ISA transactions最新文献

This paper presents a novel non-iterative data-driven distributed economic model predictive control (EMPC) scheme for distributed linear time-invariant systems. The system consists of dynamically interacting subsystems that optimize an economic objective. Each subsystem solves its local optimization problem using input-output data. It also considers assumed neighbor trajectories subject to a consistency constraint derived from Hankel matrices. A data-driven framework is presented to achieve strong duality and dissipativity with a general supply rate. Additionally, a procedure is developed to design terminal ingredients for distributed EMPC using input-output trajectories. Theoretical analysis guarantees recursive feasibility and closed-loop stability. Simulation results on a building temperature control system demonstrate the energy efficiency and effectiveness of the proposed method compared with other MPC approaches.

This study integrates the adaptive neural network technique, state and disturbance observer, and nonsingular terminal sliding mode control framework to achieve rapid trajectory tracking for a quadrotor under state uncertainty, dynamic disturbances, and unknown dynamics. The control methodology begins by modeling the quadrotor's dynamics, including inherent uncertainties and disturbances. A state observer is designed to estimate unknown states while rejecting uncertainties and disturbances, with exponential convergence of observer errors demonstrated via Lyapunov theory. The tracking error between the observed quadrotor's states and the desired trajectory is subsequently defined, and a nonsingular terminal sliding mode surface is employed to stabilize this error. Using Lyapunov-based analysis, it is proven that the sliding surfaces converge to zero at an exponential rate. Furthermore, an adaptive neural network strategy is developed to approximate unknown components, including state variables, uncertainties, and disturbances. Simulation and experimental results using a quadrotor with realistic parameters validate the proposed control strategy, confirming its robustness and effectiveness.

Multi-agent system (MAS), as the main framework for distributed resource allocation, may be subject to resource limitations, communication bandwidth constraints, and environmental changes in actual operation, which can lead to nonlinear behavior, uncertain parameters, and unbalanced communication in the MAS. To address these challenges, this paper proposes a novel distributed algorithm that includes a controller and an optimal estimator. By incorporating a dynamic compensation mechanism and tracking technique, the proposed algorithm effectively handles system nonlinearities, tracks uncertain parameters, and compensates for asymmetric communication, thereby enhancing the system's robustness and expanding its application scope. Furthermore, to further improve the communication efficiency of MAS, a dynamic event-triggered mechanism is integrated with the dynamic compensation mechanism, ensuring that MAS can effectively conserve communication resources under unbalanced communication topologies. Finally, the proposed algorithm is validated through a case study involving a five-drone cooperative formation in a nonlinear uncertain MAS.

This paper investigates the adaptive fuzzy neural networks cooperative formation control problem of multiple unmanned surface vehicles with actuator faults and unknown velocity information. Firstly, a new fuzzy neural networks-based prescribed time extended state observer is designed, which can not only estimate unknown velocity from the position information of unmanned surface vehicles, but also estimate ocean disturbances and uncertainties in the system model. Considering the actual requirements of formation operation, the formation error and velocity are constrained within the prescribed safety boundary. Secondly, an adaptive cooperative formation control method is designed by defining distributed cooperative formation error and introducing log-type barrier Lyapunov function to prevent formation error and velocity from violating prescribed constraints. In addition, the designed fuzzy neural networks adaptive law and multiplicative fault adaptive law can compensate for model uncertainties and time-varying multiplicative actuator faults. This method can ensure that multiple unmanned surface vehicles form and maintain the desired formation within a prescribed time, and the formation error converges to a smaller neighborhood of the origin within the prescribed time. Finally, the effectiveness of the proposed control method is demonstrated through multiple simulation analyses.

This work discusses the secure consensus issue of Lipschitz nonlinear multiagent systems (MASs) subject to denial-of-service (DoS) attacks. Due to the presence of DoS attacks, a new update rule for adaptive parameters has been designed. Meanwhile, a novel event-triggered (ET) strategy with an auxiliary dynamic variable is presented to cut down unnecessary communication, save energy and prolong the network lifetime. And the corresponding control strategy is developed and does not need to use any global information. Moreover, a well designed Lyapunov function is introduced for stability analysis. The sufficient criteria for achieving consensus convergence have been derived, and the Zeno phenomenon has also been avoided, where the relation between attack constraints, system matrix and triggering parameters has been clarified. Finally, the feasibility and validity of the designed ET control strategy are validated through a simulation experiment.

The control of piezoelectric nano motion stages is one of the key enabling techniques supporting high precision optical scanning tasks in nano-measurement applications. This paper proposes a hysteretic state and parameter estimation scheme for switching iterative learning control of a piezoelectric driven stage, in which the hysteresis nonlinearities of such a motion system are reflected by an asymmetric Bouc-Wen model (ABWM). Aided by only the displacement measurement, the prediction error of the hysteresis model is utilized to devise an update law for obtaining the hysteretic state observation error. The parameters of the entire ABWM are adaptively estimated by the use of the hysteretic state observation error. Based on this, a novel adaptive state observer is presented for hysteretic estimation and inverse compensation. Moreover, both the external perturbation and the unmodeled dynamics of the asymmetric hysteretic system are also considered, and the robust compensation term with an adaptive gain is introduced to address their adverse effects. The convergence of the parameter estimation errors and the state observation errors is rigorously analyzed by the Lyapunov stability theory. To boost the feedforward inverse control performance, a switching iterative learning control framework is developed to seek the high-accuracy trajectory tracking. Comprehensive experiments regarding the adaptive model prediction and the trajectory tracking are conducted on a piezoelectric nano motion stage. Both the excellent model prediction performance and the satisfactory displacement control accuracy are demonstrated by the experimental results.

Hydraulic manipulators are increasingly used in applications such as ocean exploration and forestry harvesting. Given the potential risks associated with the high output torque, remote operation offers an appropriate method for operating hydraulic manipulators. However, the master-slave heterogeneity in remote operations restricts the operational efficiency and precision, and the occlusions of the operator's field of view (FOV) can affect the accuracy of the master commands. While visual guidance can assist manipulator operation, the varying imaging conditions and potential target occlusions also limit the reliability of guidance commands. To address these challenges, this paper designs a visual guidance based shared remote operation control for hydraulic manipulators. On the one hand, the speed and accuracy of the mapped commands are balanced by the designed adaptive command mapping method. On the other hand, the reliability of the visual guidance command is enhanced through feature mismatch rectification and fault awareness. Subsequently, a dual-command fusion control method is developed to integrate the optimized commands, thereby achieving shared remote operation. Finally, various experiments are conducted to verify the proposed algorithms. The results demonstrate that the designed scheme guarantees the shared remote operation performance of hydraulic manipulators even when the operator's FOV is occluded.

This study investigates the stability analysis of linear systems with cyclical delays, where the delay interval is partitioned into subintervals that exhibit alternating cycles of monotonic increase and decrease. The Lyapunov-Krasovskii functionals (LKFs) are constructed by integrating two-loop functional frameworks based on delay product terms, which strategically leverage distinct delay monotonicity intervals. A novel delay-derivative-dependent inequality is proposed, which incorporates delay-derivative information into the stability analysis framework. By integrating the proposed delay-product-dependent LKFs with a new negative-definiteness condition (NDC) for generalized bivariate matrix polynomials, an improved stability criterion is described in the form of a linear matrix inequality (LMI). Three numerical examples demonstrate that the derived stability condition is less conservative than previous work.

Hydraulic systems are vital in industrial settings, and reliable Condition Monitoring (CM) is crucial to preventing failures. This paper introduces a Physics-informed Deep Hidden Markov Model (PiDHMM) combined with a Wasserstein Generative Adversarial Network (WGAN) for enhanced fault detection. PiDHMM improves traditional Hidden Markov Models (HMMs) by embedding physical constraints into state transitions and leveraging a Convolutional Neural Network (CNN) to model emission probabilities and capture complex sensor behavior. To address data scarcity in rare failure modes, WGAN is employed to generate realistic synthetic sensor data. The proposed framework is validated on a multi-sensor hydraulic dataset with known failure events. Comparative results show that PiDHMM outperforms both standard HMMs and deep HMMs without physics constraints, achieving a significant increase in accuracy for fault classification. The inclusion of physics-informed transitions enhances temporal consistency and interpretability, while the WGAN-based augmentation addresses issues of data imbalance, further improving model performance. These results demonstrate that the PiDHMM-WGAN approach offers a more precise, interpretable, and robust solution for hydraulic system monitoring.

Network congestion and communication resource competition are common and destructive challenges in single-master-multiple-slaves (SMMS) teleoperation systems. In this paper, an adaptive event-triggered scheduling control for SMMS teleoperation systems is proposed. The event-triggered scheme is based on the position and velocity signals of robots with an adaptive threshold to dynamically adjust the triggering frequency. A scheduling protocol is constructed to allocate the signal transmission priority for slave robots based on three triggering states. Subsequently, a cooperative controller is developed utilizing the master triggered signals and the slave triggered and scheduled signals, guaranteeing the stability and position tracking performance of the system. The system stability and the convergence of the position tracking error are proved with Lyapunov functions and validated through simulations. The proposed method guarantees the stability and position tracking performance of SMMS teleoperation systems while concurrently alleviating network congestion and communication resource competition.