Showing papers in "IEEE Transactions on Systems, Man, and Cybernetics in 2021"

Abstract: This paper examines the adaptive event-triggered fault detection problem of semi-Markovian jump systems (S-MJSs) with output quantization. First, we develop an adaptive event-triggered scheme for S-MJSs that is more effective than conventional event-triggered strategy for decreasing network transmission information. Meanwhile, we design a new adaptive law that can dynamically adjust the event-triggered threshold. Second, we consider output signal quantization and transmission delay in the proposed fault detection scheme. Moreover, we establish novel sufficient conditions for the stochastic stability in the proposed fault detection scheme with an $H_{\infty }$ performance with the help of linear matrix inequalities (LMIs). Finally, we provide simulation results to demonstrate the usefulness of the developed theoretical results.

Abstract: This paper addresses the adaptive finite-time decentralized control problem for time-varying output-constrained nonlinear large-scale systems preceded by input saturation. The intermediate control functions designed are approximated by neural networks. Time-varying barrier Lyapunov functions are used to ensure that the system output constraints are never breached. An adaptive finite-time decentralized control scheme is devised by combining the backstepping approach with Lyapunov function theory. Under the action of the proposed approach, the system stability and desired control performance can be obtained in finite time. The feasibility of this control strategy is demonstrated by using simulation results.

Abstract: This article investigates the neural network-based finite-time control issue for a class of nonstrict feedback nonlinear systems, which contain unknown smooth functions, input saturation, and error constraint. Radial basis function neural networks and an auxiliary control signal are adopted to identify unknown smooth functions and deal with input saturation, respectively. The issue of error constraint is solved by combining the performance function and error transformation. Based on the backstepping recursive technique, a neural network-based finite-time control scheme is developed. The developed control scheme can ensure that the closed-loop system is semi-globally practically finite-time stable. Finally, the validity of theoretical results is verified via simulation studies.

Abstract: An adaptive neural network (NN) event-triggered control scheme is proposed for nonlinear nonstrict-feedback multiagent systems (MASs) against input saturation, unknown disturbance, and sensor faults. Mean-value theorem and Nussbaum-type function are invoked to transform the structure of the input saturation and overcome the difficulty of unknown control directions, respectively. On the basis of the universal approximation property of NNs, a nonlinear disturbance observer is designed to estimate the unknown compounded disturbance composed of external disturbance and the residual term of input saturation. According to the measurement error defined by control signal, an event-triggered mechanism is developed to save network transmission resource and reduce the number of controller update. Then, an adaptive NN compensation control approach is proposed to tackle the problem of sensor faults via the dynamic surface control (DSC) technique. It is proved that all signals in the closed-loop system are semi-globally uniformly ultimately bounded. Finally, simulation results demonstrate the effectiveness of the presented control strategy.

Abstract: This article is concerned with the problem of the containment control with respect to continuous-time semi-Markovian multiagent systems with semi-Markovian switching topologies. Two kinds of classical control schemes, which are dynamic containment control and static containment control schemes, are utilized to address the proposed synthesis problem. Based on the linear matrix inequality (LMI) method, the dynamic containment controller and static containment controller are designed to plunge into the studied semi-Markovian multiagent systems, respectively. Moreover, the random switching topologies with the semi-Markovian process, the partly unknown transition rates, and the generally uncertain transition rates are taken into account, which can be applicable to more practical situations. Finally, the simulation results are provided to illustrate the effectiveness of the proposed theoretical results.

Abstract: This article investigates the adaptive fault-tolerant tracking control problem for a class of discrete-time multiagent systems via a reinforcement learning algorithm. The action neural networks (NNs) are used to approximate unknown and desired control input signals, and the critic NNs are employed to estimate the cost function in the design procedure. Furthermore, the direct adaptive optimal controllers are designed by combining the backstepping technique with the reinforcement learning algorithm. Comparing the existing reinforcement learning algorithm, the computational burden can be effectively reduced by using the method of less learning parameters. The adaptive auxiliary signals are established to compensate for the influence of the dead zones and actuator faults on the control performance. Based on the Lyapunov stability theory, it is proved that all signals of the closed-loop system are semiglobally uniformly ultimately bounded. Finally, some simulation results are presented to illustrate the effectiveness of the proposed approach.

Abstract: This paper focuses on the design of a reduced adaptive fuzzy tracking controller for a class of high-order stochastic nonstrict feedback nonlinear systems with full-state constraints. In the proposed approach, reduced fuzzy systems are used to approximate uncertain functions which involve all state variables and a high-order tan-type barrier Lyapunov function (BLF) is considered to deal with full-state constraints of the controlled system. With this BLF and a combination of the reduced fuzzy control and adding a power integrator, a novel control scheme is constructed to ensure that tracking error is within a very small range of the origin almost surely, meanwhile, the constraints on the system states are not breached almost surely during the operation. Two examples are proposed to show the effectiveness of the design scheme.

Abstract: Event-triggered communication mechanism (ETCM) provides an efficient way to reduce unwanted network traffic. This article studies the co-design of an ETCM and an annular finite-time (AFT) $H_{\infty }$ filter for networked switched systems (NSSs). First, the AFT definition and ETCM are presented. Second, a set of mode-dependent average dwell-time (MADT) switching rules is given. By resorting to a delay-dependent Lyapunov functional approach, some feasible AFT $H_{\infty }$ filters are designed. Third, it is proved that the filtering error system (FES) has a good performance in attenuating the external disturbances. Finally, the feasibility of the developed method is verified via simulation.

Abstract: As a generation of fuzzy sets, intuitionistic fuzzy sets (IFSs) have a more powerful ability to represent and address the uncertainty of information. Therefore, IFSs have been used in many areas. However, the distance measure between the IFSs indicating the difference or discrepancy grade is still an open question that has attracted considerable attention over the past few decades. Although various measurement methods have been developed, some problems still exist regarding the unsatisfactory axioms of distance measure or that lack discernment and cause counterintuitive cases. To address the above issues, in this article, we propose a new distance measure between IFSs based on the Jensen–Shannon divergence. This new IFS distance measure can not only satisfy the axiomatic definition of distance measure but also has nonlinear characteristics. As a result, it can better discriminate the discrepancies between IFSs, and it generates more reasonable results than do other existing measure methods; these advantages are illustrated by several numerical examples. Based on these qualities, an algorithm for pattern classification is designed that provides a promising solution for addressing inference problems.

Abstract: In this article, the $\mathcal {H}_{\infty }$ static output feedback tracking control problem is studied for discrete-time nonlinear networked systems subject to quantization effects and asynchronous event-triggered constraints. The Takagi–Sugeno (T–S) fuzzy model is utilized to represent the investigated nonlinear networked systems. A novel asynchronous event-triggered strategy is given to reduce the network communication burdens in both communication channels from the plant to the controller and from the reference model to the controller. The objective of this article is to design a quantized event-triggered tracking controller such that the resulting system is asymptotically stable and the given $\mathcal {H}_{\infty }$ tracking performance is guaranteed. The sufficient design conditions for the tracking controller are formulated in the form of the linear matrix inequalities (LMIs). Furthermore, a simulation example will be utilized to show the effectiveness of the developed design strategy.

Abstract: To deal with a class of nonlinear systems with unknown control directions, a command filter-based adaptive tracking controller is designed in this paper. In the design process, fuzzy logic system is required to handle nonlinear functions, command filter is employed to settle the explosion of complexity problem and Nussbaum function is introduced to compensate the influence of unknown directions problem. Finally, the proposed control approach guarantees that error signals converge into bounded compact sets around the origin and all closed-loop signals are bounded. The effectiveness of the presented scheme is illustrated by a simulation example.

Abstract: This article addresses the investigation of sliding-mode control (SMC) for slow-sampling singularly perturbed systems (SPSs) with Markov jump parameters. As a new attempt, the SMC strategy is considered in the study of discrete-time Markov jump SPSs. Subsequently, in order to design a sliding-mode controller to ensure the stability of the proposed system, a novel integral sliding surface is constructed, and an SMC law is synthesized to ensure the reachability of the sliding surface. Through the utilization of Lyapunov stability and SMC theory, sufficient conditions are derived to ensure the state trajectories of the system are driven to a predefined sliding surface and the closed-loop sliding mode dynamics are stochastically stable. Finally, the applicability of the proposed SMC strategy is verified by a numerical example and a practical electric circuit model.

Abstract: Salient object detection from RGB-D images is an important yet challenging vision task, which aims at detecting the most distinctive objects in a scene by combining color information and depth constraints. Unlike prior fusion manners, we propose an attention steered interweave fusion network (ASIF-Net) to detect salient objects, which progressively integrates cross-modal and cross-level complementarity from the RGB image and corresponding depth map via steering of an attention mechanism. Specifically, the complementary features from RGB-D images are jointly extracted and hierarchically fused in a dense and interweaved manner. Such a manner breaks down the barriers of inconsistency existing in the cross-modal data and also sufficiently captures the complementarity. Meanwhile, an attention mechanism is introduced to locate the potential salient regions in an attention-weighted fashion, which advances in highlighting the salient objects and suppressing the cluttered background regions. Instead of focusing only on pixelwise saliency, we also ensure that the detected salient objects have the objectness characteristics (e.g., complete structure and sharp boundary) by incorporating the adversarial learning that provides a global semantic constraint for RGB-D salient object detection. Quantitative and qualitative experiments demonstrate that the proposed method performs favorably against 17 state-of-the-art saliency detectors on four publicly available RGB-D salient object detection datasets. The code and results of our method are available at https://github.com/Li-Chongyi/ASIF-Net .

Abstract: The second-order sliding mode (SOSM) controller design problem for a class of sliding mode dynamics subject to an upper-triangular structure has been discussed in this paper. The proposed SOSM controller design involves two steps. First, a Lyapunov-based SOSM controller is developed by using the adding a power integrator technique to locally finite-time stabilize the sliding variables. Second, by combining the local SOSM controller with a saturation function, a novel SOSM controller with a saturation level is constructed. The feature of the new SOSM controller lies in that the saturation level can be tuned not only to guarantee the global convergence but also to improve the dynamic performance. Lyapunov analysis has been utilized to test the finite-time stability of the closed-loop sliding mode dynamics. The proposed method is eventually demonstrated by simulation results.

Abstract: An observer-based adaptive dynamic surface control (DSC) strategy is proposed for nonlinear nonstrict-feedback systems with time-varying disturbance, event-triggered mechanism, and actuator failures in this paper. Fuzzy logic systems are implemented to construct an observer to estimate the unmeasured states. The DSC method is exploited to solve the issue of “explosion of complexity” with the backstepping control technique. The constructed event-triggered mechanism can avoid the waste of communication resources. The barrier Lyapunov functions are utilized to address the problem of output constraint. It is demonstrated that the reference signals can be well tracked by the system output and all closed-loop signals remain semi-globally uniformly ultimately bounded. Two examples illustrate the effectiveness of the constructed controller.

Abstract: This paper examines the boundary disturbance observer-based control for a vibrating single-link flexible manipulator system possessing external disturbances. Two new boundary anti-disturbance control strategies are presented to eliminate vibration, track disturbance, and determine angle position for the flexible manipulator system. Achieving rigorous analysis with no model reduction, the derived control can ensure the angle positioning and bounded stability in the controlled system. By appropriately designing parameters, the resulting simulation results can demonstrate the control performance.

Abstract: JADE is a differential evolution (DE) algorithm and has been shown to be very competitive in comparison with other evolutionary optimization algorithms. However, it suffers from the premature convergence problem and is easily trapped into local optima. This article presents a novel JADE variant by incorporating chaotic local search (CLS) mechanisms into JADE to alleviate this problem. Taking advantages of the ergodicity and nonrepetitious nature of chaos, it can diversify the population and thus has a chance to explore a huge search space. Because of the inherent local exploitation ability, its embedded CLS can exploit a small region to refine solutions obtained by JADE. Hence, it can well balance the exploration and exploitation in a search process and further improve its performance. Four kinds of its CLS incorporation schemes are studied. Multiple chaotic maps are individually, randomly, parallelly, and memory-selectively incorporated into CLS. Experimental and statistical analyses are performed on a set of 53 benchmark functions and four real-world optimization problems. Results show that it has a superior performance in comparison with JADE and some other state-of-the-art optimization algorithms.

Abstract: Evolutionary multitasking (EMT) is a newly emerging research topic in the community of evolutionary computation, which aims to improve the convergence characteristic across multiple distinct optimization tasks simultaneously by triggering knowledge transfer among them. Unfortunately, most of the existing EMT algorithms are only capable of boosting the optimization performance for homogeneous problems which explicitly share the same (or similar) fitness landscapes. Seldom efforts have been devoted to generalize the EMT for solving heterogeneous problems. A few preliminary studies employ domain adaptation techniques to enhance the transferability between two distinct tasks. However, almost all of these methods encounter a severe issue which is the so-called degradation of intertask mapping. Keeping this in mind, a novel rank loss function for acquiring a superior intertask mapping is proposed in this article. In particular, with an evolutionary-path-based representation model for optimization instance, an analytical solution of affine transformation for bridging the gap between two distinct problems is mathematically derived from the proposed rank loss function. It is worth mentioning that the proposed mapping-based transferability enhancement technique can be seamlessly embedded into an EMT paradigm. Finally, the efficacy of our proposed method against several state-of-the-art EMTs is verified experimentally on a number of synthetic multitasking and many-tasking benchmark problems, as well as a practical case study.

Abstract: In lifelong machine learning (LML) systems, consecutive new tasks from changing circumstances are learned and added to the system. However, sufficiently labeled data are indispensable for extracting intertask relationships before transferring knowledge in classical supervised LML systems. Inadequate labels may deteriorate the performance due to the poor initial approximation. In order to extend the typical LML system, we propose a novel hierarchical lifelong learning algorithm (HLLA) consisting of two following layers: 1) the knowledge layer consisted of shared representations and integrated knowledge basis at the bottom and 2) parameterized hypothesis functions with features at the top. Unlabeled data is leveraged in HLLA for pretraining of the shared representations. We also have considered a selective inherited updating method to deal with intertask distribution shifting. Experiments show that our HLLA method outperforms many other recent LML algorithms, especially when dealing with higher dimensional, lower correlation, and fewer labeled data problems.

Abstract: This paper is coped with the robust sliding-mode-based control problem for a class of discrete nonlinear systems in the presence of mixed-delays and packet losses with uncertain missing probability. Both the time-varying state delays and the infinite distributed state delays are considered. Also, the data packet losses are modeled by a Bernoulli distributed stochastic variable with uncertain missing probability and an update rule is employed to characterize the signal transmitted to controller side. A sliding function is first constructed and the desired stochastic mean-square stability of sliding motion is ensured by providing a sufficient condition based on the matrix inequality technique. Besides, a new discrete SMC strategy is designed to guarantee that the state trajectories are driven onto the bounded band of predesigned sliding surface and maintain them therein during subsequent time. Finally, the effectiveness of the developed sliding-mode control technique is verified by some simulations with comparative results.

Abstract: This paper is concerned with the moving horizon (MH) estimation issue for a type of networked nonlinear systems (NNSs) with the so-called random access (RA) protocol scheduling effects. To handle the signal transmissions between sensor nodes and the MH estimator, a constrained communication channel is employed whose channel constraints implies that at each time instant, only one sensor node is permitted to access the communication channel and then send its measurement data. The RA protocol, whose scheduling behavior is characterized by a discrete-time Markov chain (DTMC), is utilized to orchestrate the access sequence of sensor nodes. By extending the robust MH estimation method, a novel nonlinear MH estimation scheme and the corresponding approximate MH estimation scheme are developed to cope with the state estimation task. Subsequently, some sufficient conditions are established to guarantee that the estimation error is exponentially ultimately bounded in mean square. Based on that the main results are further specialized to linear systems with the RA protocol scheduling. Finally, two numerical examples and the corresponding figures are provided to verify the effectiveness/correctness of the developed MH estimation scheme and approximate MH estimation scheme.

Abstract: Recently, deep convolutional neural networks (CNNs) have been successfully applied to the single-image super-resolution (SISR) task with great improvement in terms of both peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). However, most of the existing CNN-based SR models require high computing power, which considerably limits their real-world applications. In addition, most CNN-based methods rarely explore the intermediate features that are helpful for final image recovery. To address these issues, in this article, we propose a dense lightweight network, called MADNet, for stronger multiscale feature expression and feature correlation learning. Specifically, a residual multiscale module with an attention mechanism (RMAM) is developed to enhance the informative multiscale feature representation ability. Furthermore, we present a dual residual-path block (DRPB) that utilizes the hierarchical features from original low-resolution images. To take advantage of the multilevel features, dense connections are employed among blocks. The comparative results demonstrate the superior performance of our MADNet model while employing considerably fewer multiadds and parameters.

Abstract: With the widespread application of service-oriented architecture (SOA), a flood of similarly functioning services have been deployed online. How to recommend services to users to meet their individual needs becomes the key issue in service recommendation. In recent years, methods based on collaborative filtering (CF) have been widely proposed for service recommendation. However, traditional CF typically exploits only low-dimensional and linear interactions between users and services and is challenged by the problem of data sparsity in the real world. To address these issues, inspired by deep learning, this article proposes a new deep CF model for service recommendation, named location-aware deep CF (LDCF). This model offers the following innovations: 1) the location features are mapped into high-dimensional dense embedding vectors; 2) the multilayer-perceptron (MLP) captures the high-dimensional and nonlinear characteristics; and 3) the similarity adaptive corrector (AC) is first embedded in the output layer to correct the predictive quality of service. Equipped with these, LDCF can not only learn the high-dimensional and nonlinear interactions between users and services but also significantly alleviate the data sparsity problem. Through substantial experiments conducted on a real-world Web service dataset, results indicate that LDCF’s recommendation performance obviously outperforms nine state-of-the-art service recommendation methods.

Abstract: In brain storm optimization (BSO), the convergent operation utilizes a clustering strategy to group the population into multiple clusters, and the divergent operation uses this cluster information to generate new individuals. However, this mechanism is inefficient to regulate the exploration and exploitation search. This article first analyzes the main factors that influence the performance of BSO and then proposes an orthogonal learning framework to improve its learning mechanism. In this framework, two orthogonal design (OD) engines (i.e., exploration OD engine and exploitation OD engine) are introduced to discover and utilize useful search experiences for performance improvements. In addition, a pool of auxiliary transmission vectors with different features is maintained and their biases are also balanced by the OD decision mechanism. Finally, the proposed algorithm is verified on a set of benchmarks and is adopted to resolve the quantitative association rule mining problem considering the support, confidence, comprehensibility, and netconf. The experimental results show that the proposed approach is very powerful in optimizing complex functions. It not only outperforms previous versions of the BSO algorithm but also outperforms several famous OD-based algorithms.

Abstract: This paper investigates the leader–follower robust ${H_\infty }$ consensus of heterogeneous multiagent systems with denial of service attack, where different attack intensities are considered. A switched system model is introduced to model such an attack phenomenon. Then, sufficient conditions to guarantee the solvability of the robust output consensus problem are obtained, and the quantitative relationship between the consensus performance and attack parameter is established. It is shown that the consensus protocol design problem can be transformed into two static output feedback (SOF) control problems. It is also shown that the SOF controller gains can be determined by solving some linear matrix inequalities without the knowledge of the probability information of each attack. Finally, the effectiveness of the control protocol is demonstrated by a simulation study.

Abstract: This article discusses the control design and experiment validation of a flexible two-link manipulator (FTLM) system represented by ordinary differential equations (ODEs). A reinforcement learning (RL) control strategy is developed that is based on actor–critic structure to enable vibration suppression while retaining trajectory tracking. Subsequently, the closed-loop system with the proposed RL control algorithm is proved to be semi-global uniform ultimate bounded (SGUUB) by Lyapunov’s direct method. In the simulations, the control approach presented has been tested on the discretized ODE dynamic model and the analytical claims have been justified under the existence of uncertainty. Eventually, a series of experiments in a Quanser laboratory platform are investigated to demonstrate the effectiveness of the presented control and its application effect is compared with PD control.

Abstract: We propose a novel epidemic model based on two-layered multiplex networks to explore the influence of positive and negative preventive information on epidemic propagation. In the model, one layer represents a social network with positive and negative preventive information spreading competitively, while the other one denotes the physical contact network with epidemic propagation. The individuals who are aware of positive prevention will take more effective measures to avoid being infected than those who are aware of negative prevention. Taking the microscopic Markov chain (MMC) approach, we analytically derive the expression of the epidemic threshold for the proposed epidemic model, which indicates that the diffusion of positive and negative prevention information, as well as the topology of the physical contact network have a significant impact on the epidemic threshold. By comparing the results obtained with MMC and those with the Monte Carlo (MC) simulations, it is found that they are in good agreement, but MMC can well describe the dynamics of the proposed model. Meanwhile, through extensive simulations, we demonstrate the impact of positive and negative preventive information on the epidemic threshold, as well as the prevalence of infectious diseases. We also find that the epidemic prevalence and the epidemic outbreaks can be suppressed by the diffusion of positive preventive information and be promoted by the diffusion of negative preventive information.

Abstract: This paper studies the $ {H}_{ {\infty }}$ sliding mode control (SMC) problem for a class of discrete-time conic-type nonlinear systems with time-delays and uncertainties. The nonlinear terms satisfy the conic-type constraint condition that lies in a know hyper-sphere with an uncertain center. By choosing a proper Lyapunov candidate, sufficient conditions are derived to ensure the asymptotic stability of the sliding mode dynamics while achieving a prescribed $ {H}_{ {\infty }}$ disturbance attenuation level and finally converted into a minimization problem. The controller is constructed to guarantee the discrete-time reach condition and maintain the states on the prespecified sliding surface. A simulation result and a practical example related to the Chua’s circuit are given at last to show the validity of our SMC strategy.

Abstract: This article investigates the finite-time asynchronous control problem for continuous-time positive hidden Markov jump systems (HMJSs) by using the Takagi–Sugeno fuzzy model method. Different from the existing methods, the Markov jump systems under consideration are considered with the hidden Markov model in the continuous-time case, that is, the Markov model consists of the hidden state and the observed state. We aim to derive a suitable controller that depends on the observation mode which makes the closed-loop fuzzy HMJSs be stochastically finite-time bounded and positive, and fulfill the given $L_{2}$ performance index. Applying the stochastic Lyapunov–Krasovskii functional (SLKF) methods, we establish sufficient conditions to obtain the finite-time state-feedback controller. Finally, a Lotka–Volterra population model is used to show the feasibility and validity of the main results.

Abstract: High-dimensional and sparse (HiDS) matrices generated by recommender systems contain rich knowledge regarding various desired patterns like users’ potential preferences and community tendency. Latent factor (LF) analysis proves to be highly efficient in extracting such knowledge from an HiDS matrix efficiently. Stochastic gradient descent (SGD) is a highly efficient algorithm for building an LF model. However, current LF models mostly adopt a standard SGD algorithm. Can SGD be extended from various aspects in order to improve the resultant models’ convergence rate and prediction accuracy for missing data? Are such SGD extensions compatible with an LF model? To answer them, this paper carefully investigates eight extended SGD algorithms to propose eight novel LF models. Experimental results on two HiDS matrices generated by real recommender systems show that compared with an LF model with a standard SGD algorithm, an LF model with extended ones can achieve: 1) higher prediction accuracy for missing data; 2) faster convergence rate; and 3) model diversity.