strategy is designed. Finally, simulation examples are given to show the effectiveness of the secure observer-based guaranteed cost control method.

This paper proposes an integrated observer-based fault estimation (FE) and fault-tolerant control (FTC) framework for discrete-time Takagi–Sugeno (T–S) fuzzy systems subject to actuator faults, external disturbances, and parametric uncertainties. A robust $��{�H�}_{�\infty �}�$ fuzzy observer reconstructs the system states and actuator faults, and these estimates drive a state-feedback controller that compensates for faults online while ensuring a prescribed $��{�H�}_{�\infty �}�$ performance. Sufficient linear matrix inequality (LMI) conditions derived using a Lyapunov–Krasovskii functional (LKF) and Finsler's lemma enable tractable synthesis of both the observer and the controller. The method was applied to a lateral vehicle model, which has uncertainties, disturbances, and sudden actuator faults. The results obtained demonstrate that the closed-loop system was stable and made accurate tracking. The results of the method implemented showed accurate estimator performance and robust compensation under faults, which should be highlighted by the credible performance when used in closed loop. The results show that the method has practical application for safety-critical nonlinear systems.

Usually, autonomous driving is considered to be one of the quickest-developing, recently, cutting-edge research domains, with huge market benefits in upcoming days. However, a few drawbacks occur because of the variety and size of the vehicle conditions that can be observed from the realistic world along with a limited number of cars traveling around. In this proposed method, a new adaptive model is introduced for the Multi-domain autonomous driving data classification which is more helpful for ensuring safety in autonomous driving and also guarantees driving control. The necessary input images are garnered from standard assets. Further, the input images are processed by the Retinex network for managing a climate and light circumstances. The class imbalance of the images is handled by performing the image augmentation process using the attention-based Generative Adversarial Network (GAN). Finally, the object detection and classification task is executed by Federated Learning-based Yolo v9 with an “Adaptive Dilated Residual Attention Network (Yolo-ADRAN).” The detection and classification pipeline is adapted to the adaptive network where a parameter of this method is tune by using a Randomly-emphasized Swarm Space Hopping Algorithm (RSSHA), thus it rapidly detects and classifies the objects as of the inputs. Lastly, a presentation of the method is confirmed using the experimental analysis.

This paper investigates the cooperative output regulation (COR) problem for switched multi-agent systems (MASs) with frequent asynchronism via an event-triggered mechanism (ETM). In order to reduce the conservatism of the traditional approach that allows only one switching during the event interval, the frequent asynchronous switching strategy is introduced into the MASs with switching dynamics. This eliminates the restriction on the minimum dwell time for individual subsystems, and thus permits to switch more frequently during a triggered interval. Subsequently, a distributed event-triggered dynamic compensator is designed. This compensator effectively facilitates the MASs to realize asymptotic tracking of the reference input while successfully rejecting the external disturbance. In contrast to the current achievements, the ETM presented in this article eliminates the need for any global information and avoids persistent communication between neighboring agents. Finally, the rationality of the proposed approach is confirmed by two examples.

This paper proposes an integral sliding mode control (ISMC) method based on preview repetitive control (PRC) for continuous-time nonlinear systems with the presence of matched uncertainties, external disturbances, and norm-bounded nonlinearities. First, a two-dimensional (2D) dynamic system is constructed. Second, the linear matrix inequality (LMI)-based condition is proposed, and the PRC law is designed for the nominal system. Then, a preview repetitive integral sliding mode control (PRISMC) law is obtained by combining an integral sliding mode controller with the preview repetitive controller, which ensures the robustness of the linear system. Finally, the effectiveness of the method is verified by a numerical simulation.

This paper proposes an adaptive multi-lane fusion control strategy for a 2-D plane vehicle platoon with delayed imposed distance constraints (DIDC) and actuator faults. DIDC is a novel distance constraint strategy that allows vehicles to operate freely at the initial stage and gradually enforces the constraint, thereby avoiding excessive early control efforts and enhancing flexibility in distance regulation. To handle the discontinuity caused by transitioning from no constraints to constraints, a shifting function and a transformed error function are designed, converting constrained variables into unconstrained ones while keeping the unconstrained part unchanged. Additionally, dead-zone and actuator fault inputs are linearized, and unknown nonlinearities functions are approximated by radial basis function neural networks (RBFNNs) with weight update laws based on gradient descent to ensure fast convergence. Sliding-mode controllers are developed for position and angle to achieve multi-lane fusion and maintain platoon stability. Simulations and comparative analyses validate the effectiveness of the proposed method.

This article focuses on the dissipative iterative learning control (ILC) for parabolic partial differential equation (PDE) systems. Firstly, with the aid of dissipativity analysis, the proposed P-type learning algorithm can achieve the perfect trajectory tracking based on $��(�{�L�}^{�2�}�,�s�)�$-norm along the iteration axis. Then, by employing the Wirtinger-based inequality, a novel linear matrix inequality (LMI)-based sufficient condition is manifested for the parabolic PDE systems to satisfy the dissipativity, and a solvability criterion of the LMI is also given. Furthermore, the dissipative ILC for parabolic PDE systems in the irregular case is investigated. Finally, the effectiveness of the proposed method is illustrated and validated via simulation examples.