Fuzzy parameter-varying (FPV) systems constitute an emerging framework for analyzing nonlinear systems. However, due to the coexistence of fuzzy logic and time-varying parameters, the stabilization methods for FPV systems often suffer from severe conservatism. To address this issue, this study proposes a dimension-reduced design method to construct an FPV sliding mode control (SMC) law for general FPV systems. Furthermore, the global asymptotic stability of the resulting control system is rigorously proved by the stability theory of switching systems. Finally, two simulation cases of different dimensions demonstrate the superiority of the proposed method in terms of stability and computational complexity.

To achieve high-precision trajectory tracking for robotic manipulators subject to unknown nonlinearities and external disturbances, a backstepping sliding mode control strategy incorporating a dual-path extreme learning machine (DP-ELM) and a disturbance observer is proposed. The DP-ELM approximates the system’s unknown nonlinearities via two parallel paths: a main path driven by joint angle errors to capture dominant nonlinearities, and an auxiliary path utilizing angular velocity errors to compensate for unmodeled high-frequency residuals. To enhance approximation capability and computational efficiency, a multi-scale feature extraction mechanism and a Top-K dynamic sparse activation strategy are integrated into the network. Furthermore, an exponentially convergent disturbance observer is designed to estimate external disturbances online. Both numerical simulations in Simulink and experiments on a physical platform of a 3-degree-of-freedom (3-DOF) robotic manipulator demonstrate that the proposed scheme yields superior tracking accuracy, rapid convergence, and strong robustness against disturbances compared to existing methods.

This paper addresses the adaptive event-triggered control (ETC) problem for a class of time-varying switched systems with parametric uncertainties. A novel dual-adaptive ETC method is proposed by utilizing the initial excitation (IE) condition and the average dwell time (ADT) method. To reduce the conservatism resulting from the adoption of a common update law, different update laws are developed for all subsystems. An adaptive switching event-triggered controller is designed such that it significantly conserves communication resources. Moreover, the proposed approach guarantees the exclusion of the Zeno behavior. Furthermore, a rigorous Lyapunov stability analysis is conducted to establish the boundedness of all closed-loop signals for the time-varying switched systems. Finally, an example is provided to illustrate the effectiveness of the proposed method.

Accurate trajectory tracking control of mobile robots is a complex task due to inherent system nonlinearities, modeling uncertainties, and external disturbances. To tackle these challenges, this paper proposes a neural network-based residual learning approach (NNRes) for high-fidelity and efficient system modeling. Unlike full model learning, the NNRes framework focuses on learning only the residual error between a nominal physical model and the actual system dynamics, thus preserving prior knowledge while enhancing adaptability. This hybrid model, combining the physical model and learned residual, is embedded into a Nonlinear Model Predictive Control (NMPC) scheme to improve control accuracy and robustness. The proposed method is systematically compared with state-of-the-art methods: full model learning using NNs (NNFull) and residual learning using Gaussian Processes (GPRes). Comparative simulation results demonstrate that NNRes achieves superior trajectory tracking control performance, enhanced robustness, and lower computational cost. These findings validate the effectiveness of NNRes within NMPC frameworks and offer valuable insights for the intelligent control of complex robotic systems.

This study investigates the impact of time-varying jet reaction forces on ship-cleaning manipulators, focusing on joint torque fluctuations, high-frequency chatter, and the deterioration of end-effector positioning accuracy. An enhanced active disturbance rejection control (ADRC) method is proposed, which integrates jet reaction force modeling, joint torque feedforward compensation, and improvements to both the extended state observer (ESO) and the nonlinear state error feedback (NLSEF). Simulation results show that the enhanced ADRC reduces the joint stabilization time from 10 to 2 seconds, suppresses over 70% of the initial torque spikes, and decreases the ESO observation errors by more than 60%. Compared with the classical sliding mode control (SMC), the proposed method effectively eliminates high-frequency chattering phenomena and reduces control signal aggressiveness during the startup phase. This method improves trajectory tracking accuracy under strong disturbances while smoothing the control output and significantly reducing high-frequency joint chatter. These results confirm that the enhanced ADRC method improves stability and responsiveness, providing a robust framework for the reliable operation of ship-cleaning manipulators in high-disturbance environments.

This paper investigates the fault detection (FD) problem of interval type-2 (IT2) fuzzy systems with random sensor saturation and denial-of-service (DoS) attacks. In order to alleviate communication burden and reduce the impact of DoS attacks, a resilient memory event-triggered mechanism (RMETM) is proposed, and round robin (RR) protocol is used to coordinate the data transmission of nodes at each triggering moment. Subsequently, a switching fuzzy fault detection filter (FDF) is constructed, and the error dynamic system is established. Sufficient conditions are derived to ensure that the error dynamic system is exponentially stable in mean square with the satisfied H ∞ performance via the Lyapunov stability theory. In addition, filter gains are obtained by solving linear matrix inequalities (LMIs). Finally, two simulation examples are employed to demonstrate the effectiveness of the proposed FD method.

Centrifugal pumps are critical components in industrial processes and heating, ventilation, air conditioning, and cooling (HVAC) systems. To mitigate CO 2 emissions, it is essential to reduce the energy consumption of centrifugal pumps across various applications. Conventionally, valve-based flow control is widely adopted in industrial processes and HVAC systems; however, this approach results in excessive energy consumption when the flow demand is low. This study introduces a soft-sensor-based proportional pressure control method for centrifugal pumps. The prediction model integrates back-propagation neural networks (BPNN) with genetic algorithms (GA) to estimate pump flow rate and head without requiring feedback signals from flow meters or pressure transducers. Motor speed and power serve as input variables for the prediction model. A motor speed correction equation compensates for the slip effect in asynchronous motors. Furthermore, two PID control loops based on the GA-BPNN prediction model are implemented to independently adjust the valve opening and pump speed, enabling the pump to operate in a proportional pressure mode. The novelty lies in the soft-sensor for pump flow/head prediction using motor power/speed with slip compensation for the integrated control strategy. Experimental results demonstrate that the proposed soft-sensor-based control method achieves high control accuracy and reliability on pump flow and head. This new control method eliminates the need for physical sensors used in flow or pressure measurement and transmission feedback, thereby enabling precise control of the centrifugal pump’s operating state based on the performance prediction model. Compared with traditional valve-based flow control, this new strategy can achieve energy savings exceeding 23.5%.

In this paper, an improved predefined-time scheme is constructed to solve the tracking control problem of switched nonlinear systems with input quantization (IQ) and hysteresis nonlinearity. First, we introduce quantized control to address the communication resource wastage caused by the time-sampling mechanisms. The unexpected high-frequency components introduced by IQ affect system performance seriously. To overcome this case, a new approach is proposed to remedy this, wherein a novel quantization-filtering auxiliary system is conceived via the utilization of a band-pass filter. Subsequently, an improved PTL is developed, combining the filter parameters and quantization parameters, which has two main characteristics: (1) it has a coupled relationship between filter and quantizer, further reducing the minimum upper bound of settle time and the complexity of tuning simulation parameters and (2) compared to traditional predefined-time lemmas, it has faster convergence rates. Furthermore, a predefined-time filter is devised to address the issue of “explosion of complexity.” Finally, the effectiveness of the developed control strategy is validated through numerical and practical examples.

In order to improve the servo accuracy and dynamic performance of permanent magnet synchronous motors (PMSM), a physics-guided neural network (PGNN) model is proposed, which can be used to establish a dynamic inverse model of PMSM-based servo systems. First, based on the characteristics of PMSM, a PGNN model is constructed, which is divided into a preprocessing model part, a physical model part, and a neural network part. Then, the above model is applied to the design of feedforward controller to improve servo performance. Considering that the dynamic characteristics of servo systems are variable, this paper further proposes an adaptive method that can periodically update model parameters based on input and output data of the system, thereby improving control effectiveness. The simulation experiment results show that the PGNN model proposed in this paper has high accuracy and good generalization ability, and can accurately describe the nonlinear characteristics of PMSM. The control effect of the feedforward controller based on this model is better than traditional modeling method and the model established in our previous work. Therefore, the control method proposed in this paper holds significant value and is crucial for advancing the intelligent development of PMSM systems.

In this paper, a distributed cooperative guidance method based on the Long Short-Term Memory (LSTM) network is proposed, considering the communication packet dropouts and time delays between missiles. A cooperative guidance model with nonideal communication constraints is given, in which the Bernoulli-distribution-based packet dropout model and uniform-distribution-based time delay model are built, respectively. Furthermore, considering the fact that the exchanged information between missiles is missing due to the nonideal communication constraints, an LSTM network is established for predicting the trajectory information of neighborhood missiles to compensate for the packet dropout data. Subsequently, an LSTM-compensation-based distributed two-stage cooperative guidance law is designed by designing the time-to-go coordination variable with sampling coordination data, in which the synchronization of the time-to-go variable is achieved in the first stage by designing a cooperative guidance mechanism, and a successful attack target is obtained by using the proportional navigation guidance method in the second stage. Simulation results demonstrate that the miss distance is reduced by about 45.71% and the convergence time of the coordination variable is reduced by about 53.33% compared to the method without data compensation, which effectively improves the stability and robustness of the cooperative guidance method under a nonideal communication environment.