International Journal of Robust and Nonlinear Control最新文献

This study epitomizes the problem of procuring targeted state tracking objectives while ensuring fault tolerance, disturbance suppression and reduced network load for nonlinear networked control systems represented based on interval type-2 fuzzy framework. In detail, a generalized extended state observer (GESO) is implemented for concurrent assessments of system states alongside providing the controller with estimations of actuator faults and external disturbances to counteract their detrimental effects on the system behaviour. Further, a model reference system incorporating fuzzy membership is included to construct the indented tracking control algorithm. Specifically, the state information extracted from GESO is propagated to the established tracking control protocol by means of adaptive event-triggered mechanism to reduce network load while ensuring peak performance. Altogether, the intended targets are achieved by devising a GESO-based adaptive event-triggered model-reference tracking controller. Notably, through the employment of fuzzy membership-linked Lyapunov–Krasovskii functionals, we set forth the necessary prerequisites for obtaining asymptotic stability for the configured system specified by linear matrix inequalities (LMIs). Certainly, to substantiate the utility of assembled controller, numerical findings are conveyed through graphical plots.

The problem of actuator and sensor fault estimation for a class of nonlinear multi-agent systems with multiple exogenous disturbances and stochastic noise is studied. By considering the centralized and the distributed output feedback simultaneously, the new intermediate observer and disturbance observer are designed to estimate the system state, actuator fault, and sensor fault simultaneously. In the designed observers, the observer matching condition is not required. Different from the existing results, it is assumed that different agents have different exogenous disturbance systems. Schur decomposition is introduced to reduce the computational complexity. Specifically, if the disturbance system parameters of different agents are the same, the calculation amount is the same as that for a single agent. Finally, the effectiveness of the proposed method is verified by simulations.

In this paper, the problem of robust adaptive iterative learning control (RAILC) with switching $��\sigma �$-modifications is addressed for a class of nonlinear systems with unrepeatable uncertainties and initial errors. Different from the published learning algorithms, switching $��\sigma �$ learning is introduced to ensure the robustness of the parametric estimation. The causal contradiction caused by the switching $��\sigma �$-modification is solved by applying an open-loop learning law. Sufficient conditions for initial rectifying functions (IRFs) are given for constructing prespecified tracking error trajectories, which are adopted in RAILC algorithms to cope with initial errors. S-class function with a series convergence sequence is utilized in the controller design to guarantee the perfect tracking performance. Simulations are given to compare the switching-$��\sigma �$ and the saturated learning that demonstrate the effectiveness of the proposed learning control scheme.

In this article, we focus on how to achieve semi-global exponential stability of stochastic complex-valued complex networks with distributed delay and semi-Markov jump (SCVCNDDSMJ) under impulse control. Considering complex-valued systems, stochastic disturbance, and semi-Markov jump, the segment functions in the Banach space are redefined and Driver's derivative for the functionals are redefined basically. Considering complex networks, we construct a global functional $��\overline{�V�}�$ by the solution component to the exponentially stabilized continuous time feedback control systems and some quasi-properties of $��\overline{�V�}�$ are derived. Combining impulse differential inequalities and mathematical induction, semi-global impulsive exponential stability can be obtained when $��\overline{�V�}�$ acts on the solution of the impulse control systems. We also provide some semi-global impulsive exponential stability criteria including the estimates of the average impulse control interval, the upper bound of the impulse control gain, and the expression between the bound of the initial condition and the maximum impulse interval without imposing restrictions on the bound of the initial condition. Moreover, the theoretical results obtained are utilized in a kind of stochastic complex-valued recurrent neural networks. In the end, the validity of the obtained theoretical results is confirmed by corresponding numerical simulations.

In this article, for a class of uncertain nonlinear systems, an adaptive predefined-time output feedback tracking control problem is investigated. Both the input signal and the output signal of the system are quantized for the sake of a lighter communication burden. First of all, a dynamic filtering technology is applied to overcome the virtual control signals being directly differentiated. Secondly, a well-constructed observer powered by quantized output and input is devised, which can help resolve the difficulty of immeasurable states. Thirdly, an adaptive output feedback control scheme is put forward under the backstepping design framework. Compared with the current predefined-time design methods, the adaptive law designed in this paper is expressed as a nonlinear differential equation, which ensures the predefined-time stability. It is shown that all the signals in a predefined time interval are bounded and the controlled system is practically predefined-time stable (PPTS). Ultimately, the effectiveness of the suggested control algorithm is illustrated through two examples.

In recent years, time-varying formations have emerged as a crucial control strategy due to their distinct advantages in adapting to environmental changes. However, achieving time-varying formation tracking in heterogeneous multi-agent systems via reinforcement learning (RL), especially when the leader's state is unavailable, remains a significant challenge. This paper investigates the problem of time-varying formation tracking for heterogeneous multi-agent systems (MASs) with unknown dynamics and inaccessible leader states. A fully distributed output-feedback reinforcement learning framework is developed, which integrates adaptive observer design, optimal regulation, and data-driven policy iteration into a unified control scheme. Specifically, a fully distributed adaptive observer is proposed to estimate the leader's state using only its output, without requiring Laplacian eigenvalues or global network knowledge. Based on this observer, an output-feedback reinforcement learning controller is constructed to achieve asymptotic convergence of the formation tracking error, in contrast to existing state-feedback-based methods that only ensure bounded errors. Furthermore, a state reconstruction mechanism, originally used in synchronization problems, is extended to time-varying formation tracking, enabling policy learning directly from input-output data under unknown dynamics. Theoretical analysis and simulation studies demonstrate that the proposed framework achieves robust, scalable, and model-free time-varying formation tracking, offering clear advantages over existing approaches.

This article considers the probabilistic-constrained tracking problem of a networked nonlinear system subject to hybrid attacks on the sensor and actuator channels. Two independent Markov-based attack models are proposed for characterizing a mixture of denial-of-service, deception and replay attacks. We generalize the attack models in the sense that Markov processes are more practical and complicated than Bernoulli ones, and the attacks at the actuator end are also considered, which prevents the control input from reaching the actuator. A resilient receding-horizon control law is designed based on the seriously tampered output, consisting of a probability-based observer and a convex optimization procedure for control parameters. It is capable of mitigating the attacks on both channels and fulfilling the tracking tasks by establishing a trade-off between the volume of the constrained set and the violation probability of the tracking error.

An event-triggered sliding mode adaptive anti-disturbance switching control scheme is presented for switched networked systems (SNSs) subjected to multi-source disturbances and DoS attacks. Firstly, a switching adaptive regulator is proposed to estimate unknown parameters of the switching neural network disturbance model. This adaptive regulator provides faster convergence and is applicable to the non-switched system case as well. Then, a switching adaptive estimator is constructed to approximate the switching neural network disturbance. Additionally, a resilient event-triggered mechanism (RETM) is applied, which not only mitigates the adverse effects of DoS attacks on communication but also facilitates the enhancement of communication resource utilization. Under the average dwell time (ADT) switching signals, the controller is built to reduce the effects of disturbances modeled by the neural network and alleviate the impact of unmodeled disturbances. At last, a switched RLC example is employed to display the efficacy of the proposed event-triggered sliding mode adaptive anti-disturbance switching control scheme.

This paper investigates the adaptive set-point tracking control problem for a class of nonlinear cascaded uncertain systems with output constraint. Three kinds of uncertainties are considered in the system of interest, including nonidentical unknown control directions, additive periodic disturbance, and unmodeled dynamics. Different from the existing results, it does not require that the signs of control coefficients be identical, nor the structure of the periodic disturbance be known a priori. In view of the presence of an additive periodic disturbance and unknown control directions, the cancellation-based feedback via backstepping is inapplicable in control design. Then, we develop a repetitive learning control strategy compensating for periodic disturbance regardless of unknown control directions. The changing supply rates technique for input-to-state stable (ISS) systems and a tan-type barrier Lyapunov function are synthesized to address unmodeled dynamics with output constraint. It is shown that the tracking error converges to zero within the specified constraint, and all signals in the closed-loop system are bounded in a semi-global sense. Finally, two examples, including a two-stage chemical reactor subject to periodic disturbance, validate our theoretical results.

This article investigates a periodic event-triggered control problem for a class of uncertain switched nonlinear systems with nonstrict feedback form. Based on sampling quantized output, a common neural observer is designed to deal with the unavailable states and unknown switching signals. Also by an input logarithmic quantizer, a novel quantized periodic event-triggered control strategy is proposed to further reduce the communication load between controller and actuator. Compared with the existing quantization control research, the dual sampling quantizer proposed for the first time overcome the disadvantage of continuously monitoring the quantizing conditions. Based on the common Lyapunov function theory and the backstepping technique, a neural event-triggered output feedback controller and the adaptive laws are obtained to achieve semiglobally uniformly ultimately bounded (SGUUB) stability of the closed-loop switched systems. The tracking error can converge to a small neighborhood of the origin under arbitrary switching. Finally, the effectiveness and the applicability of the developed control strategy are verified by some simulations of a numerical example and a practical one.