This paper investigates the problem of distributed secure state estimation for cyber-physical systems under sparse sensor attacks. Traditional centralized methods struggle with high computational complexity in large-scale systems, while existing distributed approaches either require strong robustness constraints for graphs or face challenges due to local unobservability. To address these challenges, we propose a distributed adaptive saturation method. This method designs an adaptive saturation function for each node to limit the impact of attacks on the estimation residual. Additionally, it integrates a consensus protocol to enable cooperative estimation of nodes, effectively mitigating the influence of malicious attacks. Theoretical analysis shows that the algorithm ensures consensus estimation under fixed attack channels, with estimation errors ultimately bounded. Furthermore, by refining the saturation mechanism for each node, the algorithm is extended to scenarios with time-varying attacks. Simulation experiments validate that the proposed method effectively handles sparse sensor attacks, reduces computational complexity, and significantly enhances system robustness.

This article investigates the model predictive tracking control problem for nonlinear systems represented by Takagi-Sugeno (T-S) fuzzy models. First, in order to alleviate the online computational burden of the MPC algorithm, a simplified model-based predictive strategy is developed, where the nominal system is approximated by a known linear time-varying (LTV) model, enabling the online optimization to be formulated as a quadratic programming (QP) problem. Thereafter, to ensure recursive feasibility and bounded tracking errors under fuzzy approximation errors and bounded disturbances, a Lyapunov-based dynamically updated tracking error constraint is introduced. Furthermore, by converting the original ellipsoidal constraints into relaxed box constraints, computational complexity is further reduced without violating the Lyapunov-based guarantees, which is ensured by an auxiliary one-step optimizer incorporated in the online part of the proposed framework. Compared with existing results, the proposed approach maintains bounded tracking errors with relatively high computational efficiency. The effectiveness of the proposed MPC framework is then demonstrated through simulation results.

This article addresses the problem of event-triggered quantized control for cyber-physical systems under multiple-channel denial-of-service (DoS) attacks, which independently affect each communication channel. First, to conserve communication resources, a decentralized event-triggered scheme (ETS) with quantization is proposed. Specifically, a transmission interval bound is introduced in the decentralized ETS to reduce the transmission frequency, and an adjustable parameter is designed for the quantizer to decrease the number of bits. Second, the DoS intervals are redefined by applying the transmission interval bound. Subsequently, the transmission interval bound is adopted as the update interval of the controller, and an attack switched mode is developed based on the multiple-channel DoS attacks, thereby designing a control strategy. Furthermore, based on Lyapunov theory, sufficient conditions are derived that reveal the relationship among the eventtriggering parameters, quantizer parameters, and attack intensity, ensuring the closedloop stability of the system. Finally, a numerical example is employed to demonstrate the feasibility of the presented approach.

Drug-target interaction prediction plays a vital role in computer-aided drug discovery, yet existing models often suffer from poor generalization to unseen drugs or targets due to distributional shifts. To address this challenge, we propose PAIR-VAE, a novel variational autoencoder framework that incorporates a pairwise augmented interaction reconstruction mechanism. PAIR-VAE reconstructs perturbed variants of drugs and targets to generate three types of synthetic interaction pairs, which are jointly trained with the original pairs under a label-sharing co-regularized objective. To enhance prediction stability, the model introduces a mean-based variational embedding strategy to mitigate the noise caused by stochastic sampling. Experimental results on two benchmark datasets demonstrate that PAIRVAE consistently outperforms existing methods across various generalization settings. Notably, in the challenging unseen-pair scenario, it achieves up to 5% and 6% improvements in AUROC and AUPRC, respectively. Moreover, PAIR-VAE maintains strong performance under a more difficult cross-domain split, where training and test sets share minimal structural similarity, further validating its real-world generalization capability. To the best of our knowledge, PAIR-VAE is the first framework to introduce variational pairwise augmentation and label-sharing supervision for DTI prediction, offering a generalizable and robust solution for real-world drug discovery.

This paper studies the issue of dynamic event-triggered (ET) resilient control for heterogeneous multi-agent systems (MASs) under DoS attacks. Most of the existing work only considers ideal linear models and undirected graph communications. However, in practice, both disturbance and noise exist in the system model, and directed graph communication is more common. To solve this issue, a fully distributed dynamic ET control strategy with a prediction-based dynamic compensation algorithm is designed to deal with the difficulty of communication jamming. By using the Lyapunov method, it is proved that bounded consensus can be achievable in heterogeneous MASs under DoS attacks, and Zeno behavior is excluded. Furthermore, the tolerable attack intensity is quantified by attack frequency and duration. Finally, a numerical simulation is conducted to validate the efficacy of the proposed method.Note to Practitioners—In real-world scenarios, numerous complex tasks necessitate collaborative endeavors among heterogeneous intelligent agents, such as robots swarms, multiple vehicles and smart grids. Due to physical or geographical limitations, direct communication between all agents and a leader is not always feasible, thereby restricting individual agents to indirectly accessing the leader’s information. Furthermore, the practical constraints on network resources, coupled with the susceptibility to malicious DoS attacks, impede the completion of tasks by multiple agents. This paper presents a resilient dynamic ET control strategy for the consensus task of heterogeneous agents, aiming to improve system resilience while conserving network resources.

This article addresses the problem of collision avoidance and obstacle avoidance in nonlinear multiagent control using a data-driven cooperative output approach. First, to avoid collisions with obstacles, a safe reference trajectory planning method is proposed, which dynamically projects unsafe segments of a reference trajectory onto the outer boundary of obstacle regions, enabling the agent to bypass the obstacle. Second, by using only input-output information, a distributed tracking controller incorporating two dynamical barrier functions is designed, which allows the agent to adaptively decelerate within the coverage range of the barrier function to avoid collisions. Finally, sufficient conditions are provided to ensure that the agent can successfully bypass obstacles using the proposed control strategy. Simulation results confirm that the proposed method enables the agent formation to timely avoid stationary/moving obstacles, prevent collisions between agents, and effectively track the desired trajectory while maintaining formation.

This article investigates the stability analysis problem for cyber-physical systems (CPSs) under denial-of-service (DoS) attacks. Based on the real-time data characterizing the suffered DoS attack, a necessary and sufficient condition for the closed-loop stability in the presence of DoS attacks is provided. Besides, by transforming stability analysis the system under DoS into stability analysis of an auxiliary system, novel sufficient conditions for the closed-loop stability, which can be verified more easily than the necessary and sufficient condition, are provided. Based on the proposed conditions, an online stability monitoring strategy, which triggers an alarm if the closed-loop stability is not guaranteed, is proposed. Finally, two examples are given to illustrate the effectiveness of the proposed strategy.

This paper investigates the resilient control problem for cyber-physical systems (CPSs) with multiple transmission channels under denial-of-service (DoS). First, a set of partial observers is designed to estimate partial states corresponding to different channels. Second, by combining the proposed partial observers, the finite-time observer technique and a switching scheme, a modified finite-time partial observer-based controller is proposed to stabilize the system in the presence of DoS. Compared with the existing results for the CPSs with multiple transmission channels, the computational burden of designing observer-based controller is reduced, and the resilience against DoS is improved by adopting the proposed finite-time partial observers. Especially, besides stability, the proposed resilient observer-based controller also provides better disturbance rejection performance.

This article concerns the problem of observer-based event-triggered control for cyber-physical systems (CPSs) under denial-of-service (DoS) attacks. In contrast to the existing studies where DoS attacks on different channels are the same, the considered attacks compromise each channel independently. Correspondingly, a decentralized event-triggered scheme is adopted based on the tradeoff between the transmission efficiency and tolerable attack intensity with guarantees on the closed-loop stability. Inspired by the Lyapunov theory for switched systems, the proposed stabilization criteria reveals a link between the tolerable attack intensity and the event-triggering parameters. An example is finally provided to illustrate the effectiveness of the proposed approaches.

This paper investigates the problem of adaptive fuzzy tracking control for a class of switched uncertain nonlinear systems under asynchronous switching. Inspired by the discretized Lyapunov function technique, a class of time-schedule Lyapunov functions (TSLFs) is first proposed to analyze the stability under asynchronous switching. Then, based on the proposed TSLF, novel observer and adaptive laws with time-schedule observer gain and design parameters are proposed to deal with asynchronous switching. Moreover, a switched command filter with the virtual control functions as input is proposed to avoid “the explosion of complexity” and tracking error jumps, which make the stability analysis difficult. It is shown that under the proposed controller, all the signals of the closed-loop system are bounded, and the tracking errors converge to a small neighborhood of zero in spite of asynchronous switching. Finally, the effectiveness of the proposed method is illustrated through two illustrative examples.

This paper is concerned with the problem of adaptive fuzzy tracking control for a class of switched uncertain nonlinear systems. The unknown nonlinear functions are approximated by using fuzzy logic systems, and the adaptive backstepping technique is applied to a convex combination of the subsystems. To ensure the continuousness of the Lyapunov function at switching instants, common Lyapunov function (CLF) technique is adopted and the proposed CLF permits switched parameter adaptive laws. Besides, two classes of adaptive laws are proposed, one for designing control laws and one for designing switching law. Then, the common control laws are obtained with the help of the convex combination. Based on the switched adaptive laws, common control laws, a new adaptive state-dependent switching law is proposed. It is shown that the designed state-feedback controllers and switching law can ensure that all the signals remain bounded and the tracking error converges to a small neighborhood of the origin. Moreover, a modified switching law is proposed to avoid Zeno behavior or switching too quickly. Finally, two examples are provided to show the effectiveness of the proposed approaches.