As an important long-range transportation carrier in the future Internet of Aerospace Things (IoAT), hypersonic flight vehicles (HFVs) will play a significant role in intelligent transportation systems (ITS). In IoAT architectures, HFVs function as intelligent edge nodes that must operate autonomously under severe uncertainties with limited onboard computational resources. However, strong coupling effects and unmodeled dynamics pose considerable challenges for the precise modeling and efficient control of HFVs. This paper proposes a novel framework for designing high-precision modeling strategies and efficient control algorithms for HFVs. The framework is named the Online-Enhanced Koopman (OE Koopman) Model Predictive Static Programming (MPSP) framework and is developed utilizing autoencoder neural networks. By leveraging the Koopman operator to lift nonlinear systems into high-dimensional linear spaces, the precise modeling associated with HFVs can be significantly simplified. Both the offline pre-training and online correction mechanisms are integrated into the OE Koopman-MPSP framework, which overcomes the issues of low precision in analytical modeling methods and insufficient data in deep learning methods. In the offline phase, autoencoder neural networks are utilized to construct the basic Koopman operator model. In the online phase, a few-shot-based Extended Dynamic Mode Decomposition (EDMD) method is employed to build compensatory operators for adapting to environmental changes. The proposed framework is specifically designed for resource-constrained IoAT edge devices, where computational efficiency and autonomous adaptation are critical. Experimental results demonstrate the effectiveness of the developed framework and specific algorithms.

Robotic arms serve as critical actuation edge nodes in Industrial Internet of Things (IIoT)-enabled intelligent manufacturing systems. In distributed industrial cyber-physical architectures, multiple robotic manipulators are required to perform autonomous and cooperative motion planning under obstacle-rich and dynamically coupled environments. However, existing multi-agent deep reinforcement learning approaches often exhibit limited scalability, redundant observation processing, inefficient experience utilization, and unstable convergence when deployed in high-dimensional cooperative scenarios. To address these challenges, this paper proposes an Attention-based Prioritized Trajectory Multi-Agent Deep Deterministic Policy Gradient (ATP-MADDPG) framework tailored for edge-coordinated multi-arm systems in IIoT environments. The proposed framework incorporates an adaptive attention mechanism to selectively emphasize critical interaction features while suppressing redundant sensory information, thereby enhancing decision efficiency at distributed edge nodes. A prioritized sequence experience replay (PSER) strategy is further introduced to improve the utilization of cooperative trajectory data and accelerate policy evolution. In addition, curriculum learning is employed to enable progressive training and scalable policy refinement for complex multi-arm tasks. Extensive simulations demonstrate that, compared with conventional MADDPG and representative baselines, the proposed ATP-MADDPG achieves higher task success rates, improved cumulative rewards, faster convergence, and enhanced policy stability. These results validate the effectiveness of the proposed framework for distributed cooperative motion planning in IIoT-oriented robotic systems.

Prescribed performance (PP) control is widely used in the construction of consensus protocols for multiagent systems (MASs) due to its property of ensuring that the variables of interest are constrained within the prescribed range during the control process. However, when unpredictable faults such as sudden sensor faults occur, or parameters such as the sampling interval are selected improperly, it can cause singularity problems and render the PP protocol ineffective. Introducing shear mapping into the PP mechanism can resolve the singularity problems, but it requires solving complex nonlinear equations, which may heavily occupy agents' computational resources. To address this issue, we propose a novel dynamic shear mapping mechanism, based on which an event-triggered PP consensus protocol is developed for a class of heterogeneous leaderless MASs. Specifically, by constructing a dynamic shear angle related to the constraint performance functions and variables of interest, the need to solve nonlinear equations is reduced, while the hard-soft transition of performance constraint in the control process is achieved. It is proven that, under the proposed protocol, the consensus errors can strictly satisfy the PP requirements during a prescribed stage, and ultimately converge to zero asymptotically. Finally, a simulation example is provided to demonstrate the effectiveness of the proposed method.

In this paper, the consensus problem for a class of multi-agent systems is investigated, where each agent is described by an Euler-Lagrange system and only sampled-data output interaction among agents is allowed. In addition, the Euler-Lagrange system considered possesses a higher degree of uncertainty; specifically, the regression matrix is also unknown. The intricate heterogeneity, inherent uncertainties, and stringent constraints of information interactions compel us to develop a new protocol. The protocol is synthesized by organically integrating theories and techniques such as graph theory, sampled-data cooperative control, fixed-time control, and fuzzy approximation. Each agent is equipped with a first-order difference trajectory generator that updates exclusively at sampling instants using the output information received from its neighbors. The virtual trajectory serves as the tracking target for the corresponding agent systems output. A fixed-time fuzzy adaptive tracker is proposed for each agent to track the generated trajectory. The problem of analyzing the protocol's effectiveness is deconstructed into two coupled subproblems: a synchronization problem for a perturbed first-order differentiator and a practical fixed-time tracking problem for a single Euler-Lagrange system. This problem is adequately addressed primarily based on the Lyapunov function method. Finally, the effectiveness of the proposed protocol is validated through numerical simulation and ROS experiment. Code is available at:https://github.com/Consensus-EL-output/tfs.

Continuum robots are well-suited for constrained environments due to their superior flexibility and structural compliance. However, relying solely on passive compliance may lead to damage to both the robot and the surrounding environment. This work proposes a finite-time Cartesian impedance control scheme for tendon-driven continuum manipulators (TDCMs), where a second-order low-pass filter is used to adjust the reference trajectory according to the external robot tip force. The controller is designed using the command filtered backstepping method, and the finite-time stability is established by the designed Lyapunov function. In TDCM systems, the tendons operate antagonistically, and actuators often fail to quickly reach the desired tendon tension, leading to partial failures. To address this, we propose an actuator fault compensation algorithm to enhance system performance and reliability. We conducted trajectory tracking experiments on a multi-segment TDCM prototype, the results demonstrate that the designed Cartesian impedance controller achieves effective compliance control effect and high position control accuracy.

Heterogeneous multi-agent systems play a pivotal role in the Internet of Things (IoT) by enabling collaborative intelligence across diverse devices, yet they inherently struggle with discontinuous communication and inaccessible internal data across neighboring platforms. In this paper, a novel protocol design method, named target point traction method, for heterogeneous multi-agent systems is proposed. The method addresses protocol design for two types of heterogeneous multi-agent systems, with a focus on consensus which is a fundamental issue in cooperative control. It aims to overcome collaborative challenges caused by communication limits, mainly the non-interaction of continuous and internal information. First, multi-agent systems consisting of agents described by first-and second-order systems subject to disturbances are considered. Then, a more general heterogeneous nonlinear multi-agent system is investigated, where the order number and nonlinearities of each agent can be different. The existence of the protocol that can accomplish the given cooperative control task for the investigated multi-agent systems is discussed. The implementation of the protocols developed by the method only depends on the local sampled-data output interaction. Under the proposed control schemes, the system output evolution of each agent in the sampling instants is equivalent to the state evolution of a (perturbed) first-order differentiator, which can effectively reduce the conservatism in the selection of the sampling period. Finally, the validity of the developed method is verified by numerical examples.

This article studies the fuzzy-based fixed-time attitude control problem for quadrotor unmanned aerial vehicle under full-state constraints. The backstepping method is utilized to design the fixed-time attitude controller. To avoid the singularity problem, a new switching function is designed in the controller design. Barrier Lyapunov functions are employed to ensure that the system states always satisfy the constraints, and fuzzy logic systems (FLSs) are introduced to approximate the unknown nonlinear functions of the system model. It is proved that the tracking errors converge to a small region around the origin in fixed time. To validate the effectiveness of the method, software-in-loop simulation and practical flight experiments are carried out with the PX4 platform.

A collision-free formation control problem for multiple unmanned aerial vehicles (UAVs) with directed switching topologies and disturbances is investigated. A novel distributed control algorithm that uses UAVs local directed switching information is proposed for achieving the required flight formation and ensuring a safe distance between UAVs; the algorithm involves the incorporation of the APF method in the virtual leader formation scheme. Two command signals generated by the virtual position controller are transmitted to the attitude subsystem. For each UAV, an adaptive composite controller is designed by combining a fuzzy system and fast terminal sliding mode control technique to guarantee that tracking errors converge to a stable area around zero. Finally, the feasibility of the proposed composite control algorithm is demonstrated through a simulation.

This article studies the optimal synchronization of linear heterogeneous multiagent systems (MASs) with partial unknown knowledge of the system dynamics. The object is to realize system synchronization as well as minimize the performance index of each agent. A framework of heterogeneous multiagent graphical games is formulated first. In the graphical games, it is proved that the optimal control policy relying on the solution of the Hamilton-Jacobian-Bellmen (HJB) equation is not only in Nash equilibrium, but also the best response to fixed control policies of its neighbors. To solve the optimal control policy and the minimum value of the performance index, a model-based policy iteration (PI) algorithm is proposed. Then, according to the model-based algorithm, a data-based off-policy integral reinforcement learning (IRL) algorithm is put forward to handle the partially unknown system dynamics. Furthermore, a single-critic neural network (NN) structure is used to implement the data-based algorithm. Based on the data collected by the behavior policy of the data-based off-policy algorithm, the gradient descent method is used to train NNs to approach the ideal weights. In addition, it is proved that all the proposed algorithms are convergent, and the weight-tuning law of the single-critic NNs can promote optimal synchronization. Finally, a numerical example is proposed to show the effectiveness of the theoretical analysis.

This article puts forward an observer-based finite-time sampled-data stabilization scheme for a nonlinear time-delay system. To overcome the difficulties of stabilizing such nonlinear system under consideration, a reduced-order observer, whose role lies in estimating unavailable states, is formulated by relying on detectable sampled output, subsequently, a finite-time sampled-data output-feedback stabilizer (FSOS), which possesses suitable scalars and sampling period, can be developed with the help of adding a power integrator (AAPI) technique, such stabilizer can drive the formulating closed-loop system to be globally practically finite-time stable (GPFS) in the presence of uncertain time delays, which requires to be verified by means of established Lyapunov-Krasovskii functionals (LKFs). The availability of the developed scheme can be reflected by two simulations in the end.

In this study, the sampled-data consensus problem is investigated for a class of heterogeneous multiagent systems (MASs) in which each agent is described by a second-order switched nonlinear system. Owing to the heterogeneity and the occurrence of dynamic switching in the MASs, the sampled-data consensus protocol design problem is challenging. In this study, two periodic sampled-data consensus protocols and an event-triggered consensus protocol are developed. Here, we first propose a new periodic sampled-data consensus protocol that involves the local objective trajectory interaction among agents. The protocol is then improved by applying the finite-time control and sliding-mode control techniques. Notably, the improved protocol can be implemented without the transmission of constructed auxiliary dynamical variables, which is a major feature of the present study. It is shown that complete consensus of the underlying MASs can be achieved by the two proposed protocols with only sampled-data measurements. To further reduce the communication load, we introduce an event-triggered mechanism to obtain a new protocol. Finally, the effectiveness of the given schemes is demonstrated by considering a numerical example.

This article investigates a distributed fuzzy adaptive formation control for quadrotor multiple unmanned aerial vehicles (UAVs) under unmodeled dynamics and switching topologies. The UAVs dynamics model is described by the Newton–Euler formula, and the actuator faults are considered in the system model in the form of multiplicative factors and additive factors. Due to the underactuated characteristics of the UAVs, two objective attitude commands are generated by designing a virtual control signal, which are transmitted to the attitude subsystem, and then the position controller is solved. By constructing a distributed communication mechanism between UAVs, an adaptive formation control strategy is proposed, which can enable UAVs to update their position and speed online according to their neighbor information, and then achieve the required formation. In addition, a fuzzy adaptive sliding mode controller is designed to ensure that the tracking errors of UAVs converge to the neighborhood of the origin. Finally, the simulation results verify the effectiveness of the proposed control strategy.

The problem of neural adaptive distributed formation control is investigated for quadrotor multiple unmanned aerial vehicles (UAVs) subject to unmodeled dynamics and disturbance. The quadrotor UAV system is divided into two parts: the position subsystem and the attitude subsystem. A virtual position controller based on backstepping is designed to address the coupling constraints and generate two command signals for the attitude subsystem. By establishing the communication mechanism between the UAVs and the virtual leader, a distributed formation scheme, which uses the UAVs' local information and makes each UAV update its position and velocity according to the information of neighboring UAVs, is proposed to form the required formation flight. By designing a neural adaptive sliding mode controller (SMC) for multi-UAVs, the compound uncertainties (including nonlinearities, unmodeled dynamics, and external disturbances) are compensated for to guarantee good tracking performance. The Lyapunov theory is used to prove that the tracking error of each UAV converges to an adjustable neighborhood of zero. Finally, the simulation results demonstrate the effectiveness of the proposed scheme.

In this article, an adaptive finite-time tracking control scheme is developed for a category of uncertain nonlinear systems with asymmetric time-varying full-state constraints and actuator failures. First, in the control design process, the original constrained nonlinear system is transformed into an equivalent "unconstrained" one by using the uniform barrier function (UBF). Then, by introducing a new coordinate transformation and incorporating it into each recursive step of adaptive finite-time control design based on the backstepping technique, more general state constraints can be handled. In addition, since the nonlinear function in the system is unknown, neural network is employed to approximate it. Considering singularity, the virtual control signal is designed as a piecewise function to guarantee the performance of the system within a finite time. The developed finite-time control method ensures that all signals in the closed-loop system are bounded, and the output tracking error converges to a small neighborhood of the origin. At last, the simulation example illustrates the feasibility and superiority of the presented control method.

This work explores the synchronization problem of two nonidentical complex-valued inertial neural networks (CVINNs) considering time-varying delays, leakage delay, and external disturbances. The entire analysis does not use reduced-order conversion, nor does it involve the separation of real and imaginary parts, but directly focuses on the original system. First, an integral sliding-mode surface suitable for the system is proposed. Second, the efficient sliding-mode control laws are designed, under which the state trajectories of the closed-loop dynamic error systems can be driven onto the predefined sliding-mode surface in finite time. Then, not requiring the time-varying delays to be differentiable, by constructing innovative Lyapunov–Krasovskii functionals, the synchronization criteria are obtained in the forms of the linear matrix inequality techniques. Eventually, for the systems with different types of activation functions, the corresponding numerical verification and comparison are carried out.

We study the containment control problem for high-order heterogeneous nonlinear multi-agent systems under distributed event-triggered schemes. To achieve the containment control objective and reduce communication consumption among agents, a distributed event-triggered control scheme is proposed by applying the backstepping method, Lyapunov functional approach, and neural networks. Then, the results are extended to the self-triggered control case to avoid continuous monitoring of state errors. The developed protocols and triggered rules ensure that the output for each follower converges to the convex hull spanned by multi-leader signals within a bounded error. In addition, no agent exhibits Zeno behavior. Two numerical simulations are finally presented to verify the correctness of the obtained results.

This study considers an adaptive neural-network (NN) periodic event-triggered control (PETC) problem for switched nonlinear systems (SNSs). In the system, only the system output is available at sampling instants. A novel adaptive law and a state observer are constructed by using only the sampled system output. A new output-feedback adaptive NN PETC strategy is developed to reduce the usage of communication resources; it includes a controller that only uses event-sampling information and an event-triggering mechanism (ETM) that is only intermittently monitored at sampling instants. The proposed adaptive NN PETC strategy does not need restrictions on nonlinear functions reported in some previous studies. It is proven that all states of the closed-loop system (CLS) are semiglobally uniformly ultimately bounded (SGUUB) under arbitrary switchings by choosing an allowable sampling period. Finally, the proposed scheme is applied to a continuous stirred tank reactor (CSTR) system and a numerical example to verify its effectiveness.

The global output feedback stabilization problem is investigated in this paper via sampled-data control for switched nonlinear systems in the p-normal form. First, a reduced-order state observer is designed. Then, an output feedback sampled-data controller is constructed with the relaxation of some restrictions of switched nonlinear systems. The proposed controller can ensure that all states of the corresponding closed-loop system can converge to the origin. Simulation results are given to show the effectiveness of the proposed scheme.

This article investigates the sampled-data stabilization problem of a class of switched nonlinear systems. All subsystems of the considered system are allowed to be unstabilizable. To relax the restrictions on unknown nonlinear functions in some existing results, we use the nonlinear approximation ability of radial basis function neural networks. Novel mode-dependent adaptive laws and sampled-data control laws are constructed by only using the system states' information at sampling instants. A novel sampled-data switching condition is derived, which can avoid Zeno behavior effectively. To guarantee that all states of the closed-loop system (CLS) are bounded, a new allowable sampling period is deduced. Finally, we demonstrate the proposed method's effectiveness through two examples.

In this study, an improved square root adaptive cubature Kalman filter (ISRACKF) is proposed to improve the filter performance in terms of accuracy, computation efficiency, and robustness. Through the evaluated measure of non‐linearity value, the cubature rule under different accuracy levels can be adaptively selected in the dynamic process or measurement model. In this way, high accuracy can be maintained without sacrificing computation efficiency. Furthermore, the maximum correntropy criterion cost function can help improve the robustness of ISRACKF. The measure of non‐Gaussianity value is utilised to control the computation complexity of the robust iterative process as well. The stability proof of estimated state error and covariance is given. The comparison results of the target tracking problem and integrated navigation system demonstrate the superior performance of the proposed ISRACKF in this study.

This paper considers the global stabilization problem via sampled-data control for a class of switched nonlinear systems meanwhile taking into account asynchronous switching. First of all, a state feedback sampled-data controller is constructed by backstepping design method. Then, a relationship between the sampling period and the average dwell time is derived, which can guarantee that the closed-loop system is globally asymptotically stable. Finally, two simulation examples are presented to demonstrate the effectiveness of the proposed method.

A class of nonlinear system is substituted by a time-varying linear system. Cubic spline functions are used directly to identify the coefficients of time-varying linear system. Then minimum variance controller for the nonlinear system is designed. The stability of closed-loop system is proved. The effectiveness of the presented method is demonstrated by the simulation results.