ABSTRACT In this paper, the output consensus problem of heterogeneous discrete‐time multi‐agent systems with state, input, output, and communication delays is investigated. To address this problem, the concept of bivariate fundamental matrices is introduced for the considered multi‐agent system. With such a concept as a tool, an observer‐based predictor is developed for each agent to compensate for its input and output delays. Based on this predictor, a distributed control protocol is designed for each agent. In addition, a condition on the solvability of linear matrix equations associated with the proposed distributed control protocol is established. It is shown that under a switching communication network, the designed distributed control protocol ensures that output consensus is achieved among all agents. Finally, two simulation examples are presented to verify the effectiveness and advantage of the proposed protocol, including one involving a network of heterogeneous chemical reactors subject to multiple delays.
ABSTRACT This paper studies quasi‐projective synchronization (QPS) of heterogeneous fractional‐order Clifford‐valued neural networks (FO‐CVNNs) under periodic intermittent pinning control. The drive and response networks may have different self‐feedback coefficients, synaptic weights, activation functions, and external inputs. Their states lie in , so the formulation includes real‐, complex‐, quaternion‐, and higher‐dimensional Clifford‐valued networks. A nonzero real projection factor defines the target manifold . The controller reduces actuation in two independent ways: it is applied only to a pinning set and only during periodic ON‐windows with duty cycle . A multivector Lyapunov analysis and a switched fractional comparison argument yield sufficient conditions, in linear‐matrix‐inequality (LMI)‐compatible form, for full and sparse pinning. The heterogeneous error residual is derived explicitly. In particular, the previously implicit nonhomogeneity is retained and bounded rigorously by when . The results distinguish exact Mittag–Leffler convergence in the residual‐free case from practical QPS with a computable ultimate bound under persistent mismatch. Two examples in and , sensitivity studies, 500 Monte Carlo trials, and an image‐encryption illustration support the analysis. For a nominal actuation‐occupancy index , the two examples use and , respectively, compared with for continuous full‐network control. Direct comparison with recent fractional QPS, intermittent‐control, and Clifford‐network studies clarifies that the proposed framework uniquely combines heterogeneity, Clifford states, temporal intermittency, and spatial pinning.
ABSTRACT This paper presents a human–robot interaction (HRI) scheme by using an adaptive admittance control, which helps stroke patients perform rehabilitation training tasks and optimizes their performance. Considering the impact of human factors, the control structure is designed to have two control loops. In the inner loop design, a model‐free adaptive control (MFAC) method is proposed to handle the unmodeled dynamics and unknown disturbances for the desired trajectory tracking, and the convergence and boundedness of this method are strictly proved by using the compression mapping principle. Then, a task‐specific outer loop is developed to find the optimal parameters of the admittance model and transformed into an LQR problem, and a learning algorithm is utilized to solve the given problem without requiring knowledge of the human arm model. Considering the safety of HRI, the constraint of the end‐effector orientation is designed. Simulation studies indicate that the proposed strategy effectively enables stroke patients to execute active training tasks on the robotic exoskeleton.
ABSTRACT In this article, the problem of predefined‐time event‐triggered tracking control for a category of non‐affine systems with dead‐zone input constraints is investigated. First, a model transformation is introduced based on the semi‐bounded conditions, which eliminates the differentiability and boundedness restrictions on the non‐affine functions. By using the transformed model, extended neural networks (NNs) are employed to handle overflow variables arising from the transformation, while simultaneously compensating for approximation errors. Moreover, an adaptive compensation scheme is adopted to estimate unknown parameters in the dead‐zone input. By accounting for the combined effects of dead‐zone input and prescribed performance constraints, an event‐triggered mechanism is constructed that effectively reduces the burden on communication resources. Based on the iterative design, the proposed control scheme ensures that the tracking error converges to a predefined region within a fixed time. Finally, simulation results demonstrate the superior performance over existing methods.
ABSTRACT This paper addresses the challenges of low data efficiency and diminished control accuracy in complex systems with completely unknown dynamics. Conventional off‐policy adaptive dynamic programming (ADP) methods, which rely on random data sampling, often lead to the accumulation of irrelevant data, straining computational resources, and hindering learning performance. To overcome these limitations, we propose a novel framework that integrates an event‐triggered mechanism into the off‐policy ADP algorithm. This innovation replaces passive, periodic sampling with an active, intelligent paradigm that collects data only when a prescribed triggering condition is met, thereby ensuring the high informational value of each sample. We provide rigorous theoretical guarantees, including the uniform ultimate boundedness and near‐optimality of the closed‐loop system and exclude Zeno behavior. The efficacy of the proposed approach is corroborated via extensive simulations on an F‐16 aircraft model and other nonlinear systems, demonstrating significant improvements in data efficiency and control performance compared to conventional methods.
ABSTRACT In this article, the secure consensus problem for networked multi‐agent systems (MASs) under distributed denial of service (DoS) attacks is investigated. Unlike most existing approaches, which consider DoS attacks launched simultaneously or periodically across communication channels, we propose a more general and severe attack model, where distributed DoS attackers independently target different communication channels within the system. By introducing a DoS‐resilient distributed control protocol incorporating dual‐terminal dynamic event‐triggered mechanisms (DETMs), we rigorously prove that leader‐following connected MASs can achieve asymptotically stable consensus under distributed DoS attacks. More interestingly, the DETMs are implemented on each agent's controller‐to‐actuator and inter‐agent communication channels, reducing both communication and control update frequencies, which helps conserve control energy and communication resources. It is also proven that the proposed protocol effectively excludes Zeno behavior. Additionally, we extend our results to the leaderless case and guarantee secure average consensus. Numerical simulations are conducted to demonstrate the efficacy and superiority of the developed protocols.
ABSTRACT Navigating marine vessels along precise trajectories remains a formidable control challenge due to high‐order inertia, nonlinear actuator constraints, and stochastic environmental disturbances. This work presents a robust dual‐strategy framework for ship course‐keeping by integrating classical frequency‐domain design with reinforcement learning. The proposed approach consists of two methods: (i) a multi‐constraint all‐stability region (ASR)‐based controller design and (ii) an ASR‐guided Q‐learning control architecture. By using the stability boundary locus approach, the ASR is constructed in the ‐plane. In addition, control‐effort constraints are directly incorporated into the ASR framework to ensure the rudder angle remains within permissible operating limits. Controller gains are selected from the feasible intersection region formed by the imposed design constraints. These analytically derived gains are then used to initialize the Q‐learning agent, enabling the learning process to begin from a stable and constraint‐satisfying operating condition. The proposed hybrid framework utilizes a reward‐based adaptive mechanism to enhance transient and steady‐state tracking while maintaining smooth and bounded control action. Stability, robustness, and rudder saturation constraints are embedded within the controller design space, ensuring that the learning process remains confined to feasible and analytically stable solutions. As a result, the proposed approach improves tracking performance without imposing excessive control effort or actuator burden. Comparative analysis with recently reported control strategies demonstrates that the proposed approach offers substantial improvements in closed‐loop performance, as evidenced by the results presented. For normal sea states, ASR and Q‐ASR controllers achieve reductions in settling time by , , and , compared to nonlinear decoration and switching feedback methods, respectively. The proposed controller efficacy was verified on a 189‐meter Yupeng vessel via real‐time controller hardware‐in‐the‐loop (C‐HIL) experiments using the C2000 Delfino F28379D platform. The proposed framework derives optimal control policies in simulation that can be efficiently implemented on standard embedded platforms. Resulting controllers offer a practical solution for autonomous ship autopilots while improving fuel economy and reducing actuator wear during maritime operations.
ABSTRACT This paper presents a resilient learning‐based adaptive sliding‐mode consensus control framework for nonlinear multi‐agent systems subject to switching communication topologies and actuator‐side cyberattacks. A distributed sliding‐mode leader‐follower tracking consensus protocol is first developed to ensure robust coordination under network constraints. To overcome the performance limitations of heuristic gain selection in conventional adaptive sliding‐mode control, an actor‐critic reinforcement learning scheme based on heuristic dynamic programming is embedded to tune controller parameters online without requiring explicit knowledge of the plant dynamics. The critic network approximates the Hamilton‐Jacobi‐Bellman (HJB) value function, enabling near‐optimal adaptive consensus performance, while the actor network synthesizes an optimal adaptive control policy that preserves robustness. Lyapunov‐based analyses establish asymptotic reachability, tracking consensus under average‐dwell‐time switching topologies, and uniform boundedness despite actuator cyberattacks. The proposed framework is validated through theoretical analysis, real‐time experimentation on the OPAL‐RT 4610XG platform, and hardware‐in‐the‐loop experiments using a leader‐follower Quanser QUBE‐Servo rotary inverted pendulum. Results demonstrate faster consensus convergence, actuator‐friendly control action, and improved robustness compared with prevalent adaptive and learning‐based controllers.
ABSTRACT This paper presents a novel adaptive fault‐tolerant prescribed performance control scheme for trajectory tracking of underactuated unmanned surface vehicles (USVs) operating under actuator faults and external disturbances. The proposed control framework integrates prescribed performance control (PPC) with radial basis function neural networks (RBF‐NNs) and barrier Lyapunov functions (BLFs) to guarantee tracking errors remain within predefined performance bounds while compensating for both multiplicative and additive actuator faults. A key innovation lies in the incorporation of a smooth shifting function that enables the controller to handle initial constraint violations, thereby significantly expanding the operational envelope compared to conventional PPC approaches. The adaptive laws are designed to simultaneously estimate unknown system dynamics, external disturbances, and fault parameters without requiring explicit fault detection and isolation mechanisms. Comprehensive simulation studies on a full‐scale USV model demonstrate that the proposed controller achieves superior tracking performance with position errors converging to steady‐state values below 0.04 m and yaw errors below 0.03 rad, even under severe actuator degradation scenarios where effectiveness factors drop to 50% and significant bias faults are present. The controller maintains prescribed performance bounds throughout nine distinct fault events occurring over a 120‐s mission, validating its robustness and fault‐tolerant capabilities.
ABSTRACT Unmanned aerial vehicles (UAVs) have achieved remarkable advancements in the realm of intelligent transportation, while the collaboration of multiple UAVs is emerging as a pivotal area of research. In this work, the flexible performance‐based cooperative control approach is proposed for the multiple UAVs‐suspended transport systems (STSs) with tracking performance constraints and cable tension limitations. Through establishing the constraint‐limitation coordination mechanism (CLCM), this approach is capable of ensuring the tension status and preventing the tensile failure of the cables, while accomplishing the prescribed performance of both the payload trajectory tracking and the UAV formation following. All the stability of the closed‐loop system, the asymptotic convergence of the CLCM, and the accomplishment of the constraints are proved based on the positive invariant set, the finite‐time convergence, and the positive system theories. Finally, the numerical simulation with comparisons is provided to demonstrate the functionality and the feasibility of the proposed control approach.
ABSTRACT Adaptive finite‐time prescribed performance control for autonomous surface vehicles (ASVs) subject to stochastic noise and asymmetric dead‐zone output nonlinearities is investigated in this paper. First, a finite‐time prescribed performance function (FTPPF) is introduced to ensure that the tracking error converges to a predefined region within a finite time, which achieves the finite‐time tracking control and good transient and stability performance. Then, to avoid the “complexity explosion” problem associated with the backstepping approach, a filtering technique is incorporated into the controller design. Furthermore, unknown nonlinear dynamics and stochastic noise are approximated by fuzzy logic systems (FLSs), and the asymmetric dead‐zone output nonlinearity is addressed by constructing a novel dead‐zone approximation model. By constructing appropriate Lyapunov functions, it is shown that all closed‐loop signals remain bounded in probability. Finally, simulation results demonstrate the effectiveness of the proposed control strategy.
ABSTRACT This work investigates prescribed‐time rigid formation control problem for double‐integrator multi‐agent systems (MASs). The key to ensure the desired formation in a prescribed time, that can be pre‐specified arbitrarily by the designer, is to integrating a time‐mapping function into the controller design. Unlike existing works that require global coordinate system information and real‐time communication among neighbors, the method we propose only needs relative position and velocity information for each agent, and does not require real‐time communication among neighbors. It is a fully decentralized control method, which can greatly enhance the universality, especially be applicable to communication‐limited scenarios (such as communication environments without GPS). In addition, the numerical execution problem inherent in the infinite gain based prescribed‐time control method is avoided in this work by integrating the time‐varying gain and fractional‐order state feedback into the controller design. The effectiveness of the proposed algorithm is validated through both numerical simulations and physical robotic experiments.
ABSTRACT This study analyzes the dissipative synchronization problem of uncertain nonlinear complex switched partial differential systems with probabilistic time‐varying delays. In order to replicate the real‐world scenario, the concept of cyber‐attacks is examined along with time‐varying actuator fault. However, to tackle these hurdles, we design an effective secure fault‐tolerant non‐fragile memory feedback controller with the presence of interchange attack. Furthermore, Bernoulli's random variables are employed to find out the random occurrence of time‐varying delays and success ratio of cyber‐attacks. The dissipative performance index is deployed to attenuate the disturbance level. By constructing suitable Lyapunov‐Krasovskii functionals and with the aid of Jensen's integral inequalities, sufficient conditions are attained in terms of linear matrix inequality to ensure the mean‐square asymptotic synchronization of the proposed system. Ultimately, simulation results are provided to verify the superiority of the proposed control scheme over the theoretical findings.
ABSTRACT Omnidirectional mobile robots (OMRs) equipped with mecanum wheels possess superior mobility, enabling them to move in any direction and pivot in place within confined spaces. This unique advantage endows them with significant application potential across various fields. The development and application of these robots require safe motion control, which necessitates maintaining a safe distance from obstacles and achieving high‐precision trajectory tracking during motion. To achieve safe obstacle avoidance and high‐precision trajectory tracking control for OMRs while ensuring real‐time control capabilities, this article proposes a hierarchical control strategy. The high‐level module integrates nonlinear model predictive control (NMPC) and artificial potential fields (APF) to plan safe avoidance trajectories. The low‐level module uses a deep neural network (DNN) to approximate the NMPC trajectory tracking control method and employs quadratic programming (QP) to correct the outputs solved by the DNN, thus improving solving speed while maintaining control performance similar to that of NMPC. Simulation results demonstrate that the proposed scheme effectively achieves safe obstacle avoidance and trajectory tracking control.
ABSTRACT Model‐free adaptive control (MFAC) can carry out various tasks using only input and output (I/O) data, providing advantages such as lower operational costs, higher scalability, and easier implementation. In this article, to further improve tracking performance, a fractional‐order model‐free adaptive control (FOMFAC) with P‐type feedforward iterative learning controller for uncertain discrete‐time nonlinear systems is proposed. First, the FOMFAC law is designed by using an optimization input criterion function based on a fractional‐order (FO) equivalent model, which contains more input and output data information. Then, an iterative learning control (ILC) scheme is proposed as the feedforward component term to achieve fast convergence. Different from the existing plug‐in method, the stability of the whole closed‐loop control system is rigorously proved. Finally, the presented simulation evidence substantiates both the theoretical validity and practical superiority of the novel control method.
ABSTRACT Quadrupedal robots operating in unstructured environments require adaptive control systems that can handle diverse terrain conditions without prior surface characterization. This paper presents an integrated control architecture that combines Model Predictive Control (MPC) with adaptive impedance control and SINDy‐based model corrections for robust terrain‐adaptive locomotion. The system automatically detects surface properties through a four‐state contact detection mechanism and adapts control parameters in real‐time based on measured ground reaction forces. The impedance controller reduces foot slippage by 40% on challenging slopes, while an admittance control component improves force tracking accuracy by 40%–60% on compliant surfaces. SINDy corrections to angular velocity dynamics enhance yaw tracking performance by 80% compared to nominal rigid body models. Comprehensive validation in PyBullet demonstrates the system's effectiveness across diverse scenarios including slope navigation, soft surface adaptation, and complex trajectory tracking. The integrated approach establishes a robust framework for autonomous quadrupedal locomotion in unstructured environments without requiring prior terrain knowledge.
ABSTRACT This article proposes an improved command‐filtered adaptive practical fast finite‐time tracking strategy for multi‐input multi‐output (MIMO) stochastic nonlinear systems. Compared to existing approaches, it targets three critical unresolved challenges: (1) intermittent output constraints active only during finite intervals, (2) control singularity in finite‐time backstepping designs, and (3) synergistic handling of concurrent practical imperfections. To address these challenges, a shift‐barrier function combination enables seamless constraint transitions, a piecewise continuous function guarantees singularity‐free control, an innovative disturbance observer estimates unknown disturbances, and Pade approximation mitigates input delay. Theoretical analysis confirms all closed‐loop signals are bounded in probability while satisfying intermittent constraints. In the end, to demonstrate the effectiveness of the proposed strategy, the method is applied to a 2‐link flexible robotic manipulator (FRM) that tracks different trajectories.
ABSTRACT To address the challenges of longitudinal cooperative platoon control for multiple heavy‐duty trucks with time‐delay compensation under constraints, this paper presents a novel model‐free adaptive control algorithm termed TD‐cMFAC. The algorithm employs a pseudo partial derivative (PPD), a time‐varying parameter, to linearize the nonlinear dynamics of the multivehicle cooperative system using dynamic linearization techniques. Afterward, a dedicated TD‐cMFAC controller is then designed. To compensate for system time delays, a hybrid mechanism integrating a Smith predictor and a tracking differentiator (TD) is developed. The controller explicitly accounts for the input and output constraints inherent in practical control processes. The primary advantage of the TD‐cMFAC algorithm is its reliance solely on the input/output data of the multivehicle system throughout the control process while maintaining robust performance in the presence of time delays and constraints. Theoretical stability analysis confirms the robustness of the proposed method. By utilizing a MATLAB/Simulink and TruckSim co‐simulation interface, the effectiveness of the control strategy is demonstrated under complex driving is demonstrated, and its practical applicability is further validated through field tests conducted with an autonomous driving platform using heavy‐duty trucks on a test road in Tianjin, China.
ABSTRACT This paper investigates a formation‐based cooperative pursuit‐evasion (PE) problem for marine surface vehicles (MSVs) under local sensing and partially connected communication topologies. A distributed two‐layer control framework is proposed for the case where each pursuer obtains the state information of one assigned evader. At the planning layer, a distributed game‐theoretic mechanism is introduced to generate PE reference trajectories based on local PE errors and prescribed relative configurations. Although each pursuer designs its pursuit strategy according to its assigned evader, all pursuers jointly achieve encirclement of the evader formation through the prescribed relative configurations. At the control layer, a fuzzy adaptive tracking controller is developed to compensate for nonlinear hydrodynamic effects and external disturbances, and to achieve asymptotic convergence of the tracking errors for the generated reference trajectories. Simulation results validate the effectiveness of the proposed framework in the considered formation‐based PE scenario.
ABSTRACT For quadrotor UAV control in the presence of input saturation, state constraints, and system uncertainties, this paper suggests a novel fractional‐order sliding‐mode control technique. Through the employment of a state‐constrained performance function, the proposed approach effectively reduces the numerical instabilities brought on by input saturation and state limits that exist in traditional approaches. Additionally, a fractional‐order sliding surface is used to enable quick practical finite‐time convergence of the system states. In comparison to conventional sliding mode control or integer order techniques, the proposed method not only enhances system robustness and chattering suppression, but it also investigates the practical finite‐time convergence qualities and clearly identifies the convergence boundaries. The simulation findings demonstrate fast convergence and excellent tracking performance, providing significant support for realistic flight control applications.
