With the popularity of social media, users are able to express their opinions in multiple forms, such as text, audio, and video. Traditional unimodal sentiment analysis methods can no longer meet the processing requirements of such multisource heterogeneous data, which makes multimodal sentiment analysis a research hotspot. However, most existing methods rely on simple feature splicing or weighted fusion, neglecting the differences in the reliability of different modalities and failing to fully explore the intermodality consistency and difference information. In this article, we propose a multimodal competitive fusion mechanism and construct multimodal competitive fusion model (MCFM). The model first dynamically evaluates the reliability of each modality through the competition mechanism and adaptively assigns weights accordingly. Then it decomposed the modality representations into similar and dissimilar features through modality feature decomposition, supplemented by the overlap of orthogonal traffic channel attention constraints, to achieve the collaborative learning of consistency and dissimilarity features. We evaluate the proposed model on several datasets. In our experiments, we used textual modality data from the dataset with audio modality data for the experiments. The experimental results show that MCFM has 2%–3% higher binary accuracy (ACC2) than the baseline model on the sentiment classification task (with 2% higher binary accuracy under the negative/nonnegative metrics and 3% higher binary accuracy under the positive/negative metrics), and that on the regression task, MCFM’s mean absolute error on the test dataset is 3% lower than that of the baseline model.

This paper introduces a swift solution for the complementary coverage path planning issue for multiple uncrewed aerial vehicles (UAV) in 3-dimensional, non-convex, and trap-laden indoor environments. A topology graph is constructed by sampling the blueprint. To divide the indoor environment into several convex areas and form tasks for the UAVs, the Louvain (fast unfolding) algorithm is utilized twice for space partitioning and task assigning. Finally, to ascertain the path for each UAV, a tiny-scale traveling salesman problem (TSP) is formed and solved. The simulation result reveals that the proposed strategy has significantly improved efficiency in both 2-dimensional and 3-dimensional path planning, underscoring the algorithm’s practical relevance for high-rise fire rescue operations. Note to Practitioners—This paper is motivated by the requirement of urgent task assignment and route finding for multiple uncrewed aerial vehicles (UAV) in indoor searching, especially for high-rise fire rescue. A rapid task allocation and pathfinding framework that balances efficiency and optimization levels is proposed. Unlike other complex environments, community-detection-based task allocation methods are recommended for actual indoor space rather than clustering-based ones. Additionally, the degree of each node in the sampling graph can be used to rapidly determine the sensor’s detection range at that node and the distance to obstacles. Making reasonable use of this can further enhance the efficiency of task allocation and pathfinding. Extensive experimental results in actual indoor maps show the effectiveness of the proposed task allocation and pathfinding framework. Note that our method is designed only for indoor environments divided into multiple rooms by continuous walls, and the effectiveness in other types of environments cannot be guaranteed.

The situation of interest is where a vehicle is equipped with multiple sensors to measure the distance to the leading vehicle but does not need to obtain data from speed and acceleration sensors. The distance measurements are susceptible to asynchronous sampling and noise, nearly half of which may be manipulated by malicious attackers. In this situation, the event-triggered vehicle-following control problem of nonlinear autonomous vehicles with unknown parameters is studied. First, a secure event-triggered mechanism that can resist manipulation is devised to alleviate the burden of data transmission and processing caused by multiple sensors. Then, a novel adaptive sensor fusion algorithm is developed to estimate the actual distance. Subsequently, an improved adaptive observer is designed based on the event-triggered estimated distance to estimate continuous-time distance, velocity, acceleration, and system parameters. Finally, the following controller is designed using the estimated states and parameters with the help of Levant differentiators. The effectiveness of the proposed control scheme is validated through simulation studies.Note to Practitioners—This work aims to develop a secure following control method for nonlinear automated vehicles with unknown states and parameters, which can effectively handle sparse sensor problems caused by attacks, faults, saturation, etc. To address practical issues such as sampling intervals and limited computing and transmission resources, we propose a discrete sampling–event-triggered transmission–continuous estimation and control framework for continuous-time systems. Despite the presence of measurement interferences and potential corruption, the designed event-triggered mechanism and sensor fusion algorithm can be used to reduce data transmission and estimate the actual output, respectively. The designed adaptive observer can estimate continuous-time system states and parameters whether the output is obtained in a continuous, short-interval discrete or suitable event-triggered manner. This capability facilitates control design and real-time monitoring of vehicle states. Additionally, the presented backstepping control design and analysis method utilizing Levant differentiators can be applied in situations where the controlled system’s states possess at least first-order differentiability.

This paper proposes an adaptive iterative learning control (AILC) scheme for multiagent systems (MASs) to improve the containment control performance. To deal with the uncertain nonlinearity, a neural network (NN)-based iterative predictor with the same structure of MASs is constructed, where an auxiliary approximation term is further incorporated to handle the NN reconstruction error. By utilizing the output of iterative predictor, an AILC-based containment control scheme is designed under the backstepping framework. To overcome the initial error problem, distributed compensation signals are developed to compensate for containment errors and dynamic surface errors. Instead of using dynamic information of neighbors, only the output of neighbors is used in each follower such that the communication load is reduced. It is established that as iteration index approaches infinity, the iteratively convergence of containment errors is achieved, and the stability of MASs is ensured. Simulation results on manipulators verify the effectiveness of both the iterative predictor and the adaptive iterative learning containment controller.

This paper investigates an adaptive iterative learning control (AILC) method for multiagent systems (MASs) performing multiple tasks. Different from traditional results for the single task, a multiple tasks case is considered in this work, which can complete various cooperative control. It should be pointed out that only one of the multiple tasks is performed in each iteration. For multiple tasks, a neural network (NN) is employed to create a mapping relationship between the input and output of nonlinearity, which is integrated into AILC to improve the control input. Additionally, an auxiliary signal is developed to compensate for the residual error caused by NN approximation and differentiator estimation. Then, a termination condition including mean square error and desired performance is established for the AILC method. By utilizing Lyapunov stability theory, it is proven that the tracking error converges to zero without termination condition and satisfies the desired accuracy with termination condition. In the simulation, multiple single-link manipulators are used to perform three different cooperative control tasks to validate the effectiveness of the proposed approach. Note to Practitioners—Most of the existing AILC methods are investigated to perform a single task for MASs. However, practical applications of MASs, such as manipulators, unmanned vehicles, and aerospace, often require the ability to perform multiple tasks. One motivation is to effectively learn control experience from different tasks. To achieve this, an NN is designed to learn experiences, which are further incorporated into AILC method to perform multiple tasks. In addition, due to limitations in computing resources, controllers cannot be learned indefinitely in practical applications. To address this challenge, a termination condition is established to stop learning when the desired performance is met. Based on the above considerations, the proposed AILC method with termination condition can reduce computational resources and flexibly perform multiple tasks in practical applications.

The issue related to distributed containment control for multiple networked mechanical systems with inherent nonlinearities, dynamic leaders, unknown external disturbances, parameter uncertainties, and constrained network communication is investigated by designing distributed adaptive event-triggered sliding-mode controllers in this study. To lessen the number of state updates and network resource loss of networked mechanical systems, a time-varying-threshold-based adaptive event-triggered mechanism is constructed. An adaptive sliding-mode estimator is established to estimate the inaccurate states. Then, integrated with the aforementioned event-triggered strategy and adaptive sliding-mode estimator, discontinuous and continuous distributed adaptive event-triggered sliding-mode control laws without requiring each follower to get the upper bounds of the leaders’ states derivatives are, respectively, devised to compensate for the influences of nonlinearities, disturbances, and parameter uncertainties. To further attenuate the negative effects of unknown disturbances, inherent nonlinearities, and chattering, a distributed adaptive event-triggered sliding-mode control protocol with boundary layer function is designed. Eventually, the Lyapunov stability theory is utilized to testify that the adaptive error and containment error are uniformly ultimately bounded. A practical example is furnished to verify the validity of the present sliding-mode containment control strategies.Note to Practitioners—This work aims to develop the distributed containment control approach for multiple networked nonlinear mechanical systems, which is of great significance in the fields of deep space exploration, environment monitoring and joint rescue. We put forward an event-based adaptive estimation and containment control framework for multiple networked nonlinear mechanical systems, which solves practical problems such as transmission frequency, limited computation capability, and network resources. An adaptive sliding-mode estimator and adaptive event-triggered mechanism are, respectively, devised to estimate the inaccurate states and reduce data transmission frequency. Despite the effects of communication interruption and unknown disturbances, an event-triggered adaptive control protocol based on sliding-mode estimator is designed, which is applied to a planar manipulator with two degree-of-freedom.

This article focuses on the issue of the quantization-based event-triggered integral sliding-mode controller design for networked multiagent systems (MASs) encountering interferences under limited network bandwidth. An integral sliding manifold (ISM) is designed to address the effect of disturbances and ensure the desired dynamic performance of the system. We establish an event-triggered mechanism (ETM) with an exponential decay rate to conserve the limited communication resources. Then, a uniform quantizer is added to quantify the triggered state signals to lessen the network transmission burden caused by the digital network. Combining the designed ETM with a static uniform quantizer, the quantized trigger state signals are sent to decoders through the digital network to construct a quantized ISM. Subsequently, an event-triggered integral sliding-mode controller under quantization technology is developed to ensure the asymptotic average consensus of networked MASs. By testifying that every network agent has a lower positive bound, the viability of the proposed ETM is demonstrated, thereby ensuring the absence of Zeno behavior. Eventually, two simulation examples are proffered to confirm the efficacy of the quantization feedback-based event-triggered sliding-mode control methodology.

In this article, we concentrate on the adaptive fuzzy containment control approach for a class of nonlinear multiagent systems with deferred constraint and actuator failure. First, considering that not all agents can directly receive the leader signals, this article constructs a distributed prescribed-time estimator to provide each agent with a corresponding reference signal, thereby the containment problem is constructed as a tracking problem. Subsequently, with the help of the prescribed-time scaling function and the barrier function, the problem of deferred output constraint is reformulated as a boundedness problem of the new variable. By introducing several useful lemmas, the designed controller can ensure that the closed-loop signals are bounded in the presence of actuator fault in the system. In addition, through the designed fuzzy control algorithm, it can strictly guarantee that the system output converges within the ideal range after the settling time. The superiority and effectiveness of this method are verified through robots experiments.

This article addresses the problem of distributed fuzzy adaptive pinning control for multiagent systems with output saturation via adopting the dynamic event-triggered mechanism. With the framework of the backstepping technique, a new adaptive pinning control protocol is developed, where an effective fuzzy strategy and a class of tuning functions are integrated into the control protocol to reduce the required design adaptive parameters. Then, the output saturation of nonlinear multiagent systems is first addressed via the signal compensation method. Moreover, a dynamic event-triggered mechanism about control information is proposed, where two dynamic laws are designed to further increase the adjustment margin of the controller. Finally, simulation results are utilized to demonstrate the suitability and feasibility of theoretical algorithm.

In this article, we focus on the distributed group asynchronous consensus problem for cyber-physical systems (CPSs) with unknown dynamics and switching topologies. This article considers a class of networked distributed CPSs composed of cloud computing systems, and it is subjected to delay detection models and topology switching. First, an asynchronous switching observer is tailored for each group of agents to guarantee the precise acquisition of the leader's information. Further, we introduce a model-free adaptive control method to devise controllers for each group of agents, which can continue to learn adaptively only from the agent's input and output data without knowing the agent dynamics. Finally, the stability of both the observer and controller are proved, respectively. The observers' and controllers' effectiveness is further confirmed by the simulation results.

This article focuses on the optimal consensus control problem for multiagent systems (MASs) with discontinuous constraints. The case of discontinuous constraints is a particular instance of state constraints, which has been studied less but occurs in many practical situations. Due to the discontinuous constraint boundaries, the traditional barrier function-based backstepping methods cannot be used directly. In response to this thorny problem, a novel constraint boundary reconstruction technique is proposed by designing a class of switch-like functions. The technique can convert discontinuous constraint boundaries into continuous ones, and it strictly proves that when the states satisfy the transformed constraint boundaries, the original constraints are also absolutely fulfilled. Meanwhile, with the aid of the barrier function and distributed event-triggered estimator, an improved coordinate transformation is constructed, which can remove the "feasibility condition" and simplify the controller design. In addition, by introducing prediction error and revised term into the learning process of neural networks (NNs), the optimal consensus problem is resolved by constructing a modified reinforcement learning strategy. Finally, the stability of the MASs is testified through the Lyapunov stability theory, and a simulation example verifies the effectiveness of the proposed method.

A pinning-based self-regulation intermediate event-triggered (ET) funnel tracking control strategy is proposed for uncertain nonlinear multiagent systems (MASs). Based on the backstepping framework, a pinning control strategy is designed to achieve the tracking control objective, which only uses the communication weight between the agents without additional feedback parameters. Moreover, by designing a self-regulation triggered condition based on the tracking error, the intermediate triggered signal is calculated to replace the continuous signal in the controller, so as to achieve the goal of discontinuous update of the controller signal, and this mechanism does not need to add additional compensation function to the controller signal. At the same time, the funnel method is adopted to restrict the error of step $n$ and avoid the possible negative impact caused by control signal. Furthermore, the nonlinear noncontinuous faults are compensated by the disturbance observer. Then, the Lyapunov stability theorem is used to prove that all signals of the closed-loop system are semiglobally uniformly ultimately bounded (SGUUB). Finally, some simulation results confirm the effectiveness of the proposed control scheme.

In this article, the fuzzy dynamic event-triggered (DET) containment control problem for human-in-the-loop (HiTL) multiagent systems (MASs) with error constraints is investigated. Through utilization of fuzzy logic systems (FLSs), a high-gain state observer is presented to estimate unavailable states. To improve the transient performance, a fixed-time prescribed performance (FTPP) function is presented to restrict the containment errors and virtual errors. Under the backstepping control framework, a barrier Lyapunov function is designed to guarantee that the transform errors do not transgress the constraint bounds. Meanwhile, a nonlinear filter is introduced to obtain reduction calculation and enhance the control performance. Furthermore, the DET mechanism is presented to decrease the network communication burden. By directly controlling multiple leaders to form a dynamic convex hull, the proposed control strategy ensures that output signals of followers can converge to this convex hull and the containment errors can converge to the prescribed bounds within fixed time. Herein, a simulation example is presented to assess the effectiveness of the proposed control scheme.

This article concentrates on the adaptive neural control approach of n -link flexible-joint electrically driven robots. The presented control method only needs to know the position and armature current information of the flexible-joint manipulator. An adaptive observer is designed to estimate the velocities of links and motors, and radial basis function neural networks are applied to approximate the unknown nonlinearities. Based on the backstepping technique and the Lyapunov stability theory, the observer-based neural control issue is addressed by relying on uplink-event-triggered states only. It is demonstrated that all signals are semi-globally ultimately uniformly bounded and the tracking errors can converge to a small neighborhood of zero. Finally, simulation results are shown to validate the designed event-triggered control strategy.

This article is centered on the cybersecurity research of dynamic state estimation for power systems with measurement delays. Relying on mixed measurements from phasor measurement units (PMUs) and remote terminal units (RTUs), a delayed measurement model is constructed. A modified state estimator based on the Kalman filter (KF) is designed, which can obtain the optimal estimated states under measurement delays. Moreover, the measurement data transmitted from the sensor to the estimator are vulnerable to cyberattacks. Especially, false data-injection (FDI) attacks are frequently encountered in the power system state estimation (PSSE) process. In the case of measurement delays, an FDI attack strategy is designed to interfere with the state estimator and evade detection by the chi-square detector. By utilizing the attacked estimated information and the uncorrupted measurement information, two measurement residual vectors are designed. According to these two residual vectors, a chi-square-based attack detection method is proposed, which has the ability to detect the attack without being affected by the delayed measurements. The proposed KF algorithm and attack detection method are implemented on an IEEE 14-bus system and they are confirmed to be effective and feasible.

This article proposes a finite-time adaptive containment control scheme for a class of uncertain nonlinear multiagent systems subject to mismatched disturbances and actuator failures. The dynamic surface control technique and adding a power integrator technique are modified to develop the distributed finite-time adaptive containment algorithm, which shows lower computational complexity. In order to overcome the difficulty from the mismatched uncertainties, the disturbance observers are constructed based on the backstepping technique. Moreover, the uncertain actuator faults, including loss of effectiveness model and lock-in-place model, are considered and compensated by the proposed adaptive control scheme in this article. According to the Lyapunov stability theory, it is demonstrated that the containment errors are practically finite-time stable in the presence of actuator faults. Finally, a simulation example is conducted to show the effectiveness of the proposed theoretical results.

This article focuses on the sampled-data synchronization issue for a class of complex dynamical networks (CDNs) subject to noisy sampling intervals and successive packet losses. The sampling intervals are subject to noisy perturbations, and categorical distribution is used to characterize the sampling errors of noisy sampling intervals. By means of the input delay approach, the CDN under consideration is first converted into a delay system with delayed input subject to dual randomness and probability distribution characteristic. To verify the probability distribution characteristic of the delayed input, a novel characterization method is proposed, which is not the same as that of some existing literature. Based on this, a unified framework is then established. By recurring to the techniques of stochastic analysis, a probability-distribution-dependent controller is designed to guarantee the mean-square exponential synchronization of the error dynamical network. Subsequently, a special model is considered where only the lower and upper bounds of delayed input are utilized. Finally, to verify the analysis results and testify the effectiveness and superiority of the designed synchronization algorithm, a numerical example and an example using Chua's circuit are given.

This article considers the Nussbaum gain adaptive control issue for a type of nonlinear systems, in which some sophisticated and challenging problems, such as periodic disturbances, dead zone output, and unknown control direction are addressed. The Fourier series expansion and radial basis function neural network are incorporated into a function approximator to model time-varying-disturbed function with a known period in nonlinear systems. To deal with the problems of the dead zone output and unknown control direction, the Nussbaum-type function is recommended in the design of the control algorithm. Applying the Lyapunov stability theory and backstepping technique, the proposed control strategy ensures that the tracking error is pulled back to a small neighborhood of origin and all closed-loop signals are bounded. Finally, simulation results are presented to show the availability and validity of the analysis approach.

This article focuses on the event-triggered-based adaptive neural-network (NN) control problem for nonlinear large-scale systems (LSSs) in the presence of full-state constraints and unknown hysteresis. The characteristic of radial basis function NNs is utilized to construct a state observer and address the algebraic loop problem. To reduce the communication burden and the signal transmission frequency, the event-triggered mechanism and the encoding-decoding strategy are proposed with the help of a backstepping control technique. To encode and decode the event-triggering control signal, a one-bit signal transmission strategy is adopted to consume less communication bandwidth. Then, by estimating the unknown constants in the differential equation of unknown hysteresis, the effect caused by unknown backlash-like hysteresis is compensated for nonlinear LSSs. Moreover, the violation of full-state constraints is prevented based on the barrier Lyapunov functions and all signals of the closed-loop system are proven to be semiglobally ultimately uniformly bounded. Finally, two simulation examples are given to illustrate the effectiveness of the developed strategy.

The problem of leader-following consensus fault-tolerant control is investigated for multiagent systems (MASs) with time-varying nonaffine nonlinear faults, where the interactive topology is directional. In this article, to guarantee the performance consensus on tracking error, in the design process, the inherent problem of “explosion of complexity” is solved by the dynamic surface control technique. Fuzzy logic systems are employed to approximate the unknown nonlinearity effects and changes in model dynamics due to faults. A fuzzy state observer is presented to estimate the unmeasured states. According to the backstepping technique, a distributed consensus fuzzy controller is developed to guarantee that output signals of all followers and leader are synchronized. It can be proved that all variables of MASs are uniformly ultimately bounded. Finally, the validity of the control scheme is illustrated by some simulation results.

This paper concentrates on the linear least mean square (LLMS) filtered and smoothed estimators for networked linear stochastic systems. Multiple packet losses, Markovian communication constraints, and superposed process noise are considered simultaneously. In order to reduce the channel load during communication, at every step, just one transmission node is permitted to send data packets. Hence, a Markovian communication protocol is utilized to arrange the packets of these transmission nodes. Moreover, multiple data packet dropouts occur during transmission due to an imperfect communication channel. Therefore, the global observation information cannot be obtained by the state estimator. The real state of Markov chain is assumed to be unknown to the estimator except the transition probability matrix. By means of the innovation analysis approach and orthogonal projection principle, we design Kalman-like estimators in a recursive form. Finally, through simulation experiments, we verify the effectiveness and superiority of the designed algorithm.

This paper focuses on the linear minimum mean square estimator for a networked discrete time-varying linear system subject to data quantification and communication constraints. The communication limitation is that only one transmission node can get access to the shared communication channel at each time step, and that different transmission nodes in the networked systems are scheduled to transmit information according to a Markov protocol. Then the remote estimator completes the estimation with only partially available observations, which are quantified. Suppose that the Markov chain is unknown to the remote estimator. By using orthogonal projection principle and innovation analysis method, a Kalman type filter is designed in a recurrence form. It is shown that estimation performance depends on the transition probability matrix of the Markov chain, quantization error, and the shared channel weighting parameter. Finally, an illustrative example is given to show the effectiveness of the proposed method.