Neural Control Scheme Research Articles

Frequency-dependent force direction elucidates neural control of balance

BackgroundMaintaining upright posture is an unstable task that requires sophisticated neuro-muscular control. Humans use foot–ground interaction forces, characterized by point of application, magnitude, and direction to manage body accelerations. When analyzing the directions of the ground reaction forces of standing humans in the frequency domain, previous work found a consistent pattern in different frequency bands. To test whether this frequency-dependent behavior provided a distinctive signature of neural control or was a necessary consequence of biomechanics, this study simulated quiet standing and compared the results with human subject data.MethodsAiming to develop the simplest competent and neuromechanically justifiable dynamic model that could account for the pattern observed across multiple subjects, we first explored the minimum number of degrees of freedom required for the model. Then, we applied a well-established optimal control method that was parameterized to maximize physiologically-relevant insight to stabilize the balancing model.ResultsIf a standing human was modeled as a single inverted pendulum, no controller could reproduce the experimentally observed pattern. The simplest competent model that approximated a standing human was a double inverted pendulum with torque-actuated ankle and hip joints. A range of controller parameters could stabilize this model and reproduce the general trend observed in experimental data; this result seems to indicate a biomechanical constraint and not a consequence of control. However, details of the frequency-dependent pattern varied substantially across tested control parameter values. The set of parameters that best reproduced the human experimental results suggests that the control strategy employed by human subjects to maintain quiet standing was best described by minimal control effort with an emphasis on ankle torque.ConclusionsThe findings suggest that the frequency-dependent pattern of ground reaction forces observed in quiet standing conveys quantitative information about human control strategies. This study’s method might be extended to investigate human neural control strategies in different contexts of balance, such as with an assistive device or in neurologically impaired subjects.

Adaptive Fixed-Time Neural Network Tracking Control of Nonlinear Interconnected Systems.

In this article, a novel adaptive fixed-time neural network tracking control scheme for nonlinear interconnected systems is proposed. An adaptive backstepping technique is used to address unknown system uncertainties in the fixed-time settings. Neural networks are used to identify the unknown uncertainties. The study shows that, under the proposed control scheme, each state in the system can converge into small regions near zero with fixed-time convergence time via Lyapunov stability analysis. Finally, the simulation example is presented to demonstrate the effectiveness of the proposed approach. A step-by-step procedure for engineers in industry process applications is proposed.

Fixed-time composite neural learning control of state-constrained nonlinear uncertain systems

This paper presents a fixed-time composite neural learning control scheme for nonlinear strict-feedback systems subject to unknown dynamics and state constraints. To address the problem of state constraints, a new unified universal barrier Lyapunov function is proposed to convert the constrained system into an unconstrained one. Taking the unconstrained system, a modified fixed-time convergence state predictor is explored, enabling the prediction error for compensating the neural adaptive law to be obtained and improving the learning ability of online neural networks (NNs). Without employing fractional power terms or a complicated switching strategy to build the control law, a new method of constructing a smooth fixed-time dynamic surface control scheme is proposed. This overcomes the potential singularity problem and the explosion of complexity often encountered in fixed-time back-stepping designs. The representative features of our design are threefold. First, it is free of the fractional power terms, yet offers fixed-time convergence. Second, it addresses the state constraint problem without requiring a feasibility check. Third, it constructs a new state-predictor and enhances the approximation accuracy of NNs. The stability of the proposed control scheme is analyzed using the Lyapunov technique. Simulation results are presented to illustrate the effectiveness of the proposed controller.

Sliding-Mode Observer-Based Fault Reconstruction for T-S Fuzzy Descriptor Systems

This article discusses the problem of sliding-mode observer (SMO)-based fault reconstruction for T-S fuzzy descriptor systems by using the RBF neural network. First, a descriptor learning SMO, without the observability condition of fast subsystem, is developed such that the reconstruction of fault and states is achieved, simultaneously. Then, a sufficient linear matrix inequality (LMI) condition is presented to guarantee the admissibility of the sliding motion. Meanwhile, the LMI condition ensures the boundedness of the state and fault estimation errors. By utilizing robust H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> strategy, the impact of disturbance on the system can be reduced. Moreover, an RBF neural network-based sliding-mode control (SMC) strategy is adopted to estimate the nonlinearity, unknown positive constants, and ensure the reachability condition, simultaneously. Finally, an example is presented to verify the efficacy of our approach.

Frontal midline theta differentiates separate cognitive control strategies while still generalizing the need for cognitive control

Cognitive control processes encompass many distinct components, including response inhibition (stopping a prepotent response), proactive control (using prior information to enact control), reactive control (last-minute changing of a prepotent response), and conflict monitoring (choosing between two competing responses). While frontal midline theta activity is theorized to be a general marker of the need for cognitive control, a stringent test of this hypothesis would require a quantitative, within-subject comparison of the neural activation patterns indexing many different cognitive control strategies, an experiment lacking in the current literature. We recorded EEG from 176 participants as they performed tasks that tested inhibitory control (Go/Nogo Task), proactive and reactive control (AX-Continuous Performance Task), and resolving response conflict (Global/Local Task-modified Flanker Task). As activity in the theta (4–8 Hz) frequency band is thought to be a common signature of cognitive control, we assessed frontal midline theta activation underlying each cognitive control strategy. In all strategies, we found higher frontal midline theta power for trials that required more cognitive control (target conditions) versus control conditions. Additionally, reactive control and inhibitory control had higher theta power than proactive control and response conflict, and proactive control had higher theta power than response conflict. Using decoding analyses, we were able to successfully decode control from target trials using classifiers trained exclusively on each of the other strategies, thus firmly demonstrating that theta representations of cognitive control generalize across multiple cognitive control strategies. Our results confirm that frontal midline theta-band activity is a common mechanism for initiating and executing cognitive control, but theta power also differentiates between cognitive control mechanisms. As theta activation reliably differs depending on the cognitive control strategy employed, future work will need to focus on the differential role of theta in differing cognitive control strategies.

Command filter-based adaptive neural finite-time control for stochastic nonlinear systems with time-varying full-state constraints and asymmetric input saturation

This article studies a tracking control problem for a class of stochastic nonlinear systems with time-varying full-state constraints and asymmetric input saturation. Firstly, the Gauss Error Function is introduced to solve the difficulty arising from the saturation nonlinearity. Meanwhile, to overcome the problem of calculating explosion caused by the repeated differentiation of the virtual control signals, a finite-time command filter with a compensation mechanism is developed. Combining the neural networks' approximation ability and the backstepping technique, an adaptive neural finite-time control strategy is proposed for the considered system by constructing the time-varying barrier Lyapunov function. Under the proposed control strategy, it is guaranteed that all signals are bounded in a sense of mean square, the output of the system can track the reference signal within a finite time and all states will not violate the constraints. Furthermore, the stability of the closed-loop system is analysed based on the stochastic finite-time stability theory. Finally, a simulation example verifies the effectiveness of the proposed control strategy.

Adaptive neural control for uncertain switched nonlinear systems with a switched filter-contained hysteretic quantizer

This paper demonstrates an adaptive neural control strategy for the uncertain nonlinear switched systems with dwell time. The unknown nonlinear functions, hysteresis nonlinear and external disturbances are taken into account in this system. A filter-contained hysteretic quantizer is introduced to filter out the high frequency components, leading to a durable solution for the conflict between hysteresis actuator and quantization, that is the hysteresis loop deforms when the input signal goes beyond the boundary. In addition, the adaptive neural network control strategy is developed using backstepping techniques to eliminate external disturbances and unknown nonlinear function. This strategy can guarantee that the tracking error converges to an adjustable field near zero.

Neural Network-Based Active Fault-Tolerant Control Design for Unmanned Helicopter with Additive Faults

A novel adaptive neural network-based fault-tolerant control scheme is proposed for six degree-of-freedom nonlinear helicopter dynamic. The proposed approach can detect and mitigate actuators and sensors’ faults in real time. An adaptive observer-based on neural network (NN) and extended Kalman filter (EKF) is designed, which incorporates the helicopter’s dynamic model to detect faults in the actuators and navigation sensors. Based on the detected faults, an active fault-tolerant controller, including three loops of dynamic inversion, is designed to compensate for the occurred faults in real time. The simulation results showed that the proposed approach is able to detect and mitigate different types of faults on the helicopter actuators, and the helicopter tracks the desired trajectory without any interruption.

An improved neural network tracking control strategy for linear motor-driven inverted pendulum on a cart and experimental study

An improved neural network tracking control strategy for linear motor-driven inverted pendulum on a cart and experimental study

Observer‐based adaptive neural output‐feedback event‐triggered control for discrete‐time nonlinear systems using variable substitution

AbstractIn this paper, an event‐triggered adaptive neural output feedback control scheme is developed for a class of uncertain discrete‐time strict‐feedback nonlinear systems subject to immeasurable system states and network resource limitation. Ani‐step ahead predictor is synthesized to obtain the future signal of the reference orbit. By combining the neural observer and the variable substitution technology, an event‐triggered adaptive neural control scheme is developed, thereby estimating the immeasurable system states and avoiding then‐step delays of the existing controller. To promote the transient system performance, an improved triggering condition is designed to increase the number of triggering events in the transient process. The stability analysis of the closed‐loop system is divided into two parts to deal with the challenge from the simultaneous presence of state estimations, unknown system dynamics, and aperiodical controller weight updating laws. The proposed scheme achieves the state estimation, guarantees the output tracking performance with the improved transient control performance, and reduces the communication resource. Simulation studies on a numerical example and a networked robot manipulator are, respectively, implemented to show the validity of the proposed scheme.

Modelling and neural adaptive vibration control for three-dimensional Timoshenko beam with output restrictions and external disturbances

In this paper, the dynamic modelling and adaptive boundary control of a three-dimensional Timoshenko beam are presented for vibration suppression. In order to ensure that no modal information is left out, partial differential equations (PDEs) are employed to describe the nonlinear dynamic characteristics of the system by applying the extended Hamilton’s principle. Based on the established model, barrier Lyapunov functions (BLFs) are utilised for dealing with system output restrictions. For practical consideration, parameter uncertainties and external disturbances are also taken into account. Under the proposed neural adaptive control scheme, undesirable elastic vibration of the three-dimensional Timoshenko beam is restricted and constrained within given bounds even subject to external perturbations, without needing to know any system parameters. The system stability is proved by the Lyapunov’s direct method. The control performance is verified via using numerical simulations.

LVS guidance principle and adaptive neural fault-tolerant formation control for underactuated vehicles with the event-triggered input

This paper presents a novel adaptive neural event-triggered formation control strategy for underactuated surface vehicles (USVs) in the presence of marine practical constraints, i.e. the waypoints-based planned path and actuator failures. The proposed strategy is composed of the guidance model and the control one. For the guidance part, the logic virtual ship (LVS) guidance principle is developed to implement the reference path programming for the formation pattern. For merit of the adaptive virtual ship, the smooth reference can be produced for followers without using the velocity information of leader. Furthermore, the proposed adaptive neural fault-tolerant control algorithm is to guarantee that the followers can maintain their pattern and converge to the corresponding reference path. In the algorithm, the neural networks (NNs)-based observer is designed to estimate the unmeasurable velocities of followers online. With the input event-triggered mechanism, two adaptive parameters are derived to compensate for the gain uncertainty and actuator failure. And that could effectively reduce the communication burden for the channel from controller to actuator. The derived closed-loop system has been proved to be with the semi-global uniform ultimate bounded (SGUUB) stability. Finally, the superiority of the proposed strategy can be verified by the simulation experiments.

Adaptive Neural Network Fixed-Time Control Design for Bilateral Teleoperation With Time Delay.

In this article, subject to time-varying delay and uncertainties in dynamics, we propose a novel adaptive fixed-time control strategy for a class of nonlinear bilateral teleoperation systems. First, an adaptive control scheme is applied to estimate the upper bound of delay, which can resolve the predicament that delay has significant impacts on the stability of bilateral teleoperation systems. Then, radial basis function neural networks (RBFNNs) are utilized for estimating uncertainties in bilateral teleoperation systems, including dynamics, operator, and environmental models. Novel adaptation laws are introduced to address systems' uncertainties in the fixed-time convergence settings. Next, a novel adaptive fixed-time neural network control scheme is proposed. Based on the Lyapunov stability theory, the bilateral teleoperation systems are proved to be stable in fixed time. Finally, simulations and experiments are presented to verify the validity of the control algorithm.

Investigation on GaN HEMTs Based Three-Phase STATCOM with Hybrid Control Scheme.

The modern trend of decarbonization has encouraged intensive research on renewable energy (RE)-based distributed power generation (DG) and smart grid, where advanced electronic power interfaces are necessary for connecting the generator with power grids and various electrical systems. On the other hand, modern technologies such as Industry 4.0 and electrical vehicles (EV) have higher requirements for power converters than that of conventional applications. Consequently, the enhancement of power interfaces will play an important role in the future power generation and distribution systems as well as various industrial applications. It has been discovered that wide-bandgap (WBG) switching devices such as gallium nitride (GaN) high electron mobility transistors (HEMTs) and silicon carbide (SiC) metal-oxide-semiconductor field-effect transistors (MOSFETs) offer considerable potential for outperforming conventional silicon (Si) switching devices in terms of breakdown voltage, high temperature capability, switching speed, and conduction losses. This paper investigates the performance of a 2kVA three-phase static synchronous compensator (STATCOM) based on a GaN HEMTs-based voltage-source inverter (VSI) and a neural network-based hybrid control scheme. The proportional-integral (PI) controllers along with a radial basis function neural network (RBFNN) controller for fast reactive power control are designed in synchronous reference frame (SRF). Both simulation and hardware implementation are conducted. Results confirm that the proposed RBFNN assisted hybrid control scheme yields excellent dynamic performance in terms of various reactive power tracking control of the GaN HEMTs-based three-phase STATCOM system.

A Novel PID Control Strategy Based on Improved GA-BP Neural Network for Phase-Shifted Full-Bridge Current-Doubler Synchronous Rectifying Converter

In this paper, a phase-shifted full-bridge current-doubler synchronous rectifying converter (PSFB-CDSRC) based on IGBT and its control strategies are studied. In the main circuit, a current doubling synchronous rectifying circuit based on IGBT is presented to further reduce the power loss of power devices. Moreover, in the control strategy, in view of the existing researches, the basic BP neural network PID control performance of the rectifying converter still can be further improved. Therefore, this paper combines the quasi-Newton algorithm and traditional GA to propose an improved GA-BP (IGA-BP) neural network to further improve PID control performance. The simulation results demonstrate that the maximum efficiency of 5 V/500 A rectifying converter based on the proposed circuit scheme can reach 94.1%, and the rectifying converter has a good performance of excellent waveform and wide range of load. IGA-BP neural network PID control responds fast and reaches the stable state quickly in comparison with that controlled by the GA-BP neural network control strategy, and the steady-state time can be reduced by 10.5% through using IGA-BP neural network control strategy. This study can provide a valuable guidance and reference, not only in circuit scheme but also in the optimal PID control strategy for design of the high-efficiency DC/DC rectifying converter with higher power in the future.

Neural Network Supervision Control Strategy for Inverted Pendulum Tracking Control

This paper presents several control methods and realizes the stable tracking for the inverted pendulum system. Based on the advantages of RBF and traditional PID, a novel PID controller based on the RBF neural network supervision control method (PID-RBF) is proposed. This method realizes the adaptive adjustment of the stable tracking signal of the system. Furthermore, an improved PID controller based on RBF neural network supervision control strategy (IPID-RBF) is presented. This control strategy adopts the supervision control method of feed-forward and feedback. The response speed of the system is further improved, and the overshoot of the tracking signal is further reduced. The tracking control simulation of the inverted pendulum system under three different signals is given to illustrate the effectiveness of the proposed method.

Soft Computing-Based Congestion Control Schemes in Wireless Sensor Networks: Research Issues and Challenges

Wireless Sensor Networks (WSNs) are a special class of wireless ad-hoc networks where their performance is affected by different factors. Congestion is of paramount importance in WSNs. It badly affects channel quality, loss rate, link utilization, throughput, network life time, traffic flow, the number of retransmissions, energy, and delay. In this paper, congestion control schemes are classified as classic or soft computing-based schemes. The soft computing-based congestion control schemes are classified as fuzzy logic-based, game theory-based, swarm intelligence-based, learning automata-based, and neural network-based congestion control schemes. Thereafter, a comprehensive review of different soft computing-based congestion control schemes in wireless sensor networks is presented. Furthermore, these schemes are compared using different performance metrics. Finally, specific directives are used to design and develop novel soft computing-based congestion control schemes in wireless sensor networks. &nbsp;

A Tuning Method via Borges Derivative of a Neural Network-Based Discrete-Time Fractional-Order PID Controller with Hausdorff Difference and Hausdorff Sum

In this paper, the fractal derivative is introduced into a neural network-based discrete-time fractional-order PID controller in two areas, namely, in the controller’s structure and in the parameter optimization algorithm. The first use of the fractal derivative is to reconstruct the fractional-order PID controller by using the Hausdorff difference and Hausdorff sum derived from the Hausdorff derivative and Hausdorff integral. It can avoid the derivation of the Gamma function for the order updating to realize the parameter and order tuning based on neural networks. The other use is the optimization of order and parameters by using Borges derivative. Borges derivative is a kind of fractal derivative as a local fractional-order derivative. The chain rule of composite function is consistent with the integral-order derivative. It is suitable for updating the parameters and the order of the fractional-order PID controller based on neural networks. This paper improves the neural network-based PID controller in two aspects, which accelerates the response speed and improves the control accuracy. Two illustrative examples are given to verify the effectiveness of the proposed neural network-based discrete-time fractional-order PID control scheme with fractal derivatives.

Fast finite time adaptive neural network control for a class of uncertain nonlinear systems subject to unmodeled dynamics

In this paper, a fast finite time adaptive control issue is discussed for a class of uncertain nonlinear systems. The systems considered involve unmodeled dynamics as well as dynamical disturbances. First, neural networks (NNs) are introduced to deal with the difficulties caused by unknown nonlinear uncertainties, and dynamical signal functions are utilized to handle unmodeled dynamics and dynamical disturbances. Second, based on adaptive backstepping technique and a fast finite-time stability criterion, a fast finite time adaptive neural network control scheme for a class of uncertain nonlinear systems subject to unmodeled dynamics is proposed for the first time. The proposed control scheme can not only ensure that all closed-loop signals are bounded, but also has the robustness to unmodeled dynamics and dynamical disturbances. The main innovations of this work lie in the ingenious design of parameter adaptive laws and controller, and the development of fast finite-time adaptive control algorithm from a new point of view. Finally, the feasibility of the proposed control algorithm is elaborated by two simulation examples.

Adaptive Neural Output Feedback Control for MSVs With Predefined Performance

In this paper, we investigate the problem of trajectory tracking control for marine surface vehicles (MSVs), which are subject to dynamic uncertainties, external disturbances and unmeasurable velocities. To recover the unmeasurable velocities, a novel adaptive neural network-based (NN-based) state observer is constructed. To guarantee the transient and steady-state tracking performance of the system, a novel nonlinear transformation method is proposed by employing a tracking error transformation together with a newly constructed performance function, which is characterized by a user-defined settling time and tracking control accuracy. With the aid of the state observer and the nonlinear transformation method in combination with the adaptive NN technique and vector-backstepping design tool, an adaptive neural output-feedback trajectory tracking control scheme with predefined performance is developed. With regard to the developed control scheme, uncertainties can be reconstructed only by utilizing the position and heading of the MSVs. Independent designs of the state observer and the controller can be achieved, and the position tracking error can be guaranteed to fall into a predefined residual set in the user-defined time frame and remain in the above set. A rigorous stability analysis validates that all signals in the closed-loop trajectory tracking control system for MSVs are uniformly ultimately bounded. Simulation results verify the effectiveness of the developed adaptive neural output-feedback trajectory tracking control scheme.