Asian Journal of Control最新文献

An efficient identification method is proposed to estimate the time-varying parameters of a nonlinear dual-control aircraft. Based on the Extended Kalman Filter (EKF), the long-short-term memory (LSTM) neural network is introduced to adjust the filter gain to reduce the filter error caused by model uncertainty and linearization. To solve the heavy training burden of the stochastic gradient descent (SGD), RMSprop, and Adam training algorithm, an efficient training algorithm is presented, which uses Independent Extended Kalman Filter (IEKF) to transform the network weights training problem into an optimal weights identification problem to achieve the convergence speed of second-order training. To further reduce the training burden of the network, the network weights Matrix Factorization (MF) is introduced to reduce the number of learnable weights. In addition, the training performance of the LSTM neural network using MF structure and IEKF training algorithm is analyzed by the Lyapunov method, and the selection method of the artificial process noise matrices is given. Time-dependence data fitting experiments show that the efficient LSTM neural network proposed herein can greatly reduce the training burden of the network, and simulation results show that the Learnable Extended Kalman Filter (L-EKF) achieves better performance than some existing filtering methods in the time-varying parameter identification of the dual-control aircraft.

This paper focuses on the issue of adaptive event-triggered containment control for Markov jump multi-agent systems characterized by hidden Markov jump parameters. The central objective is to design an output-feedback controller for the Markov jump multi-agent system by using an adaptive event-triggered technique that not only ensures the containment control but also optimizes the utilization of network resources. The proposed method involves the use of a hidden Markov model to express the asynchrony between the states of the original Markov chain and the constructed asynchronous controller. By devising an appropriate Lyapunov–Krasovskii functional, we obtain the adequate conditions for solving the containment control problem for the chosen multi-agent system, presented as linear matrix inequalities. Ultimately, two numerical examples are given to substantiate the efficacy of the obtained theoretical results.

This paper develops an adaptive memory event-triggered fault detection approach for a class of continuous-time switched linear systems with persistent dwell-time (PDT) switching. Unlike conventional dwell-time or average dwell-time switching signals, the more common PDT switching signal is considered, which can describe more switching behaviors of dynamic systems. We aim to design an event-triggered based mode-dependent fault detection filter, thereby enabling the augmented system that satisfies the desired nonweighted $��{�H�}_{�\infty �}�$ performance. Different from the existing event-triggered fault detection mechanism, the proposed method has two advantages, that is, one is that it avoids long time without data transmission by designing the minimum real-time transmission rate and the other one is that it can reduce the false triggering caused by unknown abrupt disturbance and measurement noise. Finally, the effectiveness of the proposed fault detection approach is verified by two examples including a switched RLC circuit.

The time-varying nature of wireless channels leads to alternating data losses and fading, which pose significant challenges for ensuring the tracking accuracy of remotely controlled objects using iterative learning strategies. To address this challenge, we developed a filter to estimate the input conveyed by controllers through the time-varying channel. First, a transmission model is formulated to depict the combined effects of alternating data losses and fading. Second, the formulated model is integrated with the given control strategy to form a filtering model. Third, the filter based on projection theory is proposed. Finally, numerical results validate that feeding objects with the filtered input can significant improve the tracking accuracy.

This paper investigates consensus control of discrete-time heterogeneous multi-agent systems with multiplicative noise. As the frequent absence of velocity sensors in real industrial environments, a control protocol is developed using absolute velocity and relative position measurement information. Then, the consensus control issue is reformulated as the stability issue of stochastic difference equations. By choosing an appropriate Lyapunov function, the above stability issue is studied, and the sufficient conditions for leader-free and leader-following discrete-time heterogeneous multi-agent systems to achieve both mean-square and almost-sure consensus are derived, respectively. Finally, numerical simulations are provided to validate the effectiveness of the main results.

This paper studies the exponential input-to-state stability (ISS) of the island microgrid system with time-delay under aperiodic intermittent control. The event-triggered intermittent control (ETIC) is designed to realize the exponential ISS of the system. Under the ETIC, the island microgrid is proved to achieve the exponential ISS. Based on state-based control width (SCW), an ETIC-SCW scheme is also proposed by improving the ETIC scheme. The effectiveness of the proposed ETIC scheme is verified by simulation results. It is shown that the designed ETIC, especially the ETIC-SCW, can achieve not only the exponential ISS but also the lower minimum activation time rate than other control strategies including the time-triggered intermittent control. Therefore, the island microgrid system with time-delay may achieve a better balance between economy and stability.

For a class of higher order nonlinear systems with disturbances, this paper proposes a novel pre-defined time nonsingular sliding mode control (SMC), which ensures that the tracking error can converge to the origin in both the arrival stage and the sliding stage at a predefined time. The convergence time is independent of the controller parameters and is set by the users. In addition, this paper introduces a saturation function in the controller to solve the singular problem of traditional terminal SMC. Finally, simulations and experiments verify that the proposed method has good control performance such as fast convergence speed and strong robustness.

The full-order observer design problem for a particular class of port-controlled Hamiltonian systems is approached in this paper. The proposed full-order observer scheme belongs to the structure preserving class of dynamic estimators as it preserves the natural stability properties of the approached class of systems that are useful for the convergence analysis and exhibits a structure that is a copy of the original system plus an output corrective term. Due to the physical interpretation of port-controlled Hamiltonian systems, the proposed full-order observer is attractive because the estimated states have an immediate practical meaning. The class approached in this paper considers both linear and nonlinear dissipation terms, a quadratic Hamiltonian function and internal interconnections between two components of the state vector modulated by another state component that correspond to quadratic nonlinearities. This features lead to several structural properties that allow to carry out the convergence analysis in a relatively simple way and leads to a simple tuning procedure. The usefulness of the contribution is illustrated considering two practical applications: the Lorenz oscillator and the permanent magnet synchronous motor.

This study describes stability analysis that guarantees a part of the solutions will converge to a small ball centered in the origin for $��\varpi �$-fractional-order systems ($��\varpi �$-FOS). Such nonlinear systems are explored, and such practical stability is guaranteed by using the Lyapunov-like functions. This study delves into the theoretical underpinnings of practical stability. Additionally, it clarifies this concept through an application.

Although linear quadratic iterative learning control methods based on nonminimal state space (NMSS) models can directly observe the system state without designing a state observer, there is still room for improvement in their control performance. In order to further improve the control performance of core process indicators for batch processes, this paper presents a novel infinite horizon linear quadratic iterative learning control strategy in two-dimensional sense (2D-IHLQILC). Firstly, an extended nonminimal state space (2D-ENMSS) model in two-dimensional sense is developed to describe a batch process by incorporating process inputs, outputs, and tracking errors. Secondly, a 2D-IHLQILC scheme is designed based on this 2D-ENMSS process model. Finally, the effectiveness of the 2D-IHLQILC scheme is demonstrated by the injection speed control in injection modeling process. Compared with conventional methods based on NMSS model, the proposed 2D-IHLQILC method provides more extra degrees of freedom (DOF) to adjust the control performance and acquires improved control performance. In addition, it does not need a state observer to observe the system state.