Continual Knowledge Graph Embedding (CKGE) aims to continually learn embeddings for new knowledge, i.e., entities and relations, while retaining previously acquired knowledge. Most existing CKGE methods mitigate catastrophic forgetting via regularization or replaying old knowledge. They conflate new and old knowledge of an entity within the same embedding space to seek a balance between them. However, entities inherently exhibit multi-faceted semantics that evolve dynamically as their relational contexts change over time. A shared embedding fails to capture and distinguish these temporal semantic variations, degrading lifelong link prediction accuracy across snapshots. To address this, we propose a Multi-Faceted CKGE framework (MF-CKGE) for semantic-aware link prediction. During offline learning, MF-CKGE separates temporal old and new knowledge into distinct embedding spaces to prevent knowledge entanglement and employs semantic decoupling to reduce semantic redundancy, thereby improving space efficiency. During online inference, MF-CKGE adaptively identifies semantically query-relevant entity embeddings by quantifying their semantic importance, reducing interference from query-irrelevant noise. Experiments on eight datasets show that MF-CKGE achieves an average (maximum) improvement of 1.7% (2.7%) and 1.4% (3.8%) in MRR and Hits@10, respectively, over the best baseline. Our source code and datasets are available at: https://anonymous.4open.science/r/MF-CKGE-04E5.

The state estimation problem of linear time-invariant (LTI) systems is investigated in this article. Distributed observers are designed to estimate the complete states of LTI systems under time-varying delays. A class of edge-based switching event-triggered mechanisms (SETMs) are designed for distributed observers to continuously estimate the states of the target system by the discrete relative state information, reducing the consumption of communication/computation resources significantly. The positive lower bounds of intersampling times are guaranteed, which further avoids Zeno behaviors. A unified framework for the stability analysis of the estimation error system under the SETM is presented, and the sufficient conditions for the asymptotic state estimation of the distributed observer are derived. Finally, numerical simulations are given to illustrate the effectiveness of the proposed method.

Satellite observation is an important way to understand the earth. However, due to the problems such as satellite aging, cloud obstruction, and other interferences during the imaging and transmission process, remote sensing images inevitably produce various defects. Hence, it is necessary to quickly detect defects to calibrate the imaging system and avoid the waste of satellite resource. Current researches on defect detection in remote sensing images are not comprehensive, which only focus on partial defect categories, such as cloud and stripe. To this end, we construct the first large-scale High-resolution Remote Sensing image Defect detection dataset (HRSD). The proposed dataset contains more than 1.2 million manually annotated patches from eight different satellites, covering various common defect categories and including multiple image modalities (i.e., panchromatic and multispectral). The dataset also has rich diversity which covers different landforms in multiple regions. Furthermore, to realize the detection of multiple defect categories simultaneously, we design a feature aggregation graph network (FAGN) based on the position correlation and semantic similarity among image patches, which fully utilizes the distribution characteristics of defects to achieve accurate defect detection. Extensive experiments on the HRSD dataset demonstrated the effectiveness of FAGN. We will release the HRSD dataset and FAGN model later.

In this paper, a robust optimal three-dimensional cooperative guidance law with input saturation is proposed for intelligent aerial vehicles to intercept an unknown maneuvering target. The problem of cooperative interception is formulated as a leader-follower optimal tracking control problem based on the differential graphical game subject to a nonautonomous leading vehicle with bounded control inputs. Utilizing the backstepping method, the guidance law is divided into a feedforward part for generating the desired state signals and compensates for the impact of input saturation, and a data-driven feedback part based on the differential graphical game that regulates tracking errors due to unknown target maneuvers while optimizing interactive performance indices. The uniform ultimate bounded property of the tracking errors in the closed-loop system can be guaranteed and the predefined interactive cost function can be optimized by the proposed guidance law. Simulation examples of cooperative interception are provided to validate the effectiveness of the proposed approach.

In this article, we consider the cooperative output regulation for linear multiagent systems (MASs) via the distributed event-triggered strategy in fixed time. A novel fixed-time event-triggered control protocol is proposed using a dynamic compensator method. It is shown that based on the designed control scheme, the cooperative output regulation problem is addressed in fixed time and the agents in the communication network are subject to intermittent communication with their neighbors. Simultaneously, with the proposed event-triggering mechanism, Zeno behavior can be ruled out by choosing the appropriate parameters. Different from the existing strategies, both the compensator and control law are designed with intermittent communication in fixed time, where the convergence time is independent of any initial conditions. Moreover, for the case that the states are not available, the output regulation problem can further be addressed by the distributed observer-based output feedback controller with the fixed-time event-triggered compensator and event-triggered mechanism. Finally, a simulation example is provided to illustrate the effectiveness of the theoretical results.

This article develops a robust packetized predictive control framework to deal with the quantized-feedback control problem of networked systems subject to Markovian packet dropouts and input saturation. In the proposed framework, the Markov chain model of packet dropout is established from the link of the controller to the actuator. To deal with the quantized measurements, a robust packetized predictive control method is presented with a quantized-feedback law. The problem of unreliable transmission is addressed by proposing a packet dropout compensation strategy with a forgetting factor. An augmented Markovian jump system model is established to take the packet dropouts into account. The synthesis of packetized predictive control is then developed by minimizing a worst case cost function with respect to the model uncertainties. The recursive feasibility of the proposed controller design problem and the mean-square stability of the closed-loop systems are proved, respectively. The proposed packetized predictive control method is demonstrated by simulating a four-tank process system.

In this article, we focus on the problems of consensus control for nonlinear uncertain multiagent systems (MASs) with both unknown state delays and unknown external disturbances. First, a nonlinear function approximator is proposed for the system uncertainties deriving from unknown nonlinearity for each agent according to adaptive radial basis function neural networks (RBFNNs). By taking advantage of the Lyapunov-Krasovskii functionals (LKFs) approach, we develop a compensation control strategy to eliminate the effects of state delays. Considering the combination of adaptive RBFNNs, LKFs, and backstepping techniques, an adaptive output-feedback approach is raised to construct consensus tracking control protocols and adaptive laws. Then, the proposed consensus tracking scheme can steer the nonlinear MAS synchronizing to the predefined reference signal on account of the Lyapunov stability theory and inequality properties. Finally, simulation results are carried out to verify the validity of the presented theoretical approach.

This article considers consensus of first-order/second-order hybrid multiagent systems (MASs) based on game modeling. In the first-order hybrid MAS (HMAS), a subset of agents select the Nash equilibrium of a multiplayer game as their states at each game time and the others update their states with first-order continuous-time (C-T) dynamics. By graph theory and matrix theory, we establish sufficient and necessary conditions for consensus of the first-order HMAS with two proposed protocols. The second-order HMAS is composed of agents whose states are determined by the Nash equilibrium of a multiplayer game and agents whose states are governed by second-order C-T dynamics. Similarly, sufficient and necessary conditions are given for consensus of the second-order HMAS with two proposed protocols. Several numerical simulations are provided to verify the effectiveness of our theoretical results.

This article deals with collision-risk-based event-triggered optimal formation control problems for mobile multiagent systems. First, several collision-risk-related definitions, such as collision-free margin, moving direction angle, collision risk angle, and collision risk level, are proposed for the moving agents. Then, a collision-risk dependent, time-varying, event-triggered heterogeneous communication network topology is developed, where the agent starts obtaining information of the neighboring agents only when collision risks occur among them. Third, an anti-collision control law, which is composed of a switch function, a control force direction function, and a control strength function, is designed to guarantee the collision avoidance formation of multiagents. Fourth, to ensure the formation quality and save control cost of the multiagent system, an optimal formation control scheme with feedforward compensation is designed. Simulation results illustrate that: 1) by using the collision risk information of mobile agents, the proposed control scheme is effective to realize the collision avoidance optimal formation task and 2) the anti-collision formation controller can be implemented with incomplete information of the agents.

This article focuses on the problem of adaptive scaled consensus tracking control for uncertain nonlinear multiagent systems (MASs) subjected to unknown mixed control gains, input delays, and external disturbances. First, a new form of nonlinear MAS is presented by linear state transformation. Second, a state observer based on adaptive radial basis function neural networks is developed to estimate the unmeasured states. A command filter control scheme is deployed to address the problem of increased sharply complexity derived from the conventional backstepping design with the increase of the system order, and the filtered error is compensated by the error compensation mechanism. Third, by taking advantage of the Lyapunov–Krasovskii functionals, a compensation control strategy is intended to exclude the impact of input delays. In addition, a Nussbaum-gain function is used to deal with the problem of uncertain control direction. An adaptive output feedback control approach is raised to construct the scaled consensus tracking control protocol, error compensating signals, and adaptive laws. It is proved that the tracking errors are driven to a small residual set, and all the signal variables are bounded in the closed-loop system. Finally, the effectiveness of the proposed approach is verified by two numerical simulations.

In this article, the${H_{\infty } }$scaled consensus control problem for multiagent systems in the presence of external disturbances and mixed time delays in both input and Lipschitz nonlinearity is investigated. First, a state observer is introduced for each agent based on the output information of the agent. Then, a scaled consensus protocol is proposed via a truncated predictor output-feedback method, which can deal with the input delay. The integral terms with the mixed time delays that are contained in the transformed systems are analyzed by using the Lyapunov–Krasovskii functionals method, and sufficient conditions are obtained to achieve scaled consensus with guaranteed${H_{\infty } }$performance. An iterative procedure is utilized to calculate the linear matrix inequality. By this, the feedback gain and observer gain are then designed. Finally, a simulation example is provided to illustrate the effectiveness of the theoretical results.

This article focuses on scaled consensus tracking for a class of high-order nonlinear multiagent systems. Different from the existing results, for high-order nonlinear multiagent systems with time delays and external disturbances, a fully distributed consensus protocol is designed to drive all agents to achieve scaled consensus with preassigned ratios. The control gains are varying and updated by distributed adaptive laws. As a result, the presented protocol is independent of any global information, and thus, could be implemented in a fully distributed manner. Simultaneously, the fully distributed control protocol using an adaptive$\sigma$-modification technique is presented to deal with external disturbances, which can guarantee the tracking errors and coupling weights of all following agents are uniformly ultimately bounded. To tackle with the derivatives of the functionals with time delays, the Lyapunov–Krasovskii functional is employed to analyze and compensate them by introducing multiintegral terms. Finally, simulation examples are included to verify the effectiveness of the theoretical results.

This article considers iterative learning control (ILC) for a class of discrete-time systems with full learnability and unknown system dynamics. First, we give a framework to analyze the learnability of the control system and build the relationship between the learnability of the control system and the input-output coupling matrix (IOCM). The control system has full learnability if and only if the IOCM is full-row rank and the control system has no learnability almost everywhere if and only if the rank of the IOCM is less than the dimension of system output. Second, by using the repetitiveness of the control system, some data-based learning schemes are developed. It is shown that we can obtain all the needed information on system dynamics through the developed learning schemes if the control system is controllable. Third, by the dynamic characteristics of system outputs of the ILC system along the iteration direction, we show how to use the available information of system dynamics to design the iterative learning gain matrix and the current state feedback gain matrix. And we strictly prove that the iterative learning scheme with the current state feedback mechanism can guarantee the monotone convergence of the ILC process if the IOCM is full-row rank. Finally, a numerical example is provided to validate the effectiveness of the proposed iterative learning scheme with the current state feedback mechanism.

With the incentive to solve Nash equilibrium computation problems for networked games, this article tries to find answers for the following two problems: 1) how to accommodate hybrid games, which contain both continuous-time players and discrete-time players? and 2) are there any other potential perspectives for solving continuous-time networked games except for the consensus-based gradient-like algorithm established in our previous works? With these two problems in mind, the study of this article leads to the following results: 1) a hybrid gradient search algorithm and a consensus-based hybrid gradient-like algorithm are proposed for hybrid games with their convergence results analytically investigated. In the proposed hybrid strategies, continuous-time players adopt continuous-time algorithms for action updating, while discrete-time players update their actions at each sampling time instant and 2) based on the idea of consensus tracking, the Nash equilibrium learning problem for continuous-time games is reformulated and two new computation strategies are subsequently established. Finally, the proposed strategies are numerically validated.

In this brief, we consider the consensus problem of hybrid multiagent systems. First, the hybrid multiagent system is proposed, which is composed of continuous-time and discrete-time dynamic agents. Then, three kinds of consensus protocols are presented for the hybrid multiagent system. The analysis tool developed in this brief is based on the matrix theory and graph theory. With different restrictions of the sampling period, some necessary and sufficient conditions are established for solving the consensus of the hybrid multiagent system. The consensus states are also obtained under different protocols. Finally, simulation examples are provided to demonstrate the effectiveness of our theoretical results.

In this paper, we study the consensus problem of discrete-time and continuous-time multiagent systems with distance-dependent communication networks, respectively. The communication weight between any two agents is assumed to be a nonincreasing function of their distance. First, we consider the networks with fixed connectivity. In this case, the interaction between adjacent agents always exists but the influence could possibly become negligible if the distance is long enough. We show that consensus can be reached under arbitrary initial states if the decay rate of the communication weight is less than a given bound. Second, we study the networks with distance-dependent connectivity. It is assumed that any two agents interact with each other if and only if their distance does not exceed a fixed range. With the validity of some conditions related to the property of the initial communication graph, we prove that consensus can be achieved asymptotically. Third, we present some applications of the main results to opinion consensus problems and formation control problems. Finally, several simulation examples are presented to illustrate the effectiveness of the theoretical findings.In this paper, we study the consensus problem of discrete-time and continuous-time multiagent systems with distance-dependent communication networks, respectively. The communication weight between any two agents is assumed to be a nonincreasing function of their distance. First, we consider the networks with fixed connectivity. In this case, the interaction between adjacent agents always exists but the influence could possibly become negligible if the distance is long enough. We show that consensus can be reached under arbitrary initial states if the decay rate of the communication weight is less than a given bound. Second, we study the networks with distance-dependent connectivity. It is assumed that any two agents interact with each other if and only if their distance does not exceed a fixed range. With the validity of some conditions related to the property of the initial communication graph, we prove that consensus can be achieved asymptotically. Third, we present some applications of the main results to opinion consensus problems and formation control problems. Finally, several simulation examples are presented to illustrate the effectiveness of the theoretical findings.

This paper focuses on the finite-time consensus (FTC) problem of switched multiagent system (MAS) which is composed of continuous-time and discrete-time subsystems. Different from the existing results, each agent of this system is controlled by switching control method. To achieve consensus in finite time for the switched MAS, two types of consensus protocols (the FTC protocol and the fixed-time consensus (FdTC) protocol) are proposed. By using algebraic graph theory, Lyapunov theory and matrix theory, it is proved that the FTC problem in strongly connected network and leader-following network can be solved, respectively. When the initial states of agents are not available, the FdTC protocol is applied to solve the FTC problem. Simulations are provided to illustrate the effectiveness of our theoretical results.