This article investigates the secure consensus tracking control problem for nonlinear multiagent systems under Byzantine attacks and unknown nonlinearities. Addressing structural limitations and vulnerabilities in sparse topologies, the multihop mean-subsequence-reduced algorithm is adopted to enhance system robustness and information availability by enabling message relay through healthy intermediate nodes. Differing from existing works primarily on linear dynamics and static average consensus, this study focuses on the multihop mean-subsequence-reduced method for nonlinear state-feedback control frameworks. Radial basis function (RBF) neural networks are integrated for secure approximation of unknown nonlinearities. Using the backstepping method, a novel secure control strategy is synthesized. The proposed scheme rigorously guarantees the convergence and boundedness of the closed-loop system, ensuring accurate tracking for nonlinear multiagent systems.

With the rapid development and widespread deployment of advanced cyber-physical systems, malicious network attacks during information transmission have introduced severe challenges to their security and reliability. To address these challenges, this article provides a comprehensive review that begins with a systematic classification of major attack models threatening cyber-physical systems. On this basis, attack detection and secure control methods are jointly reviewed from both model-based and data-driven perspectives. Furthermore, the applications of security techniques for advanced cyber-physical systems are discussed across several representative domains, including power and energy systems, intelligent transportation systems, networked robotic systems, and spacecraft systems, where secure and reliable operation is essential. Finally, key challenges and promising research directions for future exploration are outlined. Overall, this article is beneficial for promoting the development of security control methods and the construction of higher-level advanced cyber-physical systems and their applications.

This article proposes a secure load frequency control scheme for multiarea power systems subject to renewable energy disturbances and false data injection attacks (FDIAs). A novel integration of high-order control barrier functions (HOCBFs) and disturbance observers (DOBs) is presented. The DOB provides real-time estimation and compensation of disturbances and the HOCBF rigorously enforces frequency safety constraints under dynamic uncertainties. A key innovation is the development of a resilient HOCBF (R-HOCBF) to counteract diverse FDIA types, including scaling, additive, and affine attacks, by guaranteeing safety over the full range of attack parameters. Stability and safety are unified within a quadratic programming formulation. Simulations on a three-area power system demonstrate effectiveness of the scheme in maintaining frequency within safe limits under concurrent renewable variations and cyber attacks, outperforming conventional PI and standalone DOB-based methods.

This work addresses the nonfragile fault-tolerant control for power cyber-physical systems (CPSs) under denial-of-service (DoS) attacks, in which the cyber attacks are considered to be strongly concealed. In the fact of power CPSs frequently subject to various attacks during the operation, the innovation is to construct a double-layer stochastic process composed of a semi-Markov chain and a sequence of observed mode to analyze the DoS attacks from the viewpoint of hidden semi-Markov chain. To address potential actuator failures and controller gain perturbations, an observed-mode-dependent nonfragile control scheme is developed. By constructing a mode-dependent Lyapunov function that incorporates both attack modes and observed modes, sufficient conditions are derived to ensure mean-square stability of the closed-loop system through the semi-Markov kernel (SMK) approach, which systematically handles the stochastic characteristics of dwell time (DT) distributions under incomplete attack mode information. Finally, a simulation example demonstrates the validity of the proposed approach.

Networked control systems (NCSs) often suffer from performance degradation due to limited communication bandwidth, which can cause data transmission conflicts and packet loss. Existing scheduling strategies may fail to simultaneously meet the real-time requirements and the importance of multisensor data, and they are particularly vulnerable under distributed denial of service (DDoS) attacks. Firstly, to address these challenges, a greedy algorithm is proposed to optimise the data transmission process to satisfy the importance and real-time requirement of sensor data. It enables the dynamic allocation of network resources, reduces the possibility of data conflict and improves the communication efficiency. Then, an observer-based control algorithm is designed to ensure the system's stability and security resilience under bandwidth constraints and DDoS attacks. Therefore, in the face of packet loss caused by data conflict, the control algorithm can maintain efficient resource scheduling capability and improve the robustness of the system. Finally, the simulation results show that the proposed dynamic resource scheduling framework can guarantee the performance of NCSs under constrained network conditions.

This paper investigates the leader-following mean-square consensus of stochastic linear multi-agent systems (MASs) via distributed impulsive control. A distributed impulsive control law relying on local neighbor information and discrete sampled data is designed. By using Lyapunov theory, stochastic analysis, algebraic graph theory and the linear matrix inequality (LMI) technique, sufficient conditions for mean-square consensus are derived. We provide methods for coupling gains and impulsive control gains and deduce the consensus result of deterministic linear MASs as a corollary. Numerical simulations demonstrate the effectiveness of the proposed control.

This paper presents an output feedback adaptive boundary control scheme for a hyperbolic ODE-PDE-ODE cascade system with significant uncertainties. The system considers that all parameters, including spatially varying ones, are unknown and that only the ODE subsystem states are measurable. To facilitate the control design, an equivalent system representation is first derived. State observers are then constructed using linear combinations of ODE-PDE cascade filters, while swapping identifiers provide online parameter estimates. By integrating these elements, an adaptive backstepping control law is designed. It is proved that the proposed adaptive control law can stabilize the studied system. Finally, the effectiveness of the derived controller is verified via numerical simulations.

This paper investigates a physics-informed deep reinforcement learning approach for the agile tracking and obstacle avoidance control of quadrotors. Traditional control methods often struggle to account for the strong aerodynamic coupling and nonlinear dynamics inherent in high-speed maneuvers. While deep reinforcement learning offers a promising alternative, standard black-box approaches frequently suffer from low sample efficiency, limited physical interpretability, and slow strategy convergence. To overcome these limitations, a physics-informed deep reinforcement learning framework built upon the proximal policy optimization algorithm is proposed. By incorporating Newton-Euler dynamics, a physics-informed neural network architecture is integrated where the drone’s governing equations are embedded as physical residual terms within the total loss function. A robust state space that accounts for measurement noise and a normalized action space are arranged to respect motor physical constraints. Furthermore, a dense reward function is formulated to balance tracking precision, flight stability, and task safety. To ensure continuous and smooth control commands, Kullback-Leibler divergence regularization and momentum-based smoothing techniques are implemented. Simulation results show that the proposed algorithm can effectively improve the tracking accuracy and obstacle avoidance safety of quadrotors.

This paper studies the output feedback resilient secure tracking control strategy for a class of spacecraft attitude-orbit integrated systems with aperiodic denial-of-service (DoS) attacks on the sensor based on the fully actuated system approach. The attitude-orbit integrated system is defined as a six-degree-of-freedom nonlinear system with strong attitude-orbit coupling characteristics. Due to the limitations of measurement accuracy and physical conditions, the states of practical systems are not fully measurable. The data interactions between spacecraft sensors and controllers are based on the communication networks, which exhibit vulnerability to DoS attacks. Some existing works fail to consider unmeasurable states and secure communication burdens. The spacecraft attitude-orbit integrated model is given based on dual quaternions, and the fully actuated system subject to aperiodic DoS attacks is constructed based on twistor, variable elimination, and order elevation techniques. A resilient secure observer is designed based on the system output, and a secure control law is formed using the estimated states. A Lyapunov function is selected for the estimation error and the tracking error, and the system stability in different attack stages is quantitatively analyzed. Finally, the simulation results verify the effectiveness of the proposed output feedback secure control method.

This article concentrates on the decentralized self-triggered control problem for state-unknown nonlinear interconnected systems within reinforcement learning framework. To simultaneously estimate the partially unknown dynamics and unavailable state of nonlinear interconnected systems, a decentralized learning observer is designed for the constructed auxiliary subsystems via output data. From the resource-efficient orientation, a novel dynamic factor-enhanced self-triggered scheme (DFSTS) with dead-zone operation is proposed to predict the data transmission and controller updating moment. Distinct from the existing schemes, the distinguishing advantages of DFSTS are twofold: 1) the successive monitoring for event-triggered condition is not required; 2) the utilization of dynamic factor and dead-zone operation leads to the larger release interval. Moreover, a DFSTS-incorporated critic-sole neural network is presented to approximate the optimal cost function with the improved weight tuning policy, in which the constraint of initial admissible control signal is obviated. It is assured that the critic weight approximation error and the target system state are uniformly ultimately bounded under the designed intelligent control strategy. Eventually, the simulation results containing the effectiveness validation and comparison analysis illustrate the practicability and superiority of the learning-observer-guided decentralized self-triggered control algorithm.

This paper investigates the synchronization issue in uncertain quaternion-valued neural networks (QVNNs), which are a subclass of hypercomplex networks, by decomposing the quaternion fields to real-valued fields. First, a reliable controller integrated with preventive and fault-tolerant mechanisms is developed to mitigate the risk of impulsive controller failure. Specifically, the impulsive controller is employed as the primary control component, whereas a state-feedback controller acts as the backup to ensure system reliability. The propsed dual-controller approach enhances robustness under various conditions and guarantees reliability against impulsive controller failures. Furthermore, the iterative optimization algorithm is proposed, which can adjust the probability of impulsive control to maximize its utilization. By fine-tuning the impulsive control parameters, the algorithm optimizes the use of the primary controller while minimizing reliance on the backup state-feedback controller, thereby improving overall system efficiency and responsiveness. Then, by employing the classical Lyapunov theory and impulsive differential Halanay inequality, several sufficient conditions for synchronization are derived. Finally, a numerical example is presented to verify the effectiveness of the theoretical results, demonstrating the robustness and reliability of the proposed controller for QVNNs.

This paper investigates the secure consensus control problem for strict-feedback nonlinear multi-agent systems with Byzantine attacks and unmeasurable states. Firstly, a fuzzy state observer is constructed to reconstruct unmeasurable states. Then, the mean-subsequence-reduced algorithm is used to design the secure control scheme, which employs a data filtering strategy rather than an agent identification strategy. Specifically, the healthy agent sorts the state information received from its neighbors and discards the extreme values, and the Byzantine attack issue is solved. The proposed control scheme can guarantee that all the signals in the closed-loop system are bounded and the consensus tracking errors converge to a small neighborhood of the origin. Finally, simulation results are given to verify the effectiveness.

Fixed-wing autonomous aerial vehicles (AAVs) performing turn maneuvers in turbulent environments face significant control challenges due to stochastic aerodynamic disturbances. This paper proposes a hybrid turbulence-aware control framework that integrates mathematical modeling, global optimization, and machine learning for adaptive autopilot gain tuning. A semi-invariant-based deterministic reduction of the stochastic lateral-directional model is developed, reducing the dimensionality from 44 to 7 equations while preserving control-relevant statistical characteristics. To determine optimal feedback gains, a Modified Survival of the Fittest Algorithm (MSoFA) is introduced, demonstrating improved global convergence reliability and achieving 5-15% better solution quality compared to conventional evolutionary methods. An offline-generated dataset of optimal gains across varying turbulence intensities and flight altitudes is then used to construct a clustered Long Short-Term Memory (C-LSTM) architecture that combines DBSCAN-based regime identification with recurrent neural approximation. The proposed neural controller achieves 98.3% gain prediction accuracy and reduces computation time by more than three orders of magnitude relative to direct optimization. Extensive numerical simulations confirm stable turn performance over altitudes ranging from 800 to 1500 m and turbulence scales from 150 to 470 m, with terminal yaw-rate deviations below 0.01%. The novelty of the proposed framework lies in the structured integration of 1) semi-invariant-based stochastic model reduction, 2) globally reliable evolutionary optimization for constrained gain tuning, and 3) regime-dependent recurrent neural approximation for real-time implementation. To the best of the authors' knowledge, such an integrated turbulence-aware tuning framework for fixed-wing AAV autopilots has not been previously reported. Note to Practitioners-This paper is motivated by the difficulty of maintaining precise flight trajectories for fixed-wing Autonomous Aerial Vehicles (AAVs) in turbulent wind conditions, which is critical for applications ranging from agricultural monitoring to package delivery. Standard autopilots often lack the adaptability for sudden gusts, while sophisticated real-time optimization typically requires computing power unavailable on small, battery-powered onboard computers. This work addresses this trade-off by introducing a hybrid control framework that shifts the heavy computational burden offline, using a robust optimization method to generate a dataset of optimal control parameters that subsequently trains a specialized neural network for onboard use. The resulting system enables real-time adaptation, operating approximately 2000 times faster than traditional mathematical optimization and capable of running on standard embedded hardware with negligible latency. The primary contribution is a practical methodology for implementing sophisticated, turbulence-aware control on resource-constrained platforms without sacrificing accuracy. However, practitioners should note that the controller's reliability is strictly limited to the operational envelopes, such as altitude ranges and turbulence intensities, defined in the training dataset; performance is not guaranteed outside these learned boundaries without retraining. Beyond aerial vehicles, this strategy of replacing computationally expensive optimization with fast neural approximation is applicable to other autonomous systems, such as underwater vehicles or ground robotics, that require rapid adaptive control in dynamic environments.

This article investigates the output feedback resilient secure tracking control problem for spacecraft attitude-orbit integrated system with denial-of-service (DoS) attacks on sensor. This system is characterized as a multidimensional complex plant that integrates six degrees-of-freedom, coupling dynamics, and complex nonlinearities. Owing to the existence of network communications, spacecraft output feedback control system displays vulnerability to DoS attacks. Some existing results require the rigorous assumption of the reliable system output signal, in which the sensor signal excludes secure communication burden. Based on dual quaternion concepts, the spacecraft relative attitude-orbit coupled dynamics subject to periodic DoS attacks is established, explicitly incorporating the attack schedule into the system. A resilient secure dual quaternion observer is designed to estimate unknown states. Then a recursive scheme is employed to design the resilient secure control law, the uniform ultimate boundedness of the closed-loop system is formally proven through rigorous Lyapunov analysis, and the secure parameter constraints are derived by the joint analysis of different attack periods. Finally, the numerical simulation is presented to verify the effectiveness of the proposed scheme.

This article addresses the event-triggered consensus control problem for nonlinear multiagent systems under resource constraints, where a unified event-triggered mechanism (ETM) is proposed to conserve system resources through discrete updates of both control laws and neural networks (NNs). First, an improved dual-level game approach is developed specifically for ETM design, treating the control law and event-triggered error as adversarial players to derive an optimal control law and a maximum allowable triggering error threshold. Unlike existing dual-level game-based ETMs, whose parameter design relies on an unknowable constant bounding the cost function gradient norm with respect to the consensus error, the proposed improved approach eliminates this dependence. Second, within the adaptive dynamic programming framework, event-triggered NNs reduce computational load by updating weights solely at triggering instants with rigorous theoretical guarantees establishing the quantitative relationship between weight estimation errors and cost functions. Finally, the unified ETM coordinates control execution and NN updates via logical "union" relations, ruling out both Zeno and singular phenomena. Simulations validate its effectiveness and superiority.

This paper investigates the modeling and output-feedback tracking control issues for the attitude-orbit integrated spacecraft based on the fully actuated system approach. As traditional spacecraft attitude and orbit control schemes treat the attitude and orbit motions as isolated subsystems, they ignore the coupling characteristics between the attitude dynamics and the orbit dynamics, failing to improve the control performance. Although attitude-orbit integrated control systems enable the integrated description and control of the six degrees-of-freedom (DoF) pose systems, they have strong nonlinearities which bring difficulties to the control design and analysis. In practical engineering circumstances, the physical conditions and the measurement accuracy make part of the system states unknown, which limits the application of the state-feedback control techniques. The development of the fully actuated control theory provides solutions to the above issues. Hence, the twistor, variable elimination, and order elevation techniques are adopted to transform the spacecraft relative dual quaternion model into the fully actuated attitude-orbit integrated system. For the output-feedback tracking control objective, a state-observer is designed to estimate the unknown states, and the tracking control law is given based on the estimation signals. A Lyapunov function is selected to analyze the stability of the closed-loop system, and the estimation and tracking errors are proved to be asymptotically convergent. Finally, a numerical simulation is carried out to verify the effectiveness of the proposed control scheme.

This paper investigates the modeling and state-feedback tracking control problems of the attitude-orbit integrated spacecraft based on the fully actuated system approach. Traditional spacecraft control strategies separate the attitude system from the orbit system, ignoring their strong coupling characteristics, which limits the control accuracy and cannot meet the requirements of complex space missions. Although mathematical tools such as dual quaternions enable the integrated description of attitude-orbit motion, the integrated models have strong nonlinearities and input-state mismatches, which bring difficulties to control law design and stability analysis. To solve these problems, the attitude-orbit integrated kinematics and dynamics are established based on the relative dual quaternion between the controlled spacecraft and the target. Then, the twistor is adopted to eliminate the redundant components in the dual quaternion system, and the six-degree-of-freedom (6-DOF) fully actuated system is obtained based on the twistor-based variable elimination and order elevation methods. For the tracking control objective, a state-feedback tracking control law is designed. The stability of the closed-loop system is rigorously analyzed via a Lyapunov function, and the tracking error is proved to be asymptotically convergent. Finally, the numerical simulation verifies the effectiveness of the proposed system model and control method.

With the proliferation of networked systems in Internet of Things (IoT) infrastructures, efficient control strategies for information-constrained environments have become increasingly critical. This article investigates the event-driven model-free optimal control problem for completely unknown nonlinear systems by means of integral reinforcement learning (IRL) and compensation mechanism. For the original unavailable dynamics, a general-form input compensator is designed to establish a partially-unknown system model. To conserve limited communication bandwidth, a dynamic event-triggered scheme (DETS) is implemented, where only state data satisfying the predefined condition are transmitted over the network. Subsequently, a single critic structure-based IRL algorithm is proposed to address the Hamilton–Jacobi–Bellman equation (HJBE) with DETS, where the critic weights are tuned by an improved gradient descent method. By incorporating an adjustable state-dependent term into the update rule, the convergence of the critic weights is ensured without an initial admissible policy. Moreover, the experience replay technique is employed to release the persistence of excitation condition. The stability analysis of the reconstructed system is conducted in accordance with the Lyapunov principle. Eventually, the feasibility and practicality of the developed control scheme are validated through a simulation example.

This paper studies the observer-based secure adaptive output feedback orbit tracking control issue for spacecraft systems with false data injection (FDI) attacks. By means of the inertial-line-of-sight coordinate system using relative distance, line-of-sight inclination and deviation angles, the model of spacecraft orbit system is constructed. An observer is designed to estimate unknown system states, thus improving the accuracy of the control system. The mechanism of parameter adaptive law is utilised to estimate the upper bound of false data injection attacks. Based on the back-stepping technique, the recursive scheme is developed to establish virtual and actual control laws step by step. The proposed secure adaptive output feedback tracking control scheme can not only guarantee that the estimation error and the tracking error converge to a small neighbourhood of the origin, but also ensure that the spacecraft orbit closed-loop system is uniformly ultimately bounded via the Lyapunov function. Finally, the effectiveness of the proposed scheme is demonstrated by a numerical simulation example.